Today's papers cluster around three methodological commitments: controlled evaluation of failure modes under realistic constraints, mechanistic diagnosis of model behavior via internal representations, and decomposition of complex objectives into tractable learning stages. The first theme spans adversarial robustness (distributed attacks in persistent codebases, typographic attacks on vision encoders, out-of-distribution performance under distribution shift), unlearning fidelity (parameter-level localization rather than output-level metrics), and safety monitoring (thresholded verifier signals calibrated via risk control), each treating evaluation as the primary lever for exposing what systems actually do rather than what they claim. The second theme, mechanistic interpretability applied to real problems, appears in circuit mining to isolate lexical encoding in vision transformers, neuron-activation-guided data selection for self-distillation, and dual-channel debate frameworks that detect emergent social objectives, all treating internal model structure as diagnostic rather than merely descriptive. The third theme decomposes learning into staged objectives: task-agnostic motor pretraining before language grounding in embodied AI, localize-then-unlearn rather than end-to-end erasure, recursive evidence replay separating retrieval from generation in long-context reasoning, and disagreement-modulated distillation that preserves student capacity while adopting teacher guidance selectively. Across domains, from code agents to vision-language models to robotics, papers privilege empirical precision over architectural novelty, favor execution-grounded metrics (pass rate, coverage, mutation score, closed-loop real-world rollouts), and treat benchmark design itself as a research contribution when it exposes gaps between claimed and actual capability.
Cole Brennan
Showing of papers
As AI coding agents become more autonomous, they increasingly ship code iteratively, with the codebase persisting across sessions. This persistence creates a new attack surface: a misaligned or prompt-injected agent can distribute attacks across pull requests (PRs) and time its payload for the PR with the best natural cover. To study the resulting dynamics, we introduce Iterative VibeCoding, a setting for AI control, the study of safely deploying capable but potentially untrusted AI. In Iterative VibeCoding, a coding agent builds software over a sequence of PRs in a persistent codebase while pursuing a covert side task. Our benchmark includes two task families: CLI tools and Flask web services, across 20 total task variations. We use Claude Sonnet 4.5 as the attack agent and GPT-4o as the monitor. We compare gradual attacks, which distribute the side task across PRs, against non-gradual attacks concentrated in a single PR. No single monitor is robust to both: which strategy evades best (success while evading the monitor) depends on the monitor type, so a defender cannot close off both gradual and non-gradual attacks with any one monitor. High evasion (>= 65%) generalizes across model attack agent backends (Sonnet 4.5, Gemini 3.1 Pro, Kimi K2.5), confirming this is a property of the persistent-state attack surface rather than a single model's capability. Evasion also remains high across state-of-the-art monitor models and the gap between gradual and non-gradual evasion widens for more capable models. We introduce a stateful link-tracker monitor that tracks suspicious buildup across PRs. On both task families, it detects gradual attacks substantially better than diff monitors that merely see more accumulated history. Combining this stronger monitor with trajectory monitors in a four-monitor ensemble reduces gradual-attack evasion from 93% under the weakest standard diff monitor to 47%.
LLMs memorize sensitive training data, including personally identifiable information (PII), creating a pressing need for reliable post hoc removal methods. Unlearning has emerged as a promising solution, with state-of-the-art(SOTA) methods often following a localize-first, unlearn-second paradigm that targets specific model parameters. However, existing benchmarks evaluate unlearning solely at the output level, leaving open the question of whether unlearning truly erases knowledge from a model's parameters or merely obfuscates it, a concern reinforced by the success of resurfacing attacks. To bridge this gap, we introduce LACUNA: the first unlearning testbed with ground-truth parameter-level localization. LACUNA injects PII of synthetic individuals into predefined parameters of 1B and 7B OLMo-based models via masked continual pretraining, enabling direct evaluation of whether unlearning targets the weights responsible for knowledge storage. We use LACUNA to benchmark current SOTA unlearning methods and find that, despite strong output-level performance, existing methods are highly imprecise and susceptible to resurfacing attacks. We further show that when localization is successful, even a simple gradient-based unlearning method achieves strong erasure and robustness to resurfacing attacks, highlighting the importance of precise unlearning. We release LACUNA to complement behavioral evaluations and drive further advances in robust, localization-based unlearning.
Many everyday programming tasks resist clean rule-based implementation, such as alerting on important log lines, repairing malformed JSON, or ranking search results by intent, and are increasingly outsourced to large language model APIs at the cost of locality, reproducibility, and price. We propose fuzzy-function programming: compiling such a function from a natural-language specification into a compact, locally-executable neural artifact. We instantiate this paradigm with Program-as-Weights (PAW), in which a 4B compiler trained on FuzzyBench, a 10M-example dataset we release, emits parameter-efficient adapters for a frozen, lightweight interpreter. A 0.6B Qwen3 interpreter executing PAW programs matches the performance of direct prompting of Qwen3-32B, while using roughly one fiftieth of the inference memory and running at 30 tokens/s on a MacBook M3. PAW reframes the foundation model from a per-input problem solver into a tool builder: invoked once per function definition, it produces a small reusable artifact whose subsequent calls per function application are cheap and offline.
Despite alignment training, LLMs remain prone to generating unsafe outputs at deployment time. Monitoring outputs online and raising an alarm when safety can no longer be assumed is therefore critical. We study a simple real-time monitor that turns a verifier signal from an external model into an alarm decision by thresholding, with the threshold calibrated via risk control. In experiments on mathematical reasoning and red teaming datasets, we show that this simple design is competitive with more advanced monitors based on sequential hypothesis testing.
Understanding and reasoning over long contexts has become a key requirement for deploying large language models (LLMs) in realistic applications. Although recent LLMs support increasingly long context windows, they often fail to use relevant evidence that is already present in the input, revealing a gap between context access and effective context utilization. In this work, we propose Recursive Evidence Replay as LLM Harness for Long-Context Reasoning (RECONTEXT), a training-free inference method for improving long-context reasoning. RECONTEXT uses model-internal relevance signals to construct a query-conditioned evidence pool and replays it before final generation while preserving the full original context. This recursive selection process separates evidence organization from answer generation without training, external memory, or context pruning. We also provide a theoretical analysis based on associative memory, which characterizes the context as a memory store, the question as a retrieval cue, attention as cue-trace association, and replay as trace reactivation. Experiments on eight long-context datasets with 128K context length show that RECONTEXT consistently improves evidence utilization across Qwen3-4B, Qwen3-8B, and Llama3-8B, achieving the best average rank on all three backbones. Code is available at https://github.com/Yanjun-Zhao/ReContext.
LLM agents will increasingly act in socially structured settings where role, audience, and relational context can shape what is advantageous or costly to say. We study whether such social structure, without any explicit objective in the prompt, changes what an agent expresses publicly relative to an off-the-record (OTR) channel elicited under the same condition. We introduce a dual-channel debate framework in which agents produce public utterances that enter the shared history alongside OTR responses that are recorded but never shown to the other participant. Across 10 models, 3 scenarios, and 5 variations within each scenario, alignment-inducing settings produce systematic public-OTR divergence in the targeted agent, with its decision divergence rising from a $\sim$3% baseline to roughly 40%. The effect is consistent across four aggregate analyses: stance, semantic similarity, natural language inference, and survey responses. In some cases, the OTR response explicitly attributes public accommodation to relational pressures, such as career risk or sponsorship obligation. The findings suggest that agent evaluation should extend beyond explicit goals and detect emergent objectives. We present a dual-channel evaluation framework and complementary behavioral measures that operationalize this assessment.
Long-form TV dramas present a formidable challenge for comprehensive video understanding, where deciphering complex storyline often relies on \textbf{speaker recognition}, the task of accurately attributing each spoken utterance to its respective character. In this paper, we advance this field through two primary contributions. (1) We introduce \textbf{DramaSR-532K}, a large-scale benchmark comprising 532K annotated dialogue lines across more than 900 unique characters, necessitating the integration of auditory, linguistic, and visual cues for speaker recognition. (2) We propose \textbf{DramaSR-LRM}, a robust approach built upon a large reasoning model (LRM). DramaSR-LRM is designed to autonomously aggregate contextual evidence via multimodal tool-use, synthesizing diverse inputs to achieve high-fidelity attribution. Experimental results demonstrate that DramaSR-LRM significantly outperforms existing baselines, particularly on short utterances where acoustic biometrics are inherently unreliable. \textit{All the data and code will be made publicly available at the project page: https://www.github.com/198808xc/DramaSR-LRM.}
On-policy self-distillation (OPSD) has emerged as a practical method for training large language models (LLMs) to reason, where a single model acts as both the teacher and the student with different levels of information access. However, recent studies have found that the teacher's dense token-level supervision, conditioned on privileged information, can lead to overfitting to in-domain patterns, suppress exploration, and hurt cross-domain generalization, while also introducing a more fundamental issue: *privileged information leakage*, where the student encodes answer-dependent shortcuts that are unavailable at test time. We introduce **DemoPSD**, a novel framework that resolves such problems through the idea of *selective adoption of teacher guidance*. Instead of fitting the full teacher distribution, DemoPSD steers the student toward a *reverse-KL barycenter target*, a weighted geometric combination of the teacher and student distributions, that naturally balances learning from the teacher with preserving the student's own reasoning capacity. We measure the difference between their distributions and use such a discrepancy to adaptively control the blending at each token position. We provably show that DemoPSD achieves **(1)** *leakage attenuation*, i.e., effective mitigation of privileged information leakage; and **(2)** *exploration preservation*, i.e., preservation of exploration capacity under dense token-level distillation. Extensive experiments on SciKnowEval across four scientific fields show that DemoPSD outperforms both GRPO and SDPO while maintaining higher training entropy and robustly generalizing to out-of-distribution GPQA benchmarks.
Machine learning interatomic potentials (MLIPs) have become a hallmark of AI for scientific simulation. While efforts on new architectures and datasets have led to increasingly accurate and general models, the choice of optimizer for training has largely remained unexplored, defaulting to Adam and its variants in the community. Here, we implement and systematically compare a class of recently proposed matrix-structured optimizers, including Muon, SOAP, and the hybrid SOAP-Muon, for training NequIP and Allegro MLIP models. We find that these optimizers can substantially outperform Adam in both convergence speed and final accuracy. SOAP and SOAP-Muon emerge as robust and consistently strong methods, while Muon only provides partial gains relative to Adam. The improvements are particularly pronounced under partial force supervision. Our results indicate that optimizer choice is an overlooked yet impactful design axis for MLIPs.
Realistic traffic simulation requires agents that imitate logged behavior and can also be steered along interpretable axes. Such controllability enables engineers to isolate variables, reproduce specific edge cases, and test autonomous systems without real-world risk. We introduce Controllable Neural Variational Agents (CNeVA), a controllable simulated-agent framework that learns to infer a per-agent Gaussian behavior latent from per-channel discounted returns via a closed-form conjugate variational update, conditioning a rectified-flow trajectory generator trained on a mixed channel-mask curriculum for classifier-free guidance. To tackle scarcity in reward signals, we propose soft eligibility gates that replace hard binary thresholds with smooth exponential decay, preserving the gradient signal for near-threshold agents. On the Waymo Open Motion Dataset, CNeVA attains competitive realism on the benchmark while exposing per-channel controllability that the higher-ranked imitation models lack. Speed- and acceleration-based steering produces monotone responses without stall-induced reward hacking. Safety controllability is monotone and substantial with the introduction of soft eligibility. We manage to achieve steerable map compliance under a context-residual return measure. Furthermore, our experiment demonstrates that steering metrics must be read alongside physical-plausibility guardrails to avoid reward-hacking confounds.
Models trained via Contrastive Language-Image Pretraining (CLIP) serve as the foundational vision encoders for most modern Large Vision Language Models (LVLMs). Despite their widespread adoption, CLIP models exhibit a critical yet underexplored failure mode: irrelevant text appearing within images confounds visual representations, biasing them toward lexical meaning rather than true visual semantics. This robustness issue, commonly described as a Typographic Attack (TA), exposes a vulnerability that poses a significant risk to safety-critical applications such as autonomous driving. To achieve interpretable and effective robustness against TA, we propose a novel, training-free mechanistic interpretability method. Our method provides sampling-based interpretations of hidden state representations and quantitatively attributes semantic versus lexical focus to individual attention heads. Through probabilistic analysis and circuit mining, we isolate specific Vision Transformer (ViT) components that disproportionately encode lexical information, thereby identifying the mechanistic source of TA. We further show that simple interventions applied directly to the identified circuits, without any additional training, can substantially improve robustness against Typographic Attacks in object classification. These interventions, such as selective adjustment of attention weights, also outperform both supervised and training-free defense methods. Our experiments demonstrate that applying the proposed intervention to the vision encoders of several state-of-the-art LVLMs yields substantial gains in Visual Question Answering accuracy under Typographic Attack interference on RIO-Bench. These results confirm both the efficacy and the generalizability of our mechanistic approach. Code is released at https://github.com/Liu-524/SamplingTAR.
In this work, we focus on SE-RRMs, a symbol-equivariant instantiation of RRMs that exhibits improved extrapolation to larger problem sizes. We propose a neuro-symbolic approach, ``Guiding with Recurrent Reasoning Models'' (G-RRM), which integrates SE-RRMs with symbolic solvers for constraint satisfaction problems. SE-RRMs act as neural solvers that generate full solution proposals and guide classical symbolic solvers, such as backtracking or SAT-based methods like Glucose 4.1 and CaDiCaL 3.0.0, that produce globally correct solutions. Centrally, we investigate when neural guidance with G-RRM improves the search efficiency of symbolic solvers. % Our experiments show that the efficacy of G-RRM depends on two conditions: first, the problem instances must have an expansive combinatorial search space to expose potential gains, and second, the solver architecture must be capable of dynamically overwriting its branching choices to recover when neural hints are imperfect. When these conditions hold, guidance drives median conflict counts to zero and yields significant wall-clock speedups: on $9\times9$ Sudoku, where the SE-RRM correctly solves $91.1\%$ of instances, backtracking accelerates by $33.3\times$ and Glucose 4.1 by $1.70\times$ (median, $p<0.001$), with Glucose 4.1 retaining a $1.17\times$ speedup on perfect-hint $25\times25$ grids. In contrast, CaDiCaL 3.0.0, whose runtime is overhead-dominated and which always respects the injected branching hints rather than overwriting them, shows no significant speedup (median $1.02\times$, n.s.) and even a small significant mean slowdown ($0.90\times$) on $9\times9$. These results delineate the regimes in which neural guidance translates into practical speedups.
Large vision-language models can reason over multimodal inputs by generating textual chains of thought (CoT). A key capability exhibited in CoT reasoning is self-reflection: revisiting earlier decisions and correcting previous errors. However, existing LVLMs often fail to properly attend to visual inputs during reflection, limiting their ability to translate feedback into grounded corrections, especially for out-of-distribution images. To address this issue, we propose a novel reinforcement learning training framework VRRL, with two components explicitly designed to elicit visually grounded self-reflection. First, we randomly mask trajectory prefixes during training to emphasize recovery from incorrect intermediate predictions rather than making early mistakes. Second, we introduce buffered roll-ins from an experience replay buffer to expose the model to diverse failure states that it must learn to correct. We evaluate our approach on visual grounding tasks involving tables and charts, as well as spatial navigation benchmarks. While off-the-shelf and conventionally fine-tuned models degrade substantially under distribution shift, our method substantially improves average out-of-distribution accuracy over standard RL and reflection-oriented fine-tuning baselines by using self-reflection effectively.
Visual token pruning is a crucial strategy for accelerating VLMs by compressing redundant image patches, yet existing methods often fail to preserve critical cues under dense instructions and fine-grained queries. In this paper, we investigate this failure and identify two underlying bottlenecks: the widespread dispersion of textual noise that corrupts dense cross-modal scoring, and the feature fragmentation inherent to standard token selection. To address these issues, we propose Entropy-Aware Dense Pruning (EADP), a framework that reformulates pruning as a structured compression problem. EADP first leverages statistical entropy to quantify and filter out textual noise, yielding a robust, fine-grained instruction relevance score. Subsequently, instead of naive Top-K selection, EADP casts token selection as a submodular maximization problem with a spatial prior, explicitly ensuring a holistic and non-redundant visual representation. Extensive experiments demonstrate that EADP improves the accuracy-efficiency trade-off of VLMs, robustly preserving fine-grained visual cues under strict token budgets while achieving SoTA performance on challenging multimodal benchmarks.
Narration is central to the audiobook listening experience, shaping how listeners engage with and understand the content. This work explores how narration qualities shape an audiobook's appeal, noting that their effects can vary by genre, title, and audience. We extract vocal and acoustic features (e.g., tone, pace, loudness) from LibriVox using pre-trained audio models and analyse their relationship with consumption data (specifically, view-rate) and their interplay with genre and title. Despite limited consumption data, we find that acoustic information alone has a robust association with appeal, even after accounting for title effects. We further validate these findings using more nuanced proprietary engagement metrics. To our knowledge, this is the first systematic computational study linking narration qualities, genre, title, and audiobook consumption, highlighting the potential of data-driven insights to improve audiobook personalisation and narrator casting.
Software tests and code evolve together: a code change should be followed by new or updated tests that record the new software behavior. Yet existing test generation and update benchmarks often isolate the test from the code change, and rely on static metadata that does not verify whether a test is executable or semantically tied to the code change. This makes it difficult to evaluate whether a test automation agent understands how a code change should propagate into the test suite. We introduce TestEvo-Bench, a benchmark of test and code co-evolution tasks mined from software repositories, with two tracks: in test generation, the agent shall write new tests to capture the new software behavior; in test update, the agent shall adapt failing existing tests to the changed software behavior. Each task is anchored to a real commit history and packaged with environment configuration to support execution-grounded metrics such as pass rate, coverage, and mutation score. TestEvo-Bench is also a live benchmark: each task records the timestamp of the test and code changes, and new tasks are periodically mined by our automated pipeline, so evaluation can be restricted to tasks postdating a model's training cutoff to reduce data leakage risk. The current snapshot contains 746 test generation and 509 test update tasks, curated from 59,950 candidate co-evolution records across 152 open-source Java projects. We experiment with four state-of-the-art agents that combine strong harnesses (Claude Code, Gemini CLI, and SWE-Agent) with strong foundation models (Claude Opus 4.7 and Gemini 3.1 Pro). Results show that they achieve up to 77.5% success rate on test generation and 74.6% on test update. However, success rate is materially lower on the most recent benchmark tasks and drops significantly under limited per-task cost.
Whether pairing people with AI helps or hurts is usually reported as a single average effect. Using a real-money prediction market (Polymarket) as an objective, externally resolved benchmark, this pilot shows that the value of human-AI collaboration depends on a specific, measurable form of human capital. Analyzed at the level of the individual forecaster, hybrid performance is trimodal: most people either deferred to the model (matching it) or used it to rubber-stamp a prior guess (performing worse than the model alone), while a minority engaged in genuine complementary reasoning and reached accuracy matching or even exceeding (i.e., lower error than) the market itself. Collaborative traits (perspective-taking, intellectual humility, and curiosity) rather than raw cognitive ability or model benchmarks, distinguished who reached that mode. The results are preliminary but statistically robust, and motivate a pre-registered replication now in preparation.
Vision-Language-Action (VLA) models are fundamentally bottlenecked by the scarcity of expert demonstrations -- triplets of observations, instructions, and actions that are costly to collect at scale. We argue that this bottleneck stems from conflating two distinct learning objectives: acquiring physical competence (how to move) and acquiring semantic alignment (what to do). Crucially, only the latter requires language supervision. Building on this Decomposition Hypothesis, we propose Task-Agnostic Pretraining (TAP), a two-stage framework that first learns transferable motor priors from cheap, unlabeled interaction data -- including discarded off-task trajectories and autonomous robot play -- via a self-supervised Inverse Dynamics objective. A lightweight second stage then grounds these priors in language using minimal expert data. On the SIMPLER benchmark, TAP matches models trained on over 1M expert trajectories while using orders of magnitude less labeled data, yielding a 10% absolute gain over standard behavior cloning. On a real-world WidowX platform, TAP retains 25% success under camera perturbations where internet-scale baselines collapse to 0%, demonstrating that task-agnostic pretraining produces robust, transferable physical representations and offers a scalable path forward for Embodied AI.
Large Language Model (LLM) social simulations are a promising research method, but they are not yet faithful enough to be adopted widely. In this work, we investigate whether the current scaling paradigm in language modeling is likely to close these gaps, or whether simulation fidelity is orthogonal to general capabilities and therefore deserving of more research attention. We use scaling laws to study the relationship between LLMs' compute scale, general capability benchmarks, and the fidelity of social simulation in three representative sub-domains: opinion modeling, behavioral simulation, and longitudinal forecasting. Surprisingly, we discover strong compute scaling in all three settings, using a suite of 85 transformer LLMs with the Qwen3 architecture pre-trained on the DCLM web text corpus under fixed-compute budgets from $10^{18}$ to $10^{20}$ FLOPs. Then we evaluate 35 larger and more capable open-weight models up to 70B parameters, allowing us to predict downstream accuracy from loss. This reveals that the majority of behavioral and opinion simulation tasks will rapidly improve with scale, particularly when they involve populations that are well-represented in English web corpora. Longitudinal forecasting and underrepresented opinions scale more slowly, especially when they are less correlated with general knowledge and reasoning benchmarks like MMLU. In behavior simulation, scaling fails to improve model calibration with human cognitive biases like risk aversion, as well as human heuristics like learning correlated rewards from related tasks. On these tasks, even fine-tuned models fail to noticeably scale up performance from 0.5B to 8B parameters. Taken together, we conclude that scale will improve social simulations in most settings, but outliers exist, and improvements will be less reliable in low-resource domains.
Diffusion transformers (DiTs) achieve state-of-the-art image and video generation, but their multi-step sampling and growing parameter count make inference expensive. Post-training quantization (PTQ) is the natural remedy, yet DiT activations shift across timesteps, prompts, and guidance branches, forcing prior methods to re-fit calibration data for every new checkpoint or modality. We present OrbitQuant, a data-agnostic weight-activation quantizer that bypasses range estimation by quantizing in a normalized, rotated basis. In this basis, a randomized permuted block-Hadamard (RPBH) rotation concentrates each coordinate around one fixed, known marginal regardless of the input, so a single Lloyd-Max codebook serves all timesteps, prompts, and layers of a given input dimension. We extend the same quantizer to weight rows offline, absorbing the rotation into the weights so that it cancels inside each linear layer and only a forward rotation on the activations remains at runtime. The same recipe transfers from image to video with no per-modality tuning. Across FLUX.1, Z-Image-Turbo, Wan 2.1, and CogVideoX, it sets the state of the art for PTQ at several low-bit settings. It also pushes PTQ of image diffusion transformers to W2A4 with usable generation quality.
Post-training large language models (LLMs) without real-world interaction feedback or human-labeled supervision remains challenging, particularly in specialized domains where expert annotations are costly to obtain. Recent annotation-free self-evolution methods address this by using the model's own outputs as supervision signals, constructing a teacher via additional context and aggregating predictions across multiple rollouts through majority voting to produce pseudo-labels. However, these approaches are not without drawbacks: SFT- and GRPO-based variants suffer out-of-domain performance degradation, while reward-based on-policy RL inflates calibration error. In this paper, we propose Neuron On-Policy Self-Distillation (Neuron-OPSD), a data-centric framework for annotation-free self-distillation that leverages internal neuron activations to guide both training-data selection and teacher context construction. The model is then trained via on-policy distillation from the teacher distribution, requiring no ground-truth labels at any stage. Across specialized-domain benchmarks, Neuron-OPSD improves in-domain task performance while preserving cross-domain generalization and mitigating calibration collapse over prior annotation-free baselines. This framework is particularly relevant to settings where online interaction or external supervision is costly or infeasible, and is conceptually distinct from offline RL approaches that rely on logged, reward-labeled trajectories.
Language models are increasingly used to quantify cultural phenomena, but what makes such measurement distinctively cultural? This paper argues that NLP work on culture is a material-discursive practice: the apparatus -- model, data, annotation, evaluation -- participates in constituting the cultural reality it measures, rather than passively recording it. Drawing on Karen Barad's concept of the agential cut -- the contingent boundary between phenomenon and instrument -- I show that the apparatus's substantive design choices draw such boundaries, and that the boundary is entangled from the start because language models have already internalized much of the cultural material they measure. I illustrate this through three case studies on television and film dialogue (measuring structure, interaction, and deviation) and three examinations of the apparatus itself (erasure of cultural markers, attunement to historical material, and agency in an agentic workflow). This big picture analysis proposes a research program that is theory-driven, empirically rigorous, and culturally contingent, treating each agential cut as a conscious commitment, at once methodological and ethical.
Recent research has introduced distributed self-supervised learning (D-SSL) approaches to leverage vast amounts of unlabeled decentralized data. However, D-SSL faces the critical challenge of data heterogeneity, and there is limited theoretical understanding of how different D-SSL frameworks respond to this challenge. To fill this gap, we present a rigorous theoretical analysis of the robustness of D-SSL frameworks under non-IID (non-independent and identically distributed) settings. Our results show that pre-training with Masked Image Modeling (MIM) is inherently more robust to heterogeneous data than Contrastive Learning (CL), and that the robustness of decentralized SSL increases with average network connectivity, implying that federated learning (FL) is no less robust than decentralized learning (DecL). These findings provide a solid theoretical foundation for guiding the design of future D-SSL algorithms. To further illustrate the practical implications of our theory, we introduce MAR loss, a refinement of the MIM objective with local-to-global alignment regularization. Extensive experiments across model architectures and distributed settings validate our theoretical insights, and additionally confirm the effectiveness of MAR loss as an application of our analysis.
We study stabilizer state testing and learning with limited coherent quantum memory. Here an algorithm sequentially receives copies of an unknown $n$-qubit state, but may keep only $k$ qubits of coherent quantum memory between measurements. With unrestricted memory, seminal work of Gross, Nezami and Walter showed how to test $n$-qubit stabilizer states using $6$ copies, which is dimension independent, unlike the learning complexity of $Θ(n)$. We show that this testing-vs-learning separation is lost under memory constraints. More concretely we show that (1) The sample complexity of testing stabilizer states in the $k$-qubit memory framework is $Θ(n-k)$. Our upper bound goes via a novel connection to the hidden shift problem and the lower bound is proven using a novel approach to average case bounds on likelihood ratios via combinatorics of the stochastic orthogonal group. (2) The sample complexity of learning stabilizer states with $k$ qubits of memory, in the non-adaptive framework, is $Θ(n^2/k)$. As a further application of our techniques, we prove an exponential lower bound for purity testing even when the memory may be left coherent throughout the protocol. Our main results identify coherent quantum memory as the resource enabling the usual separation between stabilizer testing and learning. In particular, even with $k=0.99n$ qubits of memory, there is no constant-copy stabilizer tester; furthermore for $k=cn$ qubits of memory (for $0< c < 1$), stabilizer testing is as hard as learning, with both requiring $Θ(n)$ copies.
Application Programming Interfaces (APIs) are essential in software development, enabling web services, mobile apps, and microservices. However, their widespread use introduces significant security risks, highlighting the importance of API security. This paper presents HTTP REST API Learning (HRAL), a novel unsupervised anomaly detection approach that models the structure and behavior of API endpoints directly from network traffic, without relying on predefined rules or documentation. HRAL enables robust detection of malicious activity by understanding how APIs behave and flagging deviations as potential threats. We evaluate HRAL across varying levels of OpenAPI documentation detail and compare it with existing techniques. HRAL achieves strong performance, with an average recall of 82.07% and an F1-score of 87.24%, significantly outperforming alternatives when API documentation is limited. Moreover, our results approach the effectiveness of full API document definitions. When combined with signature-based rules such as the OWASP ModSecurity CRS, our system achieves 100% detection. These results highlight HRAL's effectiveness in real-world, partially documented API environments and its potential as a foundational layer for modern API security solutions.
Autonomous agents are increasingly expected to improve executable policies through feedback, yet existing evaluations often collapse this process into a final score or confound it with open-ended software-engineering progress. We introduce Autonomous Policy Evolution, a controlled evaluation setting in which a harness-model agent repeatedly edits an executable policy system under a fixed interaction budget. We instantiate this setting in EvoPolicyGym, a benchmark built from compact interactive RL environments that evaluates how agents iteratively improve explored policies. On the EvoPolicyGym suite, GPT-5.5 achieves the strongest aggregate rank score and top-two performance on all 16 environments. Beyond leaderboard results, EvoPolicyGym also provides trajectory-level diagnostics that distinguish how agents allocate budget, convert feedback into parametric tuning. These analyses show that strong autonomous policy evolution depends not only on isolated task wins, but on discovering task-appropriate mechanisms and refining policies under bounded feedback.
Time series forecasting remains challenging when the underlying data contain rare but critical extreme events. This issue is particularly important in hydrologic forecasting, where streamflow distributions are often highly skewed and extreme peaks can have substantial impacts on flood monitoring, water resource management, and early warning systems. Although Transformer-based forecasting models have achieved strong performance by modeling long-range temporal dependencies, they typically treat all time points uniformly and may therefore underrepresent rare extreme patterns. In this paper, we propose the Extreme-Adaptive Transformer (Exformer), a forecasting framework designed to explicitly model temporal dependencies involving both normal and extreme events. Exformer introduces an extreme-adaptive attention mechanism composed of three sparse components: Local, Stride, and Extreme. The Local and Stride components capture short-term and periodic temporal dependencies, respectively, while the Extreme component selectively models event-aware dependencies between normal and extreme streamflow patterns. Experiments on four real-world hydrologic streamflow datasets show that Exformer achieves superior 3-day forecasting performance compared with state-of-the-art baselines. Our findings demonstrate that explicitly incorporating extreme-aware attention improves the forecasting capacity of Transformer models on imbalanced time series with rare but consequential events.
Agentic coding assistants are increasingly given extra capabilities, such as browser based testing tools and design oriented system prompts, on the assumption that more capability yields better software. This study tested that assumption directly. Ninety independent agent runs built the same application, a real time retrospective board, from one detailed specification, each scored on a fixed 14 criterion functional rubric (42 point maximum) and a visual quality review. The runs spanned several model generations, two agent harnesses, two reasoning effort levels, a testing tool, and two design oriented prompts. Capability tier dominated: frontier models clustered near the ceiling while a low cost local model fell to 24 to 37 points. A criterion level analysis revealed what run totals conceal. Container deployment was the dominant defect, failing first try in 44 percent of runs, with its failure rate shifting sharply across model generations while mean totals moved less than a point. The testing tool raised cost by 42 to 68 percent without improving functional score or reliability, even on interface visible criteria. Raising reasoning effort from High to xHigh lifted first try perfect runs from 28 percent to 89 percent and cut corrective prompts about five fold, for 9 to 29 percent more cost. A design oriented prompt raised visual quality, 4.5 versus 3.0 on a 5 point scale, without lifting function, and a one paragraph paraphrase of its directive reproduced the entire lift. The practical lesson is to match the fix to the failure: most first run failures came from weak reasoning, which a stronger model or more effort prevents, not from visible flaws a checking tool would catch.
Scalable and reliable grading of command-line examinations remains a challenge in computing education, where rising enrolments make manual marking difficult and rule-based autograders cannot handle partial credit, equivalent solutions, or syntactic variation. This paper evaluates whether four frontier Large Language Models (GPT, Claude Opus, Gemini, and GLM) can approximate expert judgment when grading short Linux/bash command responses. The study adopts a four-level cognitive taxonomy that combines cognitive complexity and operational impact, ranging from information retrieval (L1) and basic file manipulation (L2) to structural operations (L3) and advanced system management (L4). The models were tested with two prompt variants, a minimal baseline and a rubric-enhanced version, on 1200 real responses from second-year Computer Engineering students independently graded by three expert instructors. Gemini~3.0 Pro with rubric-guided prompting achieved the highest human-AI agreement (ICC(3,1) = 0.888, MAE = 0.10, Bland-Altman bias = -0.014). Agreement declined consistently as taxonomy level increased, with the largest discrepancies at higher levels. Across all models, rubric quality had a larger effect than provider choice, with structured prompts consistently improving agreement. These results show that question complexity is a reliable predictor of the difficulty LLMs face in grading accurately, and they establish a principled, taxonomy-based framework for determining which questions are suitable for AI-assisted grading and which require human review, while also providing a transferable evaluation protocol and prompt templates.
Reinforcement learning (RL) can overcome the demonstration-coverage limitation of imitation learning (IL) by allowing robots to improve through trial-and-error interaction beyond the states observed in demonstrations. However, deploying RL on real robots remains constrained by high interaction costs, since each physical rollout is costly and reflects only one realized action-outcome path. To address this challenge, we propose WorldSample, a physically grounded data augmentation framework for real-robot RL that closes a real-synthetic loop between physical rollouts, world-model generation, and policy improvement. Grounded on real rollouts, WorldSample generates high-fidelity synthetic transitions through a post-trained world model, which greatly lowers the visual hallucination. Specifically, rather than simply using these transitions as real-world experience, WorldSample introduces Policy-Paced Learning (PPL) to regulate the training process through sample selection and scheduling, balancing useful augmentation against value overestimation and mitigating the hallucination-induced noise. Experiments on robot manipulation tasks involving contact-rich and precise tasks show that WorldSample improves policy success rate by 28% while reducing training steps by 59% compared with baselines. Furthermore, WorldSample improves world model visual fidelity by 19.4dB in PSNR and 0.47 in SSIM over demonstration-only post-training, validating the effectiveness of the real-synthetic loop for both policy and world model performance.
Federated learning (FL) enables collaborative model training across distributed devices without sharing raw data, making it suitable for privacy-sensitive robotic sensing applications. However, multi-agent systems generate heterogeneous and non-independent and identically distributed (non-IID) multimodal sensor streams that degrade conventional FL algorithms, while classical fusion modules introduce substantial parameter overhead and communication cost. This paper proposes QFedAgent, a hybrid quantum-classical personalized FL framework for multi-agent activity recognition. The approach integrates a variational quantum circuit fusion module that models accelerometer--gyroscope interactions through quantum state encoding and entanglement, requiring only 72 quantum rotation parameters versus 33K in classical multi-layer perceptron-based fusion, achieving approximately 10x total parameter reduction. Experiments on the OPPORTUNITY dataset under subject-based non-IID partitions demonstrate 97.7% mean test accuracy, confirming that parameter-efficient quantum fusion remains competitive with conventional federated baselines.
Active Few-Shot Learning (AFSL) adapts LLMs to specialized domains by identifying the most valuable unlabeled samples for annotation and use as few-shot demonstrations, effectively reducing human annotation costs while promoting high performance. However, existing methods typically rely on output-level signals for sample identification, such as predictive entropy or semantic similarities with test-time data based on external embeddings, which often overlook models' internal dynamics, which could pinpoint specific knowledge gaps. To bridge this gap, we propose NeuFS, a Neuron-Aware Active Few-Shot Learning framework that shifts the selection paradigm from output-level proxies to models' internal dynamics. NeuFS utilizes neuron activation patterns to represent sample directly, and includes a dual-criteria selection strategy that: (1) ensures few-shot sample diversity with neuron patterns for broader example coverage, while (2) prioritizing on identifying informative and challenging few-shot samples LLMs tend to hallucinate by quantifying neuron consensus. Experiments on three datasets demonstrate that NeuFS excels in both reasoning and text classification tasks, outperforming existing AFSL baselines. Ablation studies further highlight that internal neuron activations provide a more principled and effective selection signal than external embeddings, validating the superiority of the proposed NeuFS.
Autonomous robots often need to move their camera before they can act: to inspect an object, reveal an occluded region, or obtain a view that responds to a user's intent. While vision-language navigation translates instructions to base motion and vision-language-action policies map instructions to manipulation actions, language-conditioned camera motion remains comparatively underexplored as a first-class action. We formulate language-conditioned camera motion generation: given a current RGB observation and a free-form natural-language intent, predict a relative target camera pose for the next observation. This task is inherently non-trivial: viewpoint changes are driven by latent perceptual intentions, and a valid motion may operate at different semantic granularity, from entering a room to looking around a corner, inspecting a visible object, or revealing an occluded detail. To model this structure, we mine multi-intention camera-motion supervision from egocentric video, pairing plausible intents and observation-gain descriptions with relative SE(3) target poses. We propose LIME, a vision-language camera-motion generator that combines an auto-regressive observation-gain output with a continuous flow-matching pose head. This design lets the model jointly predict what the next view should reveal while representing multi-hypothesis target views. Across experiments and downstream robotic tasks, we show that LIME can learn to actively choose camera poses from passive human video, turning ordinary egocentric recordings into supervision for intent-aware active perception.
Natural Language Processing (NLP) has traditionally been published in its core disciplinary venues like ACL. However, advances in Large Language Models (LLMs) has led to a blurring of the disciplinary lines between NLP and general Machine Learning (ML), with authors regularly publishing in venues from both fields. Here, we ask whether the disciplinary center of gravity is shifting. Using NLP research published from 2010 to 2026 and studies of both established and new authors, we find that a migration is taking place. First, comparing the pre- and post-LLM eras, established authors lost 19.2pp of share at flagship *ACL main-conference tracks while gaining 14.8pp in the newer Findings tracks, and general ML venues rose 8.6pp, even when adjusting for parallel growth in the fields. Second, among newer authors who debut with at least three first-author NLP-topic papers, the share whose work appears mostly at *ACL venues fell from 84% (2019) to 74% (2024), while the share appearing mostly at general ML venues rose from 5% to 21%. Using causal inference techniques, we estimate that these general ML venues confer a significant citation premium, which influences venue selection. Together, these results point to a significant shift in where NLP research is published.
Here we describe the quantum gas analysis and inference (Q-GAIN) Python package, which enables rapid deployment of machine learning (ML) and physics-informed analysis techniques for cold-atom experiments. Out of the box, Q-GAIN implements classification, object detection, and physics-informed metrics for feature detection in images of atomic Bose-Einstein condensates (BECs). Q-GAIN encourages a natural, module-based workflow: starting with data loading and preprocessing, followed by ML-based feature identification, and ending with conventional analysis techniques. We demonstrate this modularity by configuring Q-GAIN for three ML tasks. First, we demonstrate the basic workflow of the Q-GAIN framework by implementing the standard task of classifying handwritten digits from the MNIST dataset. Then, we re-implement our earlier soliton detection (SolDet) package in the Q-GAIN framework, enabling the detection and analysis of solitonic excitations in time-of-flight data. Finally, we develop an object-detection tool that identifies quantized vortices in images of ring-shaped BECs.
Large Language Models (LLMs) have demonstrated remarkable capabilities in 3D indoor synthesis for Manhattan environments. However, existing methods often fail to capture plausible object layout patterns in non-Manhattan settings, primarily because they struggle to model non-orthogonal spatial relationships, leading to high geometric violations and low physical fidelity. To address this challenge, we propose SPG-Layout, a novel text-driven framework designed to generate physically plausible indoor scenes within complex non-Manhattan environments. Specifically, we first utilize statistical priors of object distributions to guide the training process, enhancing environmental understanding and fidelity. Furthermore, mirroring human design workflows, we adopt a hierarchical layout strategy that prioritizes the placement of large objects, thereby substantially minimizing layout violations. By synergizing these components, SPG-Layout achieves a balanced optimization of semantic realism and physical plausibility. To evaluate performance in these complex settings, we constructed a new benchmark comprising 500 diverse non-Manhattan environments. Extensive experiments demonstrate that SPG-Layout consistently and significantly outperforms existing methods across both Manhattan and non-Manhattan environments. The code will be publicly released.
Image encoders trained with LeJEPA can deliver strong features for downstream tasks, but, like other image-level self-supervised methods, typically require large training datasets. Aligning representations at the level of objects rather than whole scenes promises greater data efficiency, but doing this in a completely self-supervised way, effectively jointly partitioning a scene and representing its objects, is unstable: the two are locked in a cyclic dependency, partitioning requires meaningful representations, while meaningful representations require consistent partitioning. We sidestep this instability by taking object masks as given during training, using cheap, off-the-shelf SAM proposals. We extend LeJEPA - whose distributional anti-collapse objective ports naturally from whole images to variable-sized sets of objects - to align object-centric representations rather than whole images. An additional instance-separating loss, which treats other objects in the same scene as negatives, further boosts downstream performance. Across two model scales and 10-100% of COCO, object-level LeJEPA outperforms image-level LeJEPA on tracking (DAVIS), classification (ImageNet-1k), segmentation (ADE20k), and re-identification (NAVI).
Decision-time planning with action-conditioned world models has become a popular paradigm for embodied control. However, the standard planning cost judges a candidate solely by how close its predicted terminal state lies to the goal, leaving the realizability of the intermediate transitions unchecked -- a predicted trajectory can look convincing while the environment rollout drifts away from it. In this paper, we propose ACID, a decision-time planning framework that introduces cycle action consistency: the action inferred backward from a predicted transition by an inverse dynamics model should recover the one that was conditioned on. We fold this per-step residual into the planning cost via a scale-invariant adaptive weight. Across four action-conditioned world models and six tasks spanning rigid and deformable manipulation, articulated control, and visual navigation, ACID consistently improves planning and matches the baseline's accuracy with substantially less planning compute.
Steering and monitoring activations in Large Language Models (LLMs) are increasingly used for both safety and interpretability. Early work assumed behaviours are encoded along single linear directions, but recent findings suggest complex behaviours, such as the refusal to answer harmful queries, live in multi-dimensional subspaces. However, existing methods for extracting these subspaces are computationally expensive, which becomes prohibitive on reasoning models who produce long reasoning traces. By adapting the Recursive Feature Machine (RFM) algorithm -- which can be computed efficiently -- with a probe-informed initialization, we are able to identify the multi-dimensional refusal subspace in seconds, on reasoning (Qwen 3) and non-reasoning (Qwen 2.5) models. While RFM allows for faster subspace identification, it also showed better performances on the ablation task than its alternatives. More work is planned to better understand the relations between subspaces found by different methods. If confirmed, RFM could be a cheap and scalable complement to existing subspace-extraction methods in LLMs.
Large Language Model (LLM) inference workloads are a rapidly growing contributor to data center energy consumption. Optimizing these deployments requires matching specific LLMs to the most efficient GPUs, but operators currently lack the tools to do so without exhaustively profiling each combination. While some predictive models exist, they still require profiling data and struggle to generalize to hardware unseen during training. To address this, we introduce \textit{WattGPU}, featuring two predictive models for mean GPU power draw and Inter-Token Latency (ITL). Our approach leverages only publicly available LLM metadata and GPU specifications, eliminating the need for hardware access or profiling while enabling generalization to unseen NVIDIA server-grade GPUs and LLMs. We evaluate our models using rigorous leave-one-GPU-out and leave-one-LLM-out cross-validation on a dataset of 42 open-source LLMs (0.1B--27B parameters) and 8 GPUs under both offline and server scenarios. The mean power draw model achieves a median absolute percentage error of $\leq3.4\%$ for offline and $\leq13.5\%$ for server scenarios on unseen GPUs, while the latency model achieves $\leq8.5\%$ in server mode, both maintaining strong GPU ranking correlations for server scenarios (Kendall $τ\geq0.76$). Compared to standard physically grounded baselines -- Load-Scaled Thermal Design Power (TDP) for power draw and roofline for latency -- our models reduce median absolute percentage error by approximately 4$\times$ on unseen LLM-GPU combinations for server scenarios or approximately 2$\times$ for completely unseen GPUs. WattGPU's data and code are publicly available at https://github.com/maufadel/wattgpu.
How can Large Language Models (LLMs) solve problems they currently cannot? Repeated sampling scales test-time compute but GPU cost grows linearly with attempts, while reinforcement learning (RL) with verifiable rewards improves single-attempt accuracy at the expense of sample diversity. Both strategies ultimately fail when the base policy has near-zero probability of producing a correct solution: no amount of sampling or gradient signal can overcome a search space that is simply too large. We take a different approach: rather than sampling harder, we make the task easier by decomposing problems into smaller, independently solvable sub-functions whose implementations can be recombined. Since off-the-shelf models are not trained for this modular generation, we introduce DecompRL, an RL algorithm that explicitly learns to decompose and implement hierarchical code structures. Recombining $k$ implementations of $n$ modules yields up to $k^{n}$ candidate solutions, shifting the bottleneck from GPU inference to cheap CPU evaluation and cutting GPU token cost by $\sim$50$\times$. On LiveCodeBench and CodeContests (Qwen~2.5~7B, Code World Model~32B), DecompRL outperforms standard and diversity-optimized RL baselines beyond $10^5$ tokens per problem, solving problems that standard generation cannot reach.
Coding agents are capable; human oversight is the bottleneck. Unconstrained agents introduce security risks, erode codebase scalability, and make human review increasingly costly. We argue that the same methods used for decades to manage large human engineering teams: access control, network policies, strict coding conventions enforced by tooling; transfer directly to coding agents, and are cheaper (in token) than recent agentic scaffolding. We sketch a start-to-end system on this principle, and report a controlled experiment in scalable oversight: a small reviewer (Gemma 4 e4b) inspects a Python codebase containing 11 inserted backdoors. Recall rises from 54.5% (unconstrained, no tools) to 90.9% (constrained substrate plus a ~200-LoC `docs` CLI), with substrate and tools contributing independently. We choose Python deliberately: substrate-level oversight gains are largest where the language gives the fewest guarantees by default; the principles extend to languages like Rust.
NASA and its data centers hold thousands of geoscience datasets and tools like Worldview, Giovanni, the Science Discovery Engine, and Harmony. Finding the right one is hard even for domain experts. We present an agentic search system, deployed as a public service for the geoscience community, that takes a natural-language research query and returns the matching datasets and tools. We demonstrate that, in the era of large language models, the latent value of knowledge graphs (KGs) can be substantially amplified through agentic search. From the NASA Earth Observation Knowledge Graph (NASA EO-KG) we derive NASA-EO-Bench, an open benchmark of 47k query-dataset pairs (21k task-based queries). A neural scorer fine-tuned on NASA-EO-Bench beats cosine and BM25 baselines. Further combining it with BM25 via score fusion raises both Recall@10 (R@10) and MRR by over 5x. On top of this supervised pipeline, we add a zero-shot agentic reranking stage that, without any additional training, lifts MRR by 28% on a stratified N=200 subset, showing that LLM reasoning is complementary to supervised retrieval.
While Vision Transformers have achieved remarkable success across computer vision and language applications, the geometric evolution of their internal representations throughout training remains insufficiently understood. Existing analyses primarily focus on attention mechanisms and downstream performance, leaving the evolution of representation geometry largely unexplored. In this work, we present Transformer Geometry Observatory-II (TGO-II), a representation geometry analysis framework designed to investigate how Transformer representations evolve during supervised training. TGO-II analyzes Vision Transformer (ViT-Small/16) representations using Centered Kernel Alignment (CKA), Singular Vector Canonical Correlation Analysis (SVCCA), Two-Nearest Neighbor Intrinsic Dimensionality (TwoNN-ID), and token covariance analysis. Our experiments reveal three key observations. First, both CKA and SVCCA progressively decrease throughout training, indicating increasing representational specialization across Transformer layers. Second, intrinsic dimensionality consistently increases before stabilizing, suggesting progressive expansion of the representation manifold into a larger set of locally accessible degrees of freedom. Third, token covariance and coupling analyses demonstrate that strong token interaction structure persists throughout training, challenging the hypothesis that increasing representational complexity arises primarily from progressive token independence. These findings suggest that representation complexity and layer specialization emerge simultaneously during training. Manifold expansion appears to occur without token decoupling. Together, these observations motivate a new hypothesis in which Vision Transformers increase representational complexity through progressively richer transformations while preserving strong token interaction structure during learning.
LLM-based retrieval-augmented generation (RAG) is increasingly used for automated fact-checking (AFC) and related tasks. By grounding LLM outputs in retrieved evidence, RAG-based systems provide transparent justifications while allowing external information to be updated independently of the underlying model. However, existing approaches often assume retrieved evidence is reliable, although real-world information may be conflicting, outdated, and can originate from unreliable or biased sources. Recent work on *source-critical reasoning* addresses this challenge through media background checks (MBCs) (Schlichtkrull, 2024), which assess the credibility of evidence sources to support downstream fact verification. However, generating MBCs relies on costly proprietary search APIs, limiting reproducibility. To mitigate this issue, we introduce MEDIAREF, a publicly available knowledge store of web-sourced documents that enables reproducible, low-cost evaluation of MBC generation across 200 media sources. We describe a reproducible methodology for constructing and updating the collection, assess widely used LLMs on the MBC generation task, and demonstrate that MEDIAREF supports higher-quality MBC generation through both automatic and qualitative evaluation.
This paper describes the participation of HULAT2-UC3M in the Spanish track of MER-TRANS 2026, a shared task on multilingual Easy-to-Read translation. Three fully automatic Spanish runs were submitted. RUN1 and RUN2 used a LangGraph-based multi-agent workflow combining Gemini 2.5 Flash and RigoChat-7B-v2, parallel generation strategies, internal quality signals, Event-Condition-Action routing, controlled editing and traceable decisions. RUN1 used the base workflow, while RUN2 activated an additional lexical-support layer based on a glossary and lexical resources. RUN3 was a RigoChat-based generate-evaluate-regenerate baseline with prompt engineering and LoRA-based adaptation. The official leaderboard reports BLEU-Orig, BLEU-Gold, SARI and BERTScore. During development, additional internal signals were also inspected, including semantic fidelity, readability, lexical simplicity, syntactic clarity and factual consistency. According to official SARI, RUN1 was the best HULAT2 run, with 44.0543 points, followed by RUN2 with 43.1049 and RUN3 with 38.5136. These results indicate that, in this task setting, signal-guided multi-agent routing outperformed the linear regeneration baseline. They also show that adding lexical support did not automatically improve reference-based scores. Further segment-level and document-level analysis are required to assess readability, factual consistency and user-oriented adequacy.
Recent advances in agentic AI are producing increasingly complex autonomous systems that integrate large language models, world models, optimization engines, specialized neural architectures, autonomous platforms, and human operators. While much current research focuses on improving reasoning capabilities, safety-critical real-time deployment also requires bounded and verifiable coordination among heterogeneous components operating concurrently under uncertainty. Software-mediated coordination presents fundamental limitations in domains where bounded latency, deterministic coordination, and enforceable safety guarantees are essential. Hence, we propose a hardware-enforced semantic coordination architecture in which selected coordination semantics are implemented directly at the hardware level via field-programmable gate arrays (FPGAs). The approach builds on the Topic-Based Communication Space Petri Net (TB-CSPN) framework, which separates semantic reasoning from interaction management. In this approach, selected TB-CSPN coordination mechanisms are mapped onto FPGA primitives, creating a hardware-native semantic coordination layer. Focus is not on acceleration, but on enforcing temporal synchronization, semantic gating, authorization constraints, and bounded coordination behavior directly in hardware. Semantic reasoning remains adaptive and software-driven, while embedded coordination semantics become deterministic.
Personalization changes what a model says to a user; we show that it can also change the reasoning trajectory used to justify the response. Modern LLMs personalize interactions by storing user attributes, preferences, and prior context, then injecting this information into future prompts. We study whether such memory reshapes reasoning on open-ended questions where no single ground-truth answer exists. To quantify this effect, we introduce DRIFTLENS, a ground-truth-free framework that maps each expressed reasoning step to a value category and measures divergence between a question's no-memory trajectory and its trajectory under injected user-attribute memory. We first validate that DRIFTLENS distinguishes content-free pragmatic noise from substantive reasoning changes. Across four LLMs and 10 user-attribute categories, including age, occupation, and disability, user-attribute memory induces medium-to-large reasoning drift above each model's pragmatic-noise floor, even when final answers remain fluent, on-topic, and plausible. We then evaluate GRPO- and DPO-based post-training methods for reducing drift. Both reduce drift, but neither uniformly dominates; effects on downstream capability, helpfulness, and instruction following are model-and reward-dependent. These results suggest that memory-induced reasoning drift is a measurable and only partly mitigated failure mode of personalized language models.
Over 285 million people worldwide live with a visual impairment, for whom everyday tasks such as avoiding obstacles, locating personal belongings, recognizing familiar faces, or handling cash remain persistent obstacles to personal autonomy. Existing assistive applications are typically limited to recognizing predefined categories, depend heavily on cloud connectivity, or require dedicated hardware. We present VisionAId, an Android application that turns a commodity smartphone into a real-time visual assistant. The system integrates six on-device deep learning models (metric monocular depth estimation, instance segmentation, visual and facial embeddings, face detection, and a custom banknote detector) running entirely through ONNX Runtime, with an optional cloud large language model (Google Gemini Flash) used only for narrative scene description and automatic object labeling. A distinctive contribution is a few-shot pipeline for personal objects: the user photographs an object from several angles, and the system later locates that specific instance in the environment, guiding the user toward it with augmented-reality markers, spatial audio, and distance-proportional haptics. All feedback is multimodal (Romanian speech synthesis, voice commands, vibration). On a reference device (Samsung Galaxy S21 Ultra), INT8 quantization reduces depth latency from ~1200 ms to ~491 ms, the custom banknote detector reaches an mAP@50 of 0.986, and metric depth is calibrated to below 1 cm of error within 3 m.
Compiler missed optimizations refer to cases in which compilers failed to optimize certain code. It takes many compiler developers' efforts to implement or patch such missed optimizations. In this paper, we present a systematic study of how well agents patch compiler missed optimizations. We identify a significant challenge that patching a missed optimization requires more than just fixing the reported case, and instead requires generalizing to similar cases. We construct a benchmark of real-world LLVM missed optimization issues and compare agent-generated patches with patches from developers in terms of optimization scope. Our results show that coding agents often optimize the given examples, but many generated patches either cover only part of the developer-intended scope or partially overlap with it; in some cases, they further generalize beyond the reference patch. We further introduce historical-knowledge augmentation techniques that leverage prior LLVM optimization pull requests through retrieval and distillation, showing that they improve developer-aligned generalization and yield practical benefits when applied to real-world IR.
LLMs stage a new form of cultural encounter that is massive, automated, and monolingual. Literary disciplines have always negotiated cultural struggles with comparative reading of literature, narratological and poetic analysis, critical theory, world literature, and translation. These tools have now become indispensable for building culturally literate AI. The essay develops a layered framework toward more nuanced textual models and pluralistic interpretations of AI, emphasizing the natural intersections of literature and AI development, connecting current debates in critical theory with structural monolingualism, and suggesting a new application of world literature approaches to address global AI textuality through macrostructure, circulation, and untranslatability.
Evaluations of LLM personas via psychometric questionnaires typically rely on aggregate scores, discarding within-instance correlation structure. We test whether this geometric structure is intrinsic or frame-dependent. Constructing within-instance correlation matrices from IPIP-50 responses, we analyze geometry on SPD manifolds under manipulated question orderings in GPT-4o simulating American and Chinese-American personas. We find that persona expression comprises two dissociable components: aggregated features (Big Five scores) degrade under randomization (21% drop) but are frame-robust; geometric features (SPD manifold) collapse under frame misalignment (42% drop) but recover substantially (to 84%) under shared frames, surpassing aggregated features (76%). This collapse-recovery pattern reveals that persona geometry is not intrinsic but a frame-dependent coordination pattern encoding information invisible to aggregation. Our findings establish a dual-nature framework for LLM personas, frame-dependent geometry versus frame-robust aggregates, necessitating frame-aware evaluation and challenging static trait conceptions.
Quantum Fast-Weight Programmers (QFWPs) store temporal information in dynamically programmed variational-circuit parameters rather than in nonlinear recurrent hidden states, offering a practical route to quantum sequence modeling. Self-Modulating QFWP improves this framework by using input-dependent gates for both new fast-weight updates and the accumulated fast-weight state, but its unbounded old-state multiplier can diverge in long-sequence regimes. We propose a bounded old-state modulation rule that applies a sign-preserving tanh gate only to the recurrent memory branch while leaving the additive update and new-update modulation unchanged. We evaluate standard QFWP, full Self-Modulating QFWP, Only-New, and Only-Old variants on two CUDA-Q quantum-dynamics forecasting tasks and on Milan SMS telecommunication activity prediction. The quantum-dynamics results show that old-state modulation is the most consistent source of improvement over Standard QFWP, and that bounding the old-state gate removes long-sequence divergence while improving aggregate robustness. On Milan SMS forecasting, the original unbounded Self-Modulating QFWP converges across the tested grid and shows its clearest gains at longer input windows, with behavior close to the Only-Old ablation. These findings identify accumulated-memory modulation as the key mechanism of Self-Modulating QFWP and bounded old-state gating as a targeted stabilization strategy.
Monocular relative pose sensing is a central perception problem in non-cooperative rendezvous and on-orbit servicing. In spacecraft images, however, weak surface texture, thin appendages, illumination changes, and partial occlusion often leave only sparse and unstable geometric evidence. This article presents GAP-GDRNet, a geometry-aware attention-enhanced framework for monocular RGB-based 6D pose sensing. The method follows the geometry-guided direct regression paradigm of GDR-Net and modifies two points in the pipeline: an attention-based feature refinement (AFR) module is placed before dense geometric prediction, and a patch-level geometric self-attention (PGSA) module is inserted into Patch-PnP. AFR reinforces global spacecraft structure together with local weak-texture cues; PGSA then relates downsampled geometric patches before final pose regression. A Blender-based annotation process supplies target masks, visible-region masks, dense model-coordinate maps, camera intrinsics, and 6D pose labels for supervised training.
LLM coding agents increasingly rely on third-party agent skills from public marketplaces, which execute with the agent's privileges and create a software supply-chain attack surface: a malicious skill can steal credentials, exfiltrate source code, or install backdoors. Existing defenses use static skill scanners based on pattern matching or LLM-as-judge analysis, but it remains unclear whether they withstand adaptive evasions that preserve malicious behavior while changing payload appearance. This paper first presents an adversarial study of existing skill scanners through SkillCloak, a payload-preserving evasion framework that keeps the attack semantics intact while transforming their visible form. SkillCloak uses two complementary strategies: Structural Obfuscation, which rewrites visible payload indicators into semantically equivalent forms, and Self-Extracting Skill (SFS) Packing, which hides malicious components from the install-time view and restores them during agent execution. Across eight scanners and 1,613 in-the-wild malicious skills, SFS Packing bypasses every scanner at over 90%, while Structural Obfuscation bypasses over 80% on most static scanners and reaches 96% on a hybrid scanner, showing that appearance-based auditing is insufficient. Motivated by this finding, we propose SkillDetonate, a behavior-centric runtime auditor that executes skills in a sandbox and detects malicious effects through OS-boundary information-flow evidence rather than install-time appearance. SkillDetonate combines on-demand closure lift, which observes instructions materialized during execution, with marker-based taint analysis, which tracks sensitive-data flows across the agent context, files, processes, and network operations. The results show that SkillDetonate detects 97% of attacks at a 2% false-positive rate and sustains 87% detection on real-world malicious skills.
Large Language Model (LLM)-based agents increasingly automate software engineering tasks through reusable skills, natural-language instruction documents that guide planning and execution. Open skill marketplaces enable users to assemble agents by co-activating community-contributed skills, but marketplace operators typically audit skills in isolation. As a result, individually benign skills may interact to redirect an agent toward unintended objectives, which we term implicit intents. Detecting such intents is challenging because the effect emerges only through skill composition, execution environments are often unavailable at admission time, and the space of possible co-activations grows exponentially with marketplace size. In this paper, we formulate implicit-intent discovery as a fuzzing problem over skill compositions, where skill compositions are the unit under test, planning artifacts expose agent intent before execution, and deviations from a skill-free baseline serve as a differential oracle. Based on this formulation, we propose skillfuzz, the first execution-free testing approach that extracts structured skill contracts and uses contract-guided Monte Carlo Tree Search to prioritize potentially conflicting compositions. Across representative skill-marketplace workloads, skillfuzz discovers over 1,000 distinct implicit intents under a fixed query budget, confirms more than 80% of the highest-risk flagged compositions during execution-time validation, and identifies substantially more high-severity implicit intents than alternative search strategies while exploring only a fraction of the pairwise interaction space they require.
Transformer architectures have shown strong potential in time series forecasting, where multi-head self-attention is widely used to capture temporal dependencies across historical timestamps. However, standard self-attention has quadratic time and memory complexity with respect to the look-back length. This cost may limit its use in resource-constrained or high-throughput forecasting systems, where fast and memory-efficient inference is important. Through qualitative and quantitative analyses, we observe that self-attention maps in time series forecasting often contain redundant patterns across different timestamps. This phenomenon can be related to the repeated temporal patterns and relatively stable temporal correlations in many real-world time series. Motivated by this observation, we propose Self-Gating Attention (SGA), a plug-and-play attention mechanism that represents the attention score with a shared learnable matrix and an input-dependent residual component. The shared matrix captures common attention patterns, while the residual component captures input-dependent variations. In this way, SGA avoids the query and key projections used in standard attention score computation, leading to linear time and score-matrix memory complexity with respect to the look-back length. We integrate SGA into several forecasting backbones and compare it with standard self-attention and lightweight attention variants on nine publicly available real-world datasets covering electricity, finance, weather, medical monitoring, human activity, and climate records. The results show that SGA improves inference efficiency on public benchmarks while maintaining competitive forecasting performance against state-of-the-art attention mechanisms. These benchmark results provide deployment-oriented evidence.
Humans can selectively attend to a target sound and estimate its direction in complex scenarios, whereas such selective localization remains challenging for current deep learning-based systems. Sound source localization (SSL) has achieved remarkable success with deep learning, yet most methods localize all active sources without selectivity. Conversely, target sound extraction (TSE) extracts sources using multimodal prompts but typically fails to preserve the multichannel spatial information required for accurate localization. To bridge this gap, we formulate the task of prompt-guided selective target sound localization and propose SelectTSL, an end-to-end architecture that localizes only the user-specified target in multi-source acoustic scenes. Specifically, we design a target-aware selective localization strategy that employs a Prompt-Guided Selective Attention Module (PGSA) to generate prompt-informed embeddings. These embeddings guide an inter-channel phase difference (IPD) enhancer to refine raw phase cues, fusing with target magnitudes to jointly estimate direction of arrival (DoA) and target-source cardinality, i.e., the number of target sound sources. This coupled design effectively focuses on the user-specified target spatial cues for selective localization and also handles time-varying numbers of target sources. Extensive experiments on both synthetic data and real-world recordings demonstrate that our proposed method consistently outperforms other baselines and exhibits robust generalization to real acoustic environments.
Hierarchical Navigable Small World (HNSW) graphs serve as the industry standard due to their logarithmic complexity and strong empirical performance. However, HNSW relies on greedy graph traversal, a heuristic that provides no theoretical guarantees of correctness. In this paper, we propose a novel "Certify-then-Rectify" framework that bridges the gap between the speed of heuristic search and the rigor of exact retrieval. Rather than discarding HNSW, our approach first employs a distribution-free statistical certifier to dynamically evaluate the quality of a standard HNSW search with minimal overhead. If certification indicates that the retrieved neighbors are of low quality, the framework safely escalates to a rigorous exact recovery algorithm. To make this exact recovery computationally feasible, we reinterpret the HNSW graph as a geometric spanner and utilize Extreme Value Theory to stochastically estimate its maximum empirical stretch factor. This allows us to mathematically bound the maximum distance of true nearest neighbors. Extensive evaluations on benchmark datasets demonstrate that our tiered framework delivers the average-case speed of HNSW while ensuring the worst-case correctness of exact search and outperforming other applicable approaches.
With the growing adoption of AI-powered coding assistants, organizations and developers are increasingly seeking to optimize their interaction with these tools. Prior research has largely focused on output quality and productivity gains, with limited attention paid to developers' well-being and interaction experiences. This paper presents a developer-centered empirical mixed-methods study to investigate how professional developers engage with Generative AI (GenAI) in their natural work environment. Controlled data collection sessions are combined with natural work periods. Results show that developers are generally satisfied with GenAI, particularly for monotonous, repetitive, and structured tasks, and report perceived efficiency and productivity gains. Copilot interaction type preferences differ by task type and complexity: While both in-code suggestions and chat-based prompting independently improve task efficiency and reduce perceived workload, combining these interaction types within a single task diminishes benefits. We propose a rule-of-thumb for selecting an interaction type based on task characteristics. During development-heavy tasks, results indicate that perceived cognitive load arises from AI interaction, while perceived productivity depends on AI output quality. Participation in this study positively influenced developers' awareness and intentional use of GenAI tools. These findings demonstrate the value of real-world, mixed-methods study designs to understand GenAI tools and developers' experiences with them.
Vibe coding democratizes software development by allowing users to generate code via natural-language (NL) interaction with large language models (LLMs). However, the code is reliable only when it faithfully implements the user's intent, which is difficult and labor-intensive for users to validate. Existing validation methods either rely on LLM-assisted automated testing, which suffers from prompt ambiguity and model fallibility, or involve users only in partial software artifacts such as prompts and test cases, which may overlook corner cases and program details. Motivated by a bug study of LLM-generated code, we find that detailed human feedback is essential, as failures often stem from underspecified requirements or subtle semantic deviations. This paper presents verifiable literate programming (VLP), a human-in-the-loop framework designed to make the review/validation process of LLM-generated code accessible to users at all programming levels. At its core, VLP proposes unambiguous NL-based documentation as a readable intermediate layer between prompts and code. The documentation demonstrates concrete program semantics and enables users to provide feedback on potential intent-code mismatches. It supports human-involved, end-to-end repair and validation via three techniques: (i) an NL-style literate language with unambiguous syntax and mostly deterministic code-to-documentation translation, (ii) LLM-based fine-grained mismatch detection that uses trace links between prompts and documentation to focus users' review effort on suspicious documentation lines, and (iii) a verification module that leverages user-validated documentation to derive API-usage checks and formal properties, which are then verified against the generated code using model checking. Our evaluation shows that VLP improves code pass@1 from 28.7%-73.2% to 65.4%-93.5% with reasonable user effort.
Autonomous-research agents have demonstrated end-to-end LLM automation in machine-learning sandboxes where execution provides calibration. Frontier physical science differs categorically: physical reasoning underlies every methodology choice, toolchains are often underdocumented, and calibration must come from external literature anchors - which unscaffolded agents cite but do not confront, hallucinating plausible, unverifiable results from internal priors. We present a pipeline that runs end-to-end from a corpus of 11,083 recent condensed-matter physics arXiv papers to a publication-grade manuscript with three substantive physics findings (here on altermagnetic piezomagnetism): the agent autonomously conceives a research direction by mapping the corpus, calibrates methodology by reproducing published references, conducts novel first-principles computations, and writes the manuscript - grounded in literature throughout, across 47 fresh-context sessions in six phases sharing only on-disk state, with 2,162 literature-consultation events. Fault tolerance emerges from redundancy: fresh-context isolation, distributed grounding, and adversarial review catch what any single session misses; pre- and post-pilot stages are fully autonomous, and pilot requires bounded human intervention only at reproduction failures - operational knowledge curation, not scientific direction. Two paired failure modes - a pre-architecture baseline and a no-pilot ablation - isolate structurally enforced numerical confrontation at calibration checkpoints as the operative grounding mechanism. The primitives, characterized failure modes, and quantified intervention pattern lay a foundation for autonomous research in high-stakes scientific domains beyond computational physics.
Several SLOG test categories explicitly involve directional distinctions (modifier position shifts, argument extraction positions), yet AM-Parser, the previous SOTA, uses an AM algebra whose operations do not encode direction. We redesign the symbolic backend around CCG directed types (deterministic CKY + single linear decoder, 30K learnable parameters). Under the same BERT-base encoder, the system achieves 75.9$\pm$6.4% LF exact match, surpassing AM-Parser (70.8$\pm$4.3%). Per SLOG's own category groupings, gains are highly directional: the CCG system outperforms AM-Parser on all 5 position-shift categories (+29.9pp), while AM-Parser outperforms on all 6 recursive-depth categories. Replacing the encoder with DeBERTa-v3-large yields 90.7$\pm$4.9%, with the largest encoder gains in recursive-depth categories, complementary to directionality's gains. Directional representations shift the bottleneck from the symbolic layer (AM-Parser's 0% category ceiling) to the neural layer, which improves with encoder upgrades.
Linear-attention and state-space language models compress the prefix into a fixed-size recurrent state, yielding O(1) memory at the cost of a lossy exact memory: when many key--value associations compete, earlier facts are overwritten and needle recall degrades. Inspired by Complementary Learning Systems, we give linear attention a hippocampal complement. HOLA (Hippocampal Linear Attention) keeps the usual delta-rule state as a compressive memory and adds a bounded exact KV cache, forming a semiparametric test-time memory: the state models linearly compressible structure, while the cache stores associations that should not be forced through that state. The cache writes without a learned eviction module, keeping tokens with large beta * ||e||, the prediction residual actually committed to the state; a decoupled RMSNorm-gamma cache read then turns these exact KV pairs into sharp retrieval rather than soft averaging. At 340M parameters trained on 15B SlimPajama tokens, HOLA lowers Wikitext perplexity from 27.32 to 22.92 (-16.1%), below a full-attention Transformer++ (26.88), and improves LAMBADA perplexity from 30.95 to 30.26. It also achieves the best linear in-context retrieval and remains much more robust than GDN or a matched HOLA+recency cache on RULER needle-in-a-haystack recall out to 32k tokens (16x its training length).
In the automotive domain, in-car scene understanding (ISU) enables the detection of safety-critical events, such as driver distraction, and supports drivers or passengers by analyzing the in-car scene and adapting the environment (e.g., ambient lighting). The industry is increasingly exploring vision-language models (VLMs) to interpret camera-recorded in-car scenes and extract information for downstream reasoning tasks. However, VLMs may generate incomplete, erroneous, or misleading scene descriptions, highlighting the need for systematic testing. Collecting real in-vehicle data is costly, difficult to scale, and often infeasible, particularly in early design stages. In this paper, we present ISU-Test, an automated testing approach that combines rendering-based scene generation with search-based testing to evaluate ISU systems. By framing testing as an optimization problem and systematically modifying scene parameters, our method generates diverse in-car scenarios and explores a wide range of configurations. We evaluate ISU-Test on both an industrial prototype and open-source VLMs across two case studies: question answering and captioning, comparing against randomized scenario generation. Results show that ISU-Test significantly outperforms the baseline, achieving up to 10 times higher failure rates and up to 3.6 times higher failure coverage.
LLM coding agents are increasingly deployed to act autonomously on real production infrastructure. They execute shell commands, modify repositories, and call operational APIs. However, completing a task is not sufficient for safety. A wrong action can cause severe consequences. Existing agent benchmarks largely emphasize task completion, leaving open how agents behave under benign but underspecified instructions. We present UnderSpecBench, a benchmark for measuring action-boundary violations in coding agents (i.e., Claude Code, Codex, and OpenCode) on DevOps tasks. UnderSpecBench includes 69 task families grounded in documented incidents, CVEs, or tool behavior and organized across four DevOps capability domains and nine operational control surfaces. To isolate underspecification from task difficulty, each task keeps the same environment and ground-truth safe action while varying the instruction along three axes: intent clarity, target certainty, and blast radius. The resulting 2,208 prompt variants are evaluated with deterministic, side-effect-based oracles that separate Safe Success, Wrong Target, and OverScope outcomes; non-action runs are further classified as clarification, refusal, or deferment. Across five agent x model configurations using OpenCode, Claude Code, and Codex, the evaluation results show that underspecification does not mainly make agents fail; it makes them guess. 55.8-67.8% of runs violate at least one boundary. Target underspecification sharply degrades action quality, while blast-radius cues barely reduce action propensity. These findings show that completion-centric evaluation can overstate safe autonomy and motivate mitigations at the model, harness, and system layer.
Neural quantum states (NQS) provide a flexible and scalable framework for approximating quantum many-body wavefunctions. Among NQS parameterizations, autoregressive models are especially attractive because they enable exact, independent sampling from the Born distribution, avoiding the autocorrelation and mixing issues of Markov chain methods. Yet their optimization remains comparatively underexplored: Adam is a scalable method but ignores function space geometry, while stochastic reconfiguration is principled but costly and numerically fragile in large models. To address this gap, we show that variational energy minimization can be viewed as an advantage policy-gradient problem over the Born distribution, motivating trust-region optimization for NQS training. We introduce Proximal Wavefunction Optimization (PWO), a principled trust-region algorithm that clips probability-ratio changes in the amplitude channel and phase increments in the phase channel. PWO avoids explicit matrix inversion, reuses samples across multiple updates, and combines the scalability of first-order optimization with theoretical guarantees. Across Ising and frustrated $J_1$-$J_2$ one- and two-dimensional spin systems, PWO improves stability and wall-clock convergence over Adam, minSR, and SPRING. Finally, we fine-tune a $1.5$B-parameter RWKV-7 model, demonstrating NQS optimization at a scale over three orders of magnitude beyond prior work.
Conventional reinforcement learning strategies for visual generation typically employ sample-wise reward functions, yet this practice frequently results in reward hacking that degrades image diversity and introduces visual anomalies. To address these limitations, we present a novel framework that finetunes generative models using distribution-wise rewards, ensuring better alignment with real-world data distributions. Unlike rewards that evaluate samples individually, distribution-wise reward accounts for the data distribution of the samples, mitigating the mode collapse problem that occurs when all samples optimize towards the same direction independently. To overcome the prohibitive computational cost of estimating these rewards, we introduce a subset-replace strategy that efficiently provides reward signals by updating only a small subset of a generated reference set. Additionally, we apply RL to optimize post-hoc model merging coefficients, potentially mitigating the train-inference inconsistency caused by introducing stochastic differential equation (SDE) in regular RL practices. Extensive experiments show our approach significantly improves FID-50K across various base models, from 8.30 to 5.77 for SiT and from 3.74 to 3.52 for EDM2. Qualitative evaluation also confirms that our method enhances perceptual quality while preserving sample diversity.
While pessimism counteracts overestimation bias in offline reinforcement learning (RL), being overly conservative has been associated with hindering certain forms of generalization. However, in this paper we demonstrate that being overly pessimistic does not inherently prevent optimal generalization in contextual MDPs (CMDPs). Instead, we argue successful generalization depends not on the amount of pessimism, but whether the pessimistic structure respects the underlying symmetries of the optimal solution. We prove that a mildly pessimistic, non-symmetric value function can generalize worse than an overly pessimistic, symmetric one. In offline RL, the structure of the pessimism is determined by the structure of the dataset coverage. As such, enforcing a symmetric value function can be non-trivial, and might require techniques such as data augmentation (DA). Inspired by our theoretical results, we argue that DA can best be applied through a consistency loss during policy extraction, rather than the common practice of (regular) offline training on an augmented dataset. This is empirically validated using IQL and CQL on a rotationally symmetric reacher environment.
In-context learning (ICL) operates via implicit gradient descent embedded in the forward pass of modern AI architectures -- Transformers, Mamba, state-space models, and MLPs. Capturing this capability in biologically plausible Spiking Neural Networks (SNNs) has remained an open challenge: existing SNNs fail the Garg-2022 benchmark at non-trivial task dimensions. We trace this failure to a structural assumption: prior SNN designs route adaptation through inference-time synaptic plasticity, viewing the dendritic compartment as a passive conduit for error or teacher signals. We challenge this assumption. The subthreshold dynamics of a single dendritic compartment already implement a complete online learning algorithm. By treating the compartment as the computational substrate rather than a passive conduit, we propose DendriCL -- a single-layer compartmental spiking architecture whose apical recurrence is structurally identical to leaky online Widrow-Hoff LMS. This dynamics-only update collapses the architectural depth required for general-purpose ICL to a single layer. DendriCL is uniquely seed-stable at super-dimensional Garg-2022 ICL -- where dense Transformers exhibit grokking-style instability and fail past moderate task dimension -- and a linear probe recovers the reference online-LMS trajectory directly from the apical membrane at R^2 = 0.93, showing the algorithm is structurally embedded in the dynamics rather than implicitly discovered during training. Taken together, ICL requires neither attention, depth, nor inference-time plasticity: a single compartment with online-LMS dynamics is sufficient.
Vision-Language Models (VLMs) have demonstrated immense promise in Spatio-Temporal Video Grounding (STVG). However, current evaluation protocols are largely confined to zero-shot assessments on general, daily-life benchmarks. This creates a critical disconnect from real-world applications in specialized fields, where models inevitably encounter rare visual concepts and complex spatio-temporal dynamics. Since exhaustive pre-training across infinite data distributions is infeasible, the ability to adapt to novel domains is essential. To bridge this gap, we introduce AnyGroundBench, a domain-adaptation benchmark designed to shift the STVG evaluation paradigm from static zero-shot testing to rigorous domain adaptation. Targeting five specialized domains (animal, industry, sports, surgery, and public security), AnyGroundBench pairs newly captured videos such as expert-annotated mouse behaviors with established datasets, unifying them through dense, high-fidelity spatio-temporal annotations. Crucially, the benchmark provides dedicated training subsets to systematically measure domain adaptability. We extensively evaluate 15 state-of-the-art VLMs, assessing their zero-shot generalization and In-Context Learning (ICL) capabilities under practical computational constraints. Ultimately, our findings reveal that current models fail in both zero-shot and ICL-based adaptation when confronted with specialized domains, exposing critical flaws in spatio-temporal reasoning that future research must address.
Most data-mixing methods assume the corpus has already been partitioned into groups, and the choice of those groups determines what a mixer can express. Existing labels, including provenance, topic or format taxonomies, and flat embedding clusters, commit to one semantic axis at one granularity; changing the resolution rebuilds the labels. We argue the bottleneck is the label system, not the mixer, and provide a hierarchical one. HERMES is a data-derived labeling substrate: a Learned Semantic Transform followed by 3-stage residual vector quantization annotates each document once into a coarse-to-fine code whose prefix length controls granularity up to approximately 130k cells. At coarse granularity HERMES sits at a plateau with KMeans-family methods on standard clustering metrics, so the contribution is the substrate, not the clusterer. On 1B-parameter, 25B-token pre-training, the hierarchy exposes an interaction fixed-granularity pipelines cannot test: at one prefix length, a combined Stage-2 rule contrast, equal-subbucket coverage versus size-proportional within-bucket quality top-30%, lifts a 16-task capability macro-average by +0.0253; at the next finer level, the same rule loses its measurable edge as candidate pools contract approximately 5x. HERMES reframes data mixture design from choosing among fixed label sets to navigating a reusable, data-derived granularity hierarchy.
Reasoning Language Models (RLMs) have significantly improved performance on complex tasks by extending the reasoning chain. However, these chains are prone to containing factual errors, particularly in knowledge-intensive tasks. To address this issue, we propose CheckRLM, a framework that improves the reliability of the reasoning process through Retrieval-Augmented Generation (RAG) by timely checking and correcting factual errors. Specifically, CheckRLM extracts factual claims from the reasoning chain to identify and localize subtle knowledge inconsistencies during inference. Upon detection of errors, a refinement mechanism performs minimal-cost yet precise corrections by leveraging external knowledge, ensuring coherence between the reasoning chain and correct knowledge. Extensive experiments demonstrate that CheckRLM substantially outperforms existing baselines, exhibiting a strong capability to mitigate error accumulation in long-horizon reasoning with lower costs. The code and data are available at https://github.com/AI9Stars/CheckRLM.
In this paper, we introduce BamiBERT, a new BERT-based pre-trained language model for Vietnamese that addresses key limitations of PhoBERT -- the current de facto Vietnamese text encoder. Trained from scratch on a 129GB corpus of general-domain Vietnamese text for 20 epochs, BamiBERT supports an extended context length of up to 2048 tokens and operates directly on raw input, eliminating the need for external word segmentation. Across 8 Vietnamese benchmarks, it achieves the best score on 11 of 15 metrics and the second-best on 3 others, setting a new state of the art among "base"-sized Vietnamese encoders and demonstrating strong cross-domain generalization. We release BamiBERT at: https://huggingface.co/Qualcomm-AI-Research/BamiBERT
Memory for a long-horizon LLM agent is a contract about what each future decision is allowed to see. The simplest contract appends past observations, tool calls, and reflections to every prompt, which makes prior context easy to access but also turns it into a jumbled mixture in which the effect of any single memory component is hard to isolate. We introduce and instrument an alternative bounded contract: every decision is made from a fresh user message assembled by typed retrieval, with no raw cross-decision transcript appended. The prompt thus stays bounded across runs of any length, and any single layer can be ablated in isolation. We instantiate the contract in Slay the Spire 2, a closed-rule stochastic deck-building game whose runs require hundreds of tactical and strategic decisions. A public online benchmark of frontier LLMs on the same game reports zero wins at the lowest difficulty across five configurations, and the developer-reported human win rate at the same difficulty is 16%; the task is hard but not saturated. Within our harness, a fixed-A0 ablation shows the largest observed difference when triggered strategic skills are enabled: the no-store baseline wins 3/10 games and adding the skill layer 6/10. At this sample size the comparison is directional rather than statistically decisive (Fisher exact p\approx0.37); a cross-backbone probe and public accumulating-context baselines are reported as operational comparisons rather than controlled tests of the contract variable itself. We release a reproducible testbed: 298 completed trajectories with condition tags, frozen memory/skill snapshots, prompt records, and analysis scripts -- an agent design and a validated, reusable methodology for studying how explicit memory layers shape long-horizon LLM-agent decisions.
The aggregation with exponential weights (AEW) estimator is not fully understood in the basic setting of model selection aggregation with squared loss. In particular, whether it is minimax-rate optimal in expectation for large enough fixed temperatures and under random design has been an open problem since its introduction, which was explicitly posed by Lecué and Mendelson (2013). In this paper, we settle this problem by showing that \emph{without} requiring a Bernstein-type assumption, the AEW indeed achieves the excess risk $T \log (M) / (n+1)$ in expectation, whenever the temperature $T$ satisfies $(L^2/T)\exp(B/T)\leq μ/2$. Here, the number of dictionary elements is $M$, the estimator has observed $n$ i.i.d. samples from any distribution, and the loss is assumed to be bounded by $B$, $L$-Lipschitz continuous and $μ$-strongly convex. For squared loss, we show that $T\geq 4 b^2$ suffices when the predictions and labels are $[0,b]$-valued. Because AEW is known to be suboptimal in expectation for temperatures below some constant, this shows that AEW has a sharp phase transition when the temperature is large enough but constant, as conjectured by Lecué and Mendelson.
Mental health disorders affect nearly one billion people globally, yet 75% of individuals in low- and middle-income countries receive no treatment due to workforce shortages, cost barriers, and stigma. Current AI-powered wellness solutions predominantly rely on single-mode conversational interfaces that suffer high abandonment rates and fail to provide measurable, immediate relief calibrated to users' dynamic emotional states. This paper presents Copewell, a novel multi-agent swarm system designed to expand access to mental wellness support through human-centered AI principles. Our architecture introduces three technical innovations: (1) a multi-source assessment framework integrating self-reported, physiological, and contextual data to mitigate algorithmic bias; (2) valence-arousal emotion mapping using Russell's Circumplex Model of Affect to route users to specialized AI agents; and (3) dual-mode intervention delivery combining conversational support with evidence-based sensory wellness protocols. We examine the sociotechnical design considerations underlying Copewell's development, including a privacy-first architecture, embedded ethical oversight through a dedicated Ethics Supervisor agent, and participatory design informed by mental health practitioners. Early practitioner engagement and beta deployment inform design decisions and identify directions for future empirical evaluation. This work contributes to responsible AI discourse by demonstrating how technical architecture can operationalize equity and safety principles from inception.
LLM-as-a-Judge has become the dominant evaluation paradigm for many natural language generation tasks, due to shortcomings of conventional metrics and high correlations with human judgment, albeit mostly in English. There are now attempts to extend LLM-as-a-Judge to multilingual settings including low-resource languages. However, LLMs have limited proficiency in low-resource languages, and there is often no adequate human validation in these settings. To highlight the scope of the problem and current practices, we explore the use of LLM-as-a-Judge evaluators in ACL Anthology papers focusing on multilingual settings and low-resource languages across a diverse set of tasks. Out of 650 papers mentioning LLM-as-a-judge, only 33 of them focus on low-resource or multilingual settings. Our in-depth analysis of these papers indicates inconsistent evaluation outcomes, a tendency to overtrust LLM judgments in multilingual settings, and the widespread reliance on a single judge model per study. To help the NLP community further, we conclude with recommendations about how to use LLM-as-a-Judge in multilingual and low-resource settings.
On-policy self-distillation (OPSD) has emerged as a promising paradigm for improving LLM reasoning, where a privileged teacher with access to reference solutions provides token-level supervision on the student's own generated trajectories. However, we find that OPSD consistently fails on long chain-of-thought (long-CoT) reasoning models, yielding at best marginal gains while destabilizing the reflective reasoning capability these models depend on. Through a novel decomposition of the teacher's supervision signal, we identify the root cause: the teacher's supervision is dominated by a reference-induced component that drives rote memorization of reference-specific shortcuts, while the question-conditioned, inference-transferable component is ignored or actively opposed. Based on this diagnosis, we propose a two-step solution. First, we construct a reference-only teacher (the same model conditioned on the reference without the question) to isolate the non-transferable component of the supervision signal; the residual after subtracting this component captures the question-conditioned, inference-transferable correction. Second, we use pointwise mutual information (PMI) as the mechanism to transform this residual into a well-formed PMI target distribution that the student can directly distill from, filtering out the reference-induced shortcut. Experiments on four long-CoT models across two datasets demonstrate consistent improvements over both the base model and standard OPSD, while preserving the models' natural epistemic behavior throughout training.
The complexity of waste disposal regulations across European countries poses significant challenges for the residents and hinders the transition to a Circular Economy. In Germany, the proper sorting and disposal of household waste remains challenging across municipalities. Consequently, substantially reducing incorrectly disposed waste is vital for improving waste management and advancing the Circular Economy. AI-based waste sorting solutions can support residents through user-friendly tools, such as mobile applications, that guide proper waste disposal. To be effective in supporting the Circular Economy, however, these solutions must be configurable to reflect the specific waste sorting scheme of individual municipalities in Germany. In the scope of this work, an evaluation and analysis are performed of two prominent classification strategies: OvA and OvR. The research uses a dataset constructed in alignment with the waste categories and sorting scheme of the city of Goslar in Germany. Moreover, this work aims to extend beyond the overall performance by examining the behavior of OvA and OvR classification strategies in identifying samples likely to be misclassified. These classification strategies are compared by applying varying confidence thresholds to identify uncertain samples for subsequent human review. This evaluation aims to balance the number of misclassifications against the human effort required for data annotation.
Vision-Language Navigation has increasingly emphasized high-level instruction reasoning, memory, global map construction, and instruction decomposition, while the low-level action representation remains comparatively underexplored. We propose CoFL-S, a low-level vision-language-action framework that predicts a language-conditioned flow field over the robot's local visible sector and generates continuous trajectories by rolling out the predicted field. To train this low-level representation, we convert each VLN-CE episode, originally a whole-episode instruction paired with an action sequence, into frame-level local supervision with aligned sub-instructions and matched action, trajectory, and dense flow-field targets. For evaluation, we introduce a continuous-time Habitat benchmark that isolates low-level action interfaces from instruction decomposition and executes all methods through a shared velocity-command controller, enabling decomposition-independent closed-loop comparison across different planner frequencies rather than fixed discrete forward-and-turn transitions in VLN-CE. Under matched encoders and training settings, CoFL-S consistently outperforms action-token and action-chunk baselines across planner frequencies in the continuous-time Habitat benchmark, and zero-shot real-world closed-loop deployment further shows its advantage over both baselines beyond simulation.
Instruction tuning for speech language models (SLMs) is substantially more challenging than for text-based large language models (LLMs), as it requires learning a new modality and a wide range of speech-specific instructions in addition to those supported by text LLMs. Existing SLM training approaches largely replicate the text LLM training paradigm by synthesizing large-scale speech pre-training and instruction-tuning datasets. However, this strategy is difficult to scale, since speech sequences are significantly longer than text sequences. In this paper, we propose SpeechCombine, an instruction-following speech language model trained without any instruction tuning, using only a single round of speech pre-training on 30k hours of data. Starting from a text LLM base model, we perform continuous pre-training on speech utterances to obtain a speech-adapted model, and then directly combine its weights with the weight difference between the instruction-tuned and base versions of the text LLM. Our results show that this simple combination strategy not only preserves the knowledge and capabilities of the original text LLM, but also effectively transfers them to the speech domain. These findings suggest a new direction for SLM training that avoids reliance on massive speech data.
Aqueous solubility is a key property in early-stage drug discovery, but most predictive models merge physicochemical descriptors and molecular graph information into a single representation, obscuring whether a prediction is driven by global chemistry, molecular structure, or both. We present an additive deep-learning framework that keeps these two sources of information separate throughout training: physicochemical descriptors are encoded by a multilayer perceptron (the chemical branch) and molecular graph topology by a graph neural network (the structural branch), with the two outputs combined only at the prediction stage through an additive model with an optional multiplicative interaction. This design provides a direct decomposition of chemical and structural components that can be examined separately after training. Furthermore, pretraining on the larger AqSolDB dataset and fine-tuning on the smaller BigSolDB2 dataset substantially improve accuracy and reduce run-to-run variations, indicating generalizability of the learned features from the data-rich settings. We further interpret the fitted model using best linear projections of the branch outputs, molecule-level embedding summaries across solubility classes, and atom-level GNNExplainer masks aggregated over functional groups. These analyses show that the chemical branch aligns with familiar physicochemical descriptors, while the structural branch captures graph-topological and functional-group patterns associated with solubility. Across both datasets, the framework attains competitive predictive performance while making the distinct roles of chemical and structural information more transparent.
The evolution toward fully autonomous telecommunications networks (Autonomous Network Levels 4-5) requires AI/ML agents to make real-time network decisions without human intervention. However, no standardized runtime mechanism exists to intercept and validate individual inference outputs before they trigger live network state changes, creating risks of erroneous autonomous decisions. This paper proposes the Guard Rail Validation (GRV) framework, a standardizable runtime architecture for intercepting and validating AI-driven decisions before execution. The framework evaluates decisions across multiple weighted dimensions -- including action scope, action type, service criticality, agent autonomy level, reversibility, and temporal behavioural patterns -- to determine a criticality level. Based on this level, graduated validation mechanisms are applied: execute-with-logging, bounds checking, independent agent validation, or multi-agent consensus. The framework additionally provides cross-agent conflict detection with criticality-weighted priority resolution and runtime conformance logging for regulatory compliance (e.g., EU AI Act Article 14). We present the architecture, algorithmic procedures, O-RAN deployment model, and evaluate threat coverage against known AI/ML attacks in telecommunications.
Predictions are increasingly used to guide high-stakes decisions, from treatment selection to policy making. To ensure reliability with imperfect predictions, uncertainty quantification methods such as conformal prediction build prediction sets with coverage guarantees. However, statistical validity alone does not immediately determine the decisions to take, nor the optimality thereof. This gap is especially delicate in counterfactual settings where the outcome that materializes depends on the action taken, so uncertainty cannot be specified independently of the decision rule. We develop a decision-theoretic framework for uncertainty-informed counterfactual decisions. We identify a novel notion of \emph{policy-coupled coverage} -- namely, coverage of the realized outcome under the action induced by the prediction sets themselves -- as the optimal and lossless interface between uncertainty and action. It plays three roles. First, it justifies acting via a natural max-min rule as minimax-optimal under distributional ambiguity. Second, optimizing prediction sets under policy-coupled coverage is equivalent both to a stronger universal-coverage formulation and to the direct risk-averse optimization over policies and utility certificates; this equivalence yields the explicit form of the population-optimal prediction sets. Third, it admits a two-stage procedure, Policy-Coupled Risk-Averse Conformal Prediction (PC-RACP), that approximates these optimal sets with rigorous finite-sample coverage. Simulations and a real email-marketing experiment confirm that PC-RACP delivers higher utility than existing approaches while maintaining valid coverage, and that ignoring the counterfactual structure of the decision problem is suboptimal for both validity and utility.
Operator learning has emerged as a powerful tool for modeling complex physical systems in functional spaces. However, their neural network-based architectures make them opaque models, obscuring the reasoning behind their predictions. In this work, we introduce a self-explainable operator learning framework that overcomes this challenge by reformulating operator learning as a linear combination of generalized functional linear models expressed through integral equations. Exploiting the additive decomposability of these integral equations, we divide the input domain into subdomains and compute localized integrals to evaluate the contribution of each region to the final prediction. This decomposition enables direct interpretability where the model explains both inputs and outputs by linking specific input regions to corresponding output patterns, thereby revealing which spatial features drive predictions. We demonstrate the framework on function-to-scalar and function-to-function mappings in fluid flow problems involving blood flow and unsteady aerodynamics. The results show that the operator most often prioritizes regions with strong feature gradients, providing physically meaningful insight into the model's decision-making process. Comparisons with established post-hoc explainability methods demonstrate qualitative agreement while highlighting the key advantage of the proposed approach: explainability is embedded directly within the operator structure itself and does not require an external tool. Therefore, our framework provides a mathematically transparent and physically interpretable approach to uncover relationships within data, fostering trust in machine learning for scientific applications by enabling more informed data-driven analysis of physical systems.
The rapid deployment of AI systems across high-stakes domains has created urgent demand for standardized evaluation, yet the field remains fragmented across competing risk taxonomies that catalog risks without showing how an audit is executed. At least 74 AI risk taxonomies exist, and almost all stop at the catalog. The hard part of auditing is not naming a risk but operationalizing it: turning it into a test run against a real system, a measured value, a calibrated severity, and a defensible grade. This paper leads with that bridge. We present the operationalization layer Eticas has built and run, shown end to end on a single risk (PII leakage) against a public benchmark, and then the open taxonomy that makes the method scale. On GPT-4-0314, a disclosure risk that seven external frameworks require be controlled is measured at 0%, 51%, and 84% disclosure as adversarial conditioning increases, mapping through calibrated severity bands to a subcategory grade of E with a SYSTEMIC pattern. Around this example, the Eticas AI Risk Taxonomy v2.0.0 organizes 76 active subcategories across 10 categories and 20 sub-groups, with mappings to 18 external frameworks across compliance, reference, and academic tiers. Its category and sub-group layer is published under CC BY 4.0 as open semantic infrastructure with stable URIs and SKOS/JSON-LD distributions, and a worked subcategory example shows the operational layer down to its severity thresholds. The contribution is the demonstrated bridge from concept to graded finding, anchored by a clean separation of risks from the mechanisms by which they surface, and framed by an open-core model in which the conceptual scaffold is open and the methodology calibration is the practitioner layer. This is the infrastructure the AI auditing field needs: shared, open, and demonstrably operable.
Mutual information (MI)-inspired feature learning techniques are capable of generating low-dimensional embeddings that retain nonlinear dependence structures, but direct estimations of MI suffer from noisy probability distribution estimates in the low-data regime. The H-Score objective, computed from second-order statistics, provides a practical proxy metric for training feature extraction networks. We prove that H-Score is invariant to invertible transformations in the unrestricted functional setting, but becomes sensitive to input basis rotations under constrained approximation classes. Consequently, we study unitary preconditioning for H-Score networks and show that selecting an appropriate basis rotation reduces finite-width truncation error by concentrating predictive dependence into fewer dominant modes. We identify the fast Fourier transform (FFT) as an effective data-independent, low-cost preconditioner for approximately stationary processes, where spectral structure induces concentration of the cross-covariance singular value spectrum. We introduce training-free metrics based on spectral entropy and cumulative dependence energy to quantify basis suitability and predict downstream inference gains prior to network training. Experiments across eight multivariate datasets demonstrate that FFT preconditioning is particularly useful in resource-constrained regimes, achieving up to 50% normalized mean squared error (NMSE) reduction, while the proposed metrics correlate with observed performance gains and correctly identify cases where spectral preconditioning is detrimental.
Human-AI teaming has received increasing attention in the literature. However, the range of studies conducted in multiple domains make it difficult to understand what types of teams are being studied, and in what ways are they similar/different from one another. In this study, we analyse 53 papers on human-AI teams and categorise them into five main clusters based on psychological taxonomies of teaming; AI Assistant, Ad-hoc Dependency, Ad-hoc Forced Dependency, Paired Equanimity, and Group Equanimity. Each cluster represents a unique combination of holistic team-level characteristics, indicating there are multiple disparate team types studied under the same definition. In turn, this raises the question of whether insights are truly transferable between papers. We conclude with guidance on how to identify the types of human-AI teams studied, a checklist for reporting a human-AI team in research work, and ways in which the field can be further synthesised.
The society and emerging risk-based regulatory frameworks for AI underscore the need for rigorous risk assessment to ensure safe and reliable AI systems. In response to this imperative, this paper presents an overview of AI risk assessment (identification and analysis) and management methodologies. It begins by reviewing the worldwide regulatory landscape that drives the need for systematic AI risk assessment. Then we characterize the spectrum of AI-related risks identified in the literature, from technical failures to ethical and social impacts. Subsequently, it reviews key risk assessment methodologies proposed for AI systems, focusing on general frameworks. The paper highlights best practices and illuminates methodological gaps, highlighting areas for further research on AI risk assessment.
We study online resource allocation when both rewards and consumption sizes may be continuously distributed. Requests arrive sequentially and must be accepted or rejected irrevocably under fixed resource capacities. Each request belongs to one of finitely many observable types; conditional on an observable request type, both the reward and the scalar size are random, and the realized size scales a fixed type-specific resource-consumption vector. The model allows the deterministic fluid relaxation to be degenerate. We show that additive regret is governed by the size-weighted mass of requests whose value-to-size ratios lie near the active acceptance cutoffs. We formalize this quantity through an active weighted-mass exponent p. When p > 1, this cutoff mass is thin, and the problem is genuinely hard: every online policy must incur regret of order at least $T^{1/2 - 1/(2p)}$, and this holds for every p > 1. A sample-path marginal policy matches this lower bound up to polylogarithmic factors; and when p = 1, so that the mass grows linearly near the cutoff, it attains $O((\log T)^2)$ regret. For example, if the size and the value-to-size ratio are independent and uniformly distributed, then p = 1; if instead the size and the reward are independent and uniformly distributed, then p = 2. Thus the policy achieves $o(\sqrt{T})$ regret throughout this regularity class without any fluid non-degeneracy assumption, allowing both primal degeneracy and dual non-uniqueness.
Physics-informed neural networks (PINNs) have emerged as a promising route to solve partial differential equations, yet they have struggled to reach the precision of classical solvers. The obstacle is increasingly understood to be one of optimisation, owing to the severely ill-conditioned loss landscape. We present $\textbf{DSGNAR}$: Doubly-Sketched Gauss-Newton with Adaptive Ratio, a scalable second-order optimisation framework that confronts this ill-conditioning and, in doing so, obtains unprecedented accuracy and speed. $\textbf{DSGNAR}$ couples a doubly-sketched Gauss-Newton model with a novel strategy that carefully controls both regularisation and step length. Across a suite of problems spanning nonlinear, chaotic, multi-scale, high-dimensional, and Navier-Stokes, the framework greatly improves on the state of the art: able to attain relative $\ell_2$ errors as low as $3\times10^{-16}$ in double precision, improve contemporary results by five orders of magnitude on the canonical Burgers' equation, and as much as eight orders on a high-dimensional Poisson problem, while remaining markedly faster. We further show that, in single precision, solutions at the limit of round-off error can be obtained very quickly: Burgers' equation to $\ell_2^{\text{rel}} = 4.75 \times 10^{-7}$ in under ten seconds. The framework is also robust to the choice of architecture, arithmetic precision, and initial hyperparameters. The code is available at https://www.github.com/wephy/physics-informed-neural-networks
Distributed machine learning enables collaborative model training without centralizing data, but it also exposes learning processes to privacy leakage and malicious manipulation. Existing defenses typically address these threats in isolation and are often tailored to specific learning paradigms or model architectures, limiting their applicability in realistic deployments. In particular, federated learning and decentralized learning exhibit distinct adversarial surfaces that are rarely addressed within a unified framework. In this paper, we present a model-agnostic framework for adversary-resistant distributed learning that jointly addresses privacy preservation and malicious behavior across both federated and decentralized settings. Our approach combines paradigm-specific defense mechanisms with GPBACC, a privacy-enhancing coded computing technique applicable to arbitrary machine learning models. For federated learning, we integrate robust aggregation strategies to mitigate the impact of malicious participants, while for decentralized learning we employ approximate decode-and-compare and group testing techniques to enable lightweight verification and adversary isolation without relying on a trusted aggregator. Crucially, we evaluate the proposed framework through an explicit, attack-driven analysis. We implement representative privacy attacks and malicious behaviors, and empirically demonstrate that the combination of GPBACC with robust aggregation and verification mechanisms significantly reduces privacy leakage and improves resilience against active adversaries. These results suggest that privacy-enhancing coded computing, when combined with appropriate adversary-resistance strategies, provides a practical and deployable foundation for secure distributed machine learning.
Software development is a complex task that demands cooperation among agents with diverse roles. Large language models (LLMs) have enabled autonomous multi-agent software development frameworks that leverage role-based collaboration to automate requirements analysis, coding, testing, and refinement. However, existing approaches typically assume that intermediate agent outputs are equally reliable, leaving them vulnerable to hallucination propagation, where incorrect decisions generated in early development phases are transferred to downstream agents and negatively impact final software quality. To address this challenge, we propose UA-ChatDev, an uncertainty-aware multi-agent software development framework that integrates uncertainty quantification into agent interactions. It introduces a lightweight uncertainty estimation mechanism based on token-level log probabilities to assess the confidence of agent responses and employs phase-aware threshold calibration to selectively trigger retrieval-based verification when uncertainty exceeds acceptable levels. Extensive experiments on the SRDD benchmark demonstrate that UA-ChatDev consistently outperforms existing single-agent and multi-agent software development frameworks across completeness, executability, consistency, and overall quality metrics. Further ablation studies and communication analyses verify that uncertainty-aware interactions enhance code execution reliability.
Deep learning has achieved remarkable performance in medical image segmentation, yet it suffers from critical limitations: mathematical intractability, substantial parameter requirements, and lack of clinical interpretability. We propose RadiomicNet, a novel two-stream hybrid architecture that enhances standard deep learning by integrating handcrafted radiomics features directly into the segmentation learning process. The key contribution is the Radiomics Attention Gate (RAG), which leverages Gray-Level Co-occurrence Matrix (GLCM) and Local Binary Pattern (LBP) features to modulate skip-connection attention in a lightweight MobileNetV2-based encoder-decoder, providing ante-hoc interpretability without post-hoc approximations. A novel Radiomics Consistency Loss further enforces alignment between texture complexity and prediction uncertainty, reducing Expected Calibration Error (ECE) from 0.142 to 0.118. RadiomicNet achieves a Dice Similarity Coefficient (DSC) of 0.763 +/- 0.231 on the Breast Ultrasound Images (BUSI) dataset and 0.854 +/- 0.112 on Kvasir-SEG, outperforming U-KAN by 1.2% and 1.8%, respectively (p < 0.05, Wilcoxon signed-rank test), with only 3.27M parameters, 9.5x fewer than standard U-Net and 4.3x fewer than U-KAN. Gradient-based feature importance analysis reveals that GLCM dissimilarity (15.24%), GLCM energy (14.56%), and LBP entropy (11.49%) are the dominant radiomics cues, providing clinically meaningful explanations for segmentation decisions. The proposed approach demonstrates that compact, interpretable models grounded in domain knowledge can deliver state-of-the-art segmentation performance with substantially reduced computational overhead.
Large language models (LLMs) exhibit remarkable reasoning capabilities, but their task-specific fine-tuning is notoriously plagued by overconfidence, severely hindering trustworthy deployment. We propose Data-Adaptive Lower-Rank Adaptation (DALorRA), a simple and effective variational Bayesian sparse framework that shifts the paradigm of uncertainty quantification from the dense parameter space to the lightweight rank level of low-rank adaptation (LoRA). With the insight that LoRA essentially aggregates multiple rank-one components that may provide superfluous model capacity, DALorRA imposes stochastic masking on rank dimensions, enabling Bayesian regularization of model capacity during training and ensemble-like calibration during inference. Extensive experiments demonstrate DALorRA's excellent calibration of LLMs without compromising reasoning accuracy.
Multiple-choice medical benchmarks are increasingly saturated, and recent rubric-based evaluations such as HealthBench have shown that open-ended clinical performance is far from solved - its "Hard" subset top score remains 32%. We present a small, deliberately difficult evaluation dataset of five clinician-authored clinical scenarios spanning four specialties (anaesthesia, internal/family medicine, emergency medicine, and obstetrics), each accompanied by an atomic, weighted, MECE rubric (25-62 criteria per task; 184 criteria total) authored from a clinician-drafted golden answer. We evaluate three frontier models: GPT 5.4, Claude Opus 4.7, and Gemini 3.1 Pro. Mean rubric pass rates were 0.47 (Claude), 0.39 (GPT), and 0.37 (Gemini). The central finding is an inversion of clinical priority: the highest-weighted (weight-5, critical) criteria passed at only 32.4-41.7%, while low-stakes weight-1 criteria passed at 80-90%. 56 of 108 critical (weight-5) criteria (52%) were satisfied by no model. Three LLM autoraters reproduced expert met/not-met labels on 92.8-94.7% of 552 graded criteria. We position this as a methods-and-preliminary-findings contribution: the five tasks demonstrate a scalable, defensible pipeline ready to develop into a large-scale benchmark.
The rapid advancements in using neural networks as implicit data representations have attracted significant interest in developing machine learning methods that analyze and process the weight spaces of other neural networks. However, efficiently handling these highdimensional weight spaces remains challenging. Existing methods often overlook the sequential nature of layer-by-layer processing in neural network inference. In this work, we propose a novel approach using dynamic graphs to represent neural network parameters, capturing the temporal dynamics of inference. Our Dynamic Neural Graph Encoder (DNG-Encoder) processes these graphs, preserving the sequential nature of neural processing. Additionally, we also leverage DNG-Encoder to develop INR2JLS (Implicit Neural Representation to Joint Latent Space) for facilitate downstream applications, such as classifying Implicit Neural Representations (INRs). Our approach demonstrates significant improvements across multiple tasks, surpassing the state-of-the-art INR classification accuracy by approximately 10% on the CIFAR-100-INR.
This paper studies additive regret in the multi-secretary problem, defined as the gap between the expected offline prophet reward and the reward of the best online policy. Prior work established \(O(\log T)\) regret for bounded-density distributions with connected support and \(O((\log T)^2)\) upper bounds for bounded-density distributions with support gaps. It was unknown whether the extra logarithmic factor is necessary even in the one-resource model. We prove that it is necessary. For a mixture of two separated uniform distributions at the critical capacity, the optimal regret grows at least on the order of \((\log T)^2\). Thus the existing \(O((\log T)^2)\) upper bounds for bounded-density gapped instances, including those implied by network revenue management models with continuous rewards, are tight in this simplest specialization. The same framework also yields a matching lower bound for gapped distributions whose gap-facing densities vanish near the support edges; this companion result is given in the appendix. The proofs use Bellman certificates: feasible solutions to a relaxation of the exact Bellman recursion. This framework converts lower bounds into explicit certificate constructions and identifies why support gaps permit larger regret.
Alzheimers disease (AD) is a brain disorder that develops slowly and mainly affects memory, thinking, language, and daily activities. It is one of the most common causes of dementia and creates many difficulties for patients as well as their families. In the early stage, the symptoms are often mild and may look like normal ageing. For this reason, many people are diagnosed late, when the disease has already progressed. At present, there is no complete cure for AD. Still, early detection can help doctors manage the condition better and take suitable steps at the right time. In this study, a machine learning model is proposed to detect the early stages of Alzheimers disease using clinical details, neuropsychological test scores, and neuroimaging-related measures. The data used in this work is collected from the Alzheimers Disease Neuroimaging Initiative (ADNI). As the dataset has missing values, iterative imputation is applied to fill them. The dataset also has class imbalance, which is handled using Borderline SVM-SMOTE. After that, feature selection is carried out using wrapper-based and embedded methods so that only important features are used for training. The selected features are divided into training and testing sets, and feature scaling is applied. A stacking ensemble model is developed using Logistic Regression, Extra Trees, Bagging KNN, and LightGBM as base classifiers. Along with this, an artificial neural network is also trained on the same dataset. The performance of these models is compared using precision, recall, F1-score, and AUC-ROC. This study aims to find the best classifier and also identify important biomarkers that may help in the early diagnosis of Alzheimers disease.
Most LP-from-text benchmarks are static datasets of word problems written and labeled by hand. Once such a dataset is released, its size is fixed, its difficulty is fixed, and every problem can leak into the training data of future LLMs. We present \textbf{A$^{2}$utoLPBench}, a benchmark for testing LLM-driven agents on linear programming problems written in plain text. We first pick a feasible point and dual, then write down a problem for which that point is optimal and the objective value is known. The answer is known by construction, with no solver call and no human annotator. The evaluation environment bundles a reference solver-critic baseline and a Docker image whose usage instructions are written for an LLM-driven agent to read. With these in place, any agent can run the benchmark and get a calibrated score with one command. Because the benchmark is a generator rather than a fixed dataset, it has properties no fixed dataset can match: an unlimited supply of fresh problems, a difficulty knob set by $(n,m)$, ground-truth answers correct by construction, low LLM-side cost per problem relative to human authoring, repeatable scores across independent batches, and resistance to training-data leakage when fresh post-cutoff seed ranges are used.
Chemical language models (CLMs) are trained with linearized representations such as SMILES, yet it remains unclear which chemically meaningful substructures they encode. To foster a better understanding of CLMs, we conduct a systematic study and probe for 78 molecular substructures across eight pre-trained and six randomly initialized models. We furthermore study how fine-tuning on chemical downstream tasks affects the learned representations of molecular substructures. Our results show that pre-training generally improves molecular structure awareness of CLMs, particularly in the upper layers. Moreover, randomly initialized models already encode ring structures well in the first layer. Our analysis on two chemical downstream tasks further reveals that, interestingly, fine-tuning affects task-relevant molecular substructures more than others, indicating that the changes in the representations follow chemical theory.
We study timestep allocation for score-based diffusion sampling, where a learned reverse-time dynamics is discretized on a finite grid. Uniform and hand-crafted schedules are standard choices, but they rely on fixed prescriptions and can therefore be suboptimal. To address this limitation, we propose Adaptive Reparameterized Time (ART), a continuous-time control formulation that learns a time change by treating the speed of the sampling clock as the control, so that a uniform grid on the learned clock induces adaptive timesteps in the original diffusion time. Based on a leading-order Euler error surrogate, ART provides a principled objective for allocating timesteps along the sampling trajectory. To solve this deterministic control problem, we introduce ART-RL, an auxiliary randomized formulation with Gaussian policies that turns schedule learning into a continuous-time reinforcement learning problem. We prove that the randomized ART-RL formulation is equivalent to ART at the optimizer level, in the sense that its optimal Gaussian policy recovers the optimal ART time-warping rate through its mean. We further establish policy evaluation and policy improvement characterizations and derive trajectory-based moment identities that yield implementable actor--critic updates for learning the schedule. Across experiments ranging from controlled low-dimensional settings to image generation, ART-RL can be plugged into existing diffusion samplers by changing only the timestep grid, consistently improving sample quality over strong baseline schedules at matched budgets while leaving the rest of the sampling pipeline unchanged. The learned schedules also exhibit broad generalization, transferring without retraining across sampling budgets, datasets, solvers, pipelines, and representation spaces.
Scientific machine learning papers typically make computational claims, e.g., that the relative mean square error is less than 5% or that the 95% predictive credible interval covers the test data. A coding agent can be prompted to replicate those claims from paper materials alone, but the prompt does not by itself reliably preserve progress or check whether generated evidence supports the paper's claims. We introduce Paper-replication, a workflow that makes each selected paper claim a target with recorded evidence, and implement it as a coding-agent skill. The workflow makes the agent record those targets, reconstruct the paper's method, run computational experiments, link generated outputs to provenance and comparisons with the paper's claims, record where matched evidence appears in the replication report, and pass validation checks before completion. We evaluate Paper-replication on twelve independent runs across four scientific machine learning papers. All twelve workspaces pass the completion gate, and all 158 recorded targets are matched with report coverage. Even in this completed workspace state, repeated runs differ in how papers are divided into targets, in numerical fidelity to the source papers, in elapsed replication time, in the number of intermediate executions replaced before final evidence is accepted, and in the rules used to accept evidence. Paper-replication makes completion depend on workspace evidence and validation checks rather than on the agent's final message.
Restoring archival film remains a fundamentally challenging problem due to the absence of paired training data and the lack of standardized evaluation benchmarks. Pristine versions of deteriorated footage are physically unrecoverable, requiring supervised methods to rely on synthetic data that often fail to capture the complex, temporally coherent nature of real film degradation. At the same time, existing real-world datasets are limited in scale, quality, and accessibility, hindering reliable evaluation and fair comparison across methods. We address both limitations with AbsoluteDegradation, a physics-inspired, modular pipeline for synthesizing realistic film degradations, and a new large-scale archival benchmark. The proposed pipeline models the analog-to-digital process as a structured composition of artifact families, incorporating signal-dependent grain, parametric scratches, and temporally coherent camera motion, enabling controlled generation of diverse degradation regimes. In parallel, we introduce a curated dataset of 81,576 high-resolution frames sourced from real archival footage, designed for consistent evaluation under real-world conditions. Together, these contributions provide a unified framework for training and benchmarking restoration models. Extensive experiments across multiple architectures show that models trained with AbsoluteDegradation generalize better to real-world footage, while the proposed benchmark reveals systematic failure modes of current methods. We hope this work establishes a foundation for reproducible and domain-authentic evaluation in archival film restoration.
Penile measurement is clinically relevant across male reproductive and urogenital health, including conditions such as micropenis, congenital and endocrine disorders, and sexual or urinary dysfunction. However, quantitative assessment of penile size has relied mainly on external length or circumference measurements, which are difficult to standardize, sensitive to measurement conditions, and unable to capture the internal portion of the penis. MRI enables volumetric assessment of the whole penis in vivo, but automated segmentation has not previously been established at population scale. Automated whole-organ volumetry would enable high-throughput phenotyping for multi-omics and clinical studies of male reproductive disease. Here, we present a deep learning framework for whole-penis segmentation in multi-channel DIXON MRI. Using a newly curated expert-annotated training dataset ($n = 145$ subjects; $13,050$ annotated slices) and a double-annotated independent test benchmark ($n = 24$ subjects; $2,160$ double-annotated slices), we optimized a 3D nnU-Net architecture. The model achieved a 5-fold cross-validation Dice score of $0.90$ and performed at observer-level accuracy on the independent test set (Dice: $0.92$; Hausdorff distance: $3.58$). We deployed the model in $34,412$ UK Biobank participants, enabling automated quantification of total penile tissue, including both external and internal components. Longitudinal evaluation in 2,282 men demonstrated high inter-session reproducibility ($r = 0.87$). This framework establishes a reproducible and population-scalable method for MRI-based assessment of penile anatomy and provides an open technical resource for future studies in urological imaging and male reproductive health. The trained model weights will be publicly released.
Conventional traction control architectures intervene only after the adhesion limit of a tire has already been breached. This paper investigates whether Rolling Split Conformal Prediction , monitoring the volatility of non-conformity residuals from a per-driver Random Forest model of expected slip behavior , can serve as a statistically grounded pre-incident warning signal, ahead of gross traction loss. Unlike an earlier internal draft of this work, the evaluation reported here corrects a confound in the slip proxy (vehicle speed is included as an explicit model feature, not left implicit in the target's denominator), uses every racing lap for each driver rather than only the fastest lap, and is scored against real, timestamped incident labels extracted from FIA Race Control Messages and track-limits lap deletions rather than narrated post-hoc. The result is negative: across 19 drivers and 55,563 test-phase telemetry samples, the rolling-volatility detector achieves a mean precision of essentially 0.0 and mean recall of 0.0 against 14 ground-truth incidents, while flagging on average 15.3% of all samples as anomalous , too high a false-alarm rate for any early-warning use. A static 95th-percentile threshold baseline performs no better in any way that would justify the added complexity of the conformal-volatility formulation. Residual autocorrelation diagnostics show the split-conformal exchangeability assumption is violated for every driver (Ljung-Box p < 0.001, n = 19/19), which is one plausible driver of the high false-alarm rate. We report this as a methodologically rigorous negative finding, diagnose its likely causes, and outline what a genuinely predictive version of this approach would require.
As Large Language Models (LLMs) and agentic systems become integrated into real-world applications, ensuring their safety and security is critical. Guardrail systems that detect and block malicious instructions sent to and from an LLM are an essential component of AI security. However, researchers conducting black-box adversarial emulation against production AI systems often struggle to determine whether a guardrail block or an LLM rejection has occurred. This distinction is important because the techniques used to bypass guardrails can differ substantially from those used to bypass LLM safety alignment, and has a material impact on attack technique selection and optimization. We propose the first black-box guardrail reconnaissance methodology, which detects the presence of a guardrail within a target AI system through behavioral monitoring of HTTP, lexical, and timing signals, assuming only black-box access and zero prior knowledge of the guardrail or AI system. Experiments demonstrate that our approach detects guardrail presence with 100% accuracy, with statistically significant behavioral separation between benign and malicious interactions (q < 0.001). Our approach further identifies the content categories a guardrail is designed to block, and distinguishes guardrail blocks from LLM rejection on unseen prompts with an average F1 score of 98%.
While Large Multimodal Models excel in comprehension, high-throughput inference engines lack native support for multimodal generation. This is severe in Speech Language Models, where generating multi-layered audio tokens via decoupled AR+NAR or synchronous Multi-Token Prediction (MTP) with delay-pattern interleaving conflicts with standard single-stream loops. We present a vLLM-based inference pipeline for unified speech understanding and generation. We extend autoregressive decoding to natively execute delay-pattern de-interleaving and coordinated multi-stream sampling, integrating an on-GPU acoustic decoder for end-to-end waveform synthesis. Crucially, we overcome the shared intuition that Classifier-Free Guidance (CFG) halves throughput. By co-scheduling paired conditional and unconditional requests within a continuous batch, our CFG implementation sustains 80% of non-CFG throughput, absorbing dual-request and logit merging overheads. We open-source our framework.
Scientific Fitness Coaching (SFC) is typically delivered by human professionals, making it costly and inaccessible to many. While recent advances in Large Language Models (LLMs) show considerable promise for more inclusive fitness coaching, directly deploying prevailing general-purpose LLMs in SFC reveals critical limitations. These models often lack sufficient domain-specific knowledge integration, leading to weak performance on complex SFC scenarios. In this paper, we introduce FitOne, a series of fitness LLMs (with 8B and 32B parameters) designed to improve reliability and domain specialization for SFC applications. Built upon the Qwen3 foundation models, FitOne is developed through a three-stage post-training pipeline consisting of continual pre-training, supervised fine-tuning, and reinforcement learning, using large-scale, high-quality datasets derived from rigorous knowledge engineering. We conduct comprehensive evaluations of FitOne on professional fitness certification exams, including ACSM-EP and NSCA-CSCS, as well as general capabilities such as knowledge reasoning and instruction following. Experimental results show that, while retaining strong general capabilities, FitOne-8B/32B achieves average improvements of up to 10.09%/9.29% and 12.73%/7.01% on the ACSM-EP and NSCA-CSCS exams, respectively, compared with the Qwen3 base models. Furthermore, in-depth ablation studies confirm the necessity of each training stage, highlighting the pipeline's effectiveness in balancing domain expertise enhancement with general ability retention. We believe this research advances LLM systems toward more reliable fitness intelligence and will inspire future research on developing domain-specific LLMs.
Autonomous AI agents increasingly depend on external knowledge stores, yet most retrieval pipelines provide relevance without durable guarantees of provenance, version identity, integrity, traceability, or point-in-time reconstruction. We formalize this as context governance and present ContextNext, an open specification and reference implementation for governed AI-consumable knowledge vaults. ContextNext does not replace Retrieval-Augmented Generation (RAG); it supplies the governance layer beneath retrieval, determining which artifacts are approved, current, attributable, and integrity-verified before retrieval systems operate over them. The specification combines typed Markdown documents with metadata, deterministic set-algebraic selectors, contextnest:// URI references, SHA-256 hash-chained version histories, graph-level checkpoints, source nodes for live data through the Model Context Protocol (MCP), and audit traces of agent context consumption. These mechanisms let organizations reconstruct which knowledge versions informed an agent output and whether those versions were AI-eligible when consumed. We report first empirical results from two controlled experiments. In a stale-version attack isolating the governance-versus-retrieval failure mode, governed selection strictly Pareto-dominates BM25 sparse retrieval, with higher answer-quality pass rate (97% versus 93-90%) at about one-third the input-token cost. In a retrieval-determinism experiment over a 1,060-document corpus, deterministic selectors and BM25 return stable document sets across repeated identical queries (Jaccard 1.0), while a dense+HNSW baseline is non-deterministic on 80% of queries (mean Jaccard 0.611, worst case 0.210). These results suggest that context governance addresses failure modes retrieval quality alone is not designed to resolve. We release a core engine, CLI, and MCP server under open licenses.
Large language models (LLMs) are increasingly used as cheap, scalable judges that compare candidate outputs pairwise -- to rank responses, select models, or triage papers. Yet LLM judges are both noisy and systematically biased: they favor verbose or well-formatted answers and exhibit position effects, so simply aggregating their votes recovers a ranking of presentation, not of true quality. We study the practical goal of identifying the \topk{} items under a fixed comparison budget, and make two contributions. First, we cast judging as Bayesian inference over latent quality with explicit, judge-specific bias covariates (verbosity, position), regularized by a shrinkage prior so that the data decide which biases a given judge actually exhibits. Second, we introduce a \topk-aware active acquisition rule that chooses the next comparison to maximally reduce uncertainty about \topk{} \emph{membership}, rather than about the full ranking. On a controlled benchmark with known ground-truth quality, judged by sixteen real LLMs spanning open and proprietary families (Llama, Qwen, Phi-4, GPT-4o-mini/5.1/5.5, Gemini, DeepSeek, and Claude Haiku/Sonnet/Opus), naive aggregation plateaus at a wrong \topk{} on biased judges regardless of budget, while our bias-aware model recovers it; \topk-aware acquisition reaches this ceiling with far fewer comparisons than round-robin or a global-uncertainty (D-optimal) rule. Bias is real but heterogeneous and capability-dependent: cheap and mid-tier judges carry a strong verbosity bias that our model corrects (lifting recall from $\sim$$0.5$--$0.6$ to $0.84$--$1.0$), whereas the frontier judges we tested show little bias and already rank accurately, so bias-aware modeling changes little there.
Classifying heterogeneous omics data remains a fundamental challenge in computational biology, particularly in high-dimensional, small-sample settings where nonlinear interactions dominate and class imbalance further complicates reliable prediction of minority phenotypes. While traditional kernel methods rely on feature abundance, they fail to leverage the known interaction landscapes of biological systems. In this work, we propose a structured Gaussian process classification framework that integrates graph-encoded biological pathways directly into the kernel construction. By propagating information along known interaction networks and combining this with abundance-derived features, the resulting classifier captures both quantitative measurements and topological context. We benchmark our proposed methodology on three publicly available gut and fecal microbiome datasets. To address severe class imbalance, we evaluate complementary strategies, including data-level resampling, threshold calibration, and confusion-matrix-based adjustments, and report minority-class performance alongside accuracy. The hybrid approach yields a performance gain over unstructured baselines and matches the performance of established benchmarks for similar datasets. Furthermore, the probabilistic nature of the framework naturally provides calibrated predictive uncertainty, enabling robust differentiation between confident predictions and ambiguous samples.
Large kernel depthwise convolutions achieve strong performance but suffer from significant degradation as kernel size grows due to irregular memory access from gather-based computation; while Large Kernel Acceleration (LKA) helps on small feature maps, it becomes counterproductive on large feature maps, even slower than non-accelerated implementations. We propose Windowed Batch Matrix Multiplication (WBMM), which partitions input into contiguous windows and indexes a compact relative position bias table to construct weight matrices, enabling regular memory access via batched matrix multiplication. This yields a unique property: WBMM's throughput improves with larger windows, opposite to depthwise convolutions that degrade with larger kernels. Operator-level benchmarks show WBMM with 14x14 windows outperforms 5x5 depthwise convolution baselines in speed while providing a 7.8x larger per-layer receptive field. Combined with inter-block cross-window communication and hierarchical window reparameterization, WBMM achieves comparable or higher accuracy on ImageNet-1K, COCO, and ADE20K with 1.31-1.88x training speedup, and demonstrates consistent advantages across GPU, CPU, and edge devices without requiring specialized acceleration kernels. Our code is available at http://github.com/wansong-s/WBMM
Flow-matching vision-language-action policies generate robot action chunks through an iterative transport process, creating an opportunity for test-time guidance without retraining the base policy. We study this opportunity in Guided Action Flow, an inference-time framework that keeps a pretrained SmolVLA policy frozen and uses a learned action-chunk critic to guide its reverse-time flow sampler. The critic is trained from real success and failure rollouts, can condition on task-description features from the frozen SmolVLA language pathway, and is used only through action gradients during sampling. We evaluate the approach on LIBERO manipulation tasks. A single-task critic improves success from 68.0% to 82.0% on one seed window and from 82.0% to 86.0% on another. A multi-family task-description critic improves validation success from 46.0% to 56.0%, while the locked held-out test gain is positive but modest, from 65.0% to 67.5%. These results support the feasibility of Q-guided inference for frozen flow-matching VLA policies, while showing that critic generalization and uncertainty-aware guidance remain the central bottlenecks.
We propose an improved Fourier Neural Operator (FNO) for modeling two-dimensional Rayleigh-Bénard convection by predicting time increments instead of full solutions, achieving higher accuracy than a standard FNO baseline. The resulting model is compact (314k parameters, 1.26 MB) and fast (7 ms inference), while maintaining similar accuracy as demonstrated in previous benchmarks. We show that although FNOs generalize to finer meshes, accuracy remains limited by the resolution of the training data.
Hierarchical state-space models (HSSMs) offer a promising approach to long-horizon prediction by segmenting sequences into temporal chunks. However, their performance hinges on how chunk boundaries are determined. While prior HSSMs typically rely on fixed-length chunking or similarity-based boundary detection, these methods often misalign with the intrinsic temporal structure of the data. We argue that chunking should instead be driven by prediction errors, which more directly indicate when longer-range context becomes necessary. Nevertheless, integrating surprise-based chunking into HSSMs introduces critical challenges, including hierarchical collapse during end-to-end training and the absence of surprise signals during open-loop prediction. To address these issues, we propose Surprise-based Nested Temporal Abstraction (SUNTA), a method that employs a decoupled training strategy to preserve surprise signals and uses internal inconsistency as a top-down surprise metric to determine chunk boundaries within imagined rollouts. Experiments on video prediction tasks in 2D and 3D environments demonstrate that SUNTA outperforms baselines, uniquely maintaining accurate predictions over 250 timesteps, whereas all baselines degrade within the first 10 timesteps.
Wave Function Collapse (WFC) is a widely used procedural content generation method that learns local adjacency constraints from example inputs to generate larger outputs. In this paper, we explore combining WFC with evolutionary search by evolving the small input examples used by WFC rather than directly evolving complete levels. In this approach, WFC acts as a genotype-to-phenotype mapping. The generated levels are then evaluated through domain-specific fitness functions. We evaluate the method in two domains with different relationships between local and global structure: Maze connectivity maps and Zelda-style dungeon layouts. Our results show that evolutionary optimization over WFC inputs improves generation quality in domains where properties emerge from local relationships, while domains requiring global constraints remain challenging. These findings suggest that evolutionary search can effectively guide WFC generation when target objectives align with local structure.
We present HaloGuard 1.0, an open-weights implementation of the constitutional-classifier paradigm for input safety. It achieves state-of-the-art performance on English and multilingual prompt-safety benchmarks at roughly one-tenth the model size of current leading open guard models. The safety constitution is the organising structure of the corpus: a natural-language constitution of 46 policies and 2,940 subcategories drives synthetic data generation, with exhaustive one-to-one paired counterfactuals that hold topic and vocabulary fixed while flipping intent, a two-tier harmless design that separately targets boundary and baseline false positives (FPs), and balanced multilingual materialisation across 46 languages that treats language as a surface form appearing on both sides of the boundary rather than as an adversarial signal. Across seven prompt-safety benchmarks, HaloGuard 1.0-0.8B attains the best average F1 (90.9) of any open guard we evaluate, outperforming baselines up to 27B parameters (over 30 times larger) while holding false-positive rate (FPR) to 4.3 and false-negative rate (FNR) to 9.5. The HaloGuard 1.0-4B variant reaches average F1 of 92.1 and FPR of 3.5, spending its extra capacity on precision rather than recall. A structured adjudication of the remaining failures indicates that most apparent missed-harm cases are benchmark mislabels rather than genuine model misses. An always-on adversarial red-teaming protocol continuously hardens the guard against both content-level and agentic attacks. We release the models as open weights.
Long-context reasoning requires models to locate, revise, and synthesize evidence distributed across lengthy inputs. Existing long-context RL methods usually reward final answers or static evidence extraction, offering little feedback on how intermediate actions change the model's evidence state. We propose Maven, a reinforcement learning framework with an editable evidence memory. Maven defines an answer-conditioned evidence-state value and rewards action-level state transitions: add actions are credited by marginal gain and hindsight contribution, link actions by evidence synergy, and drop actions by improved answer support after removing misleading evidence. These rewards are assigned to the corresponding action spans in GRPO. Across Llama and Qwen models on LongBench v2, LongReason, and RULER, Maven outperforms outcome-only RL and evidence-identification baselines, producing more sufficient evidence sets and lower distractor retention. Our results show that long-context RL benefits from optimizing stateful evidence navigation rather than one-shot evidence extraction.
Large language models (LLMs) are increasingly deployed in domains requiring guardrails to detect unsafe, off-topic, or adversarial prompts. Existing guardrails predominately rely on fine-tuning to build classifiers, which often suffer from low generalization and high inference latency. We present kNNGuard, a training-free guardrail that utilizes the activation space of an off-the-shelf LLM. Given a small bank of 50 safe and unsafe prompts, kNNGuard extracts hidden activations and performs multi-layer kNN fusing activation-space and embedding-space scores for classification. Across six domains spanning topical and security prompts, kNNGuard achieves competitive or superior F1 compared to fine-tuned state-of-the-art guardrails while running 2.7x faster than the best comparable guardrail, and 10x faster than a fine-tuned safety classifier without gradient updates or fine-tuning. Domain adaptation requires only updating the labeled bank, which can be constructed in under 10 seconds and several orders of magnitude faster than established guardrails. We also analyze the impact of system prompts, layer selection, and integration into production LLM pipelines as a configurable, low-latency guardrail.
Ensuring model reliability in Explainable AI requires a global assessment of the hypothesis space. We propose a formal framework for the exhaustive analysis of optimal and near-optimal decision trees, called Algebraic Decision Tree Counting (ADTC). Inspired by Algebraic Model Counting (AMC) in knowledge representation, ADTC reformulates diverse analytical tasks, such as optimization, counting, and sampling, into a unified sum-of-products computation over a semiring $R$. While the hypothesis space of decision trees is doubly exponential with respect to the maximum depth $Δ$, our dynamic programming algorithm achieves $O^*(n^{O(Δ)})$ time complexity in the number of features $n$, where $O^*$ suppresses polynomial factors. To handle complex constraints consisting of multiple tree metrics, we introduce model behavior tensors that aggregate semiring values via convolution products over a tensor semiring. This algebraic approach efficiently constructs a model profile that captures the global landscape and trade-offs between criteria such as accuracy, size, and fairness. We demonstrate the utility of our software, emtrees, on real-world datasets, illustrating how ADTC facilitates evidence-based model selection in sensitive domains.
Graph Neural Networks (GNNs) have been widely used to capture spatial functional connectivity patterns to improve electroencephalography (EEG)-based depression recognition performance. However, the functional connectivity of brain networks in patients with depression exhibits an inherent hierarchical structure, making it difficult to capture accurate connection patterns. To address these issues, this paper proposes a novel model named Sample-Adaptive Hyperbolic Graph Neural Network (SA-HGNN), which aims to accurately extract the authentic hierarchical structure of depression-affected brain networks. Specifically, the proposed model comprises three core modules. First, a Sample-Adaptive Graph Construction module dynamically constructs personalized brain network topologies to capture more complex spatial relationships within the brain network. Second, hyperbolic graph convolution is employed to overcome the representation bottlenecks of Euclidean space, leveraging hyperbolic geometry to precisely capture latent hierarchical relationships within the brain network. Finally, an Attention Pooling module adaptively filters out highly redundant noise channels in EEG signals, effectively mitigating the interference of inherent noise on the authentic hierarchical topology. Extensive experiments on public EEG datasets demonstrate the superior performance of our method across resting-state and task-related paradigms, validating its robustness to noise and efficacy in capturing abnormal functional connectivity patterns in brain networks of patients with depression.
File-level copying is a widespread but ungoverned form of software reuse. Copying files across repositories reduces supply-chain visibility: it removes the four observable signals a package manager provides for a declared dependency (provenance, maintenance, security, and compliance) with no mechanism to restore them. To characterize the scale and consequences of this unmanaged reuse, we present a mixed-method study of copying across the entire open-source ecosystem using World of Code (WoC). From a 0.1% commit sample, we extract 690,500 copy events and retain 3,912 rationale-bearing copy commits for intent labeling. We show that the 13 axial copy forms, spanning vendored dependencies, hardware/driver synchronization, scaffolding, UI assets, and direct source-code reuse, are unreliable proxies for developer intent: among rationale-bearing commits, hardware/driver copies are predominantly fork-maintenance work (78%), while dependency-vendoring copies more often signal upstream bypass (70%) than offline availability. These visibility gaps are form-specific: security and license risk concentrate in complementary copy forms. Copied sources are frequently stale (median 155 days; 38.5% over one year old) and seldom record a recoverable origin (4.3% documented), let alone a checkable version (2.0% versioned); even vendored copies record where they came from only 10% of the time. Security risk concentrates in vendored dependencies: 17,314 CVE-risk copy commits in the full-WoC graph, 88% in the dependency-vendoring form; 80% score CVSS >= 7.0 and upstream-fix adoption is only 47%-84%. License risk concentrates in direct source-code reuse: 41,777 pre-validation candidates, 66% in the source-code form, with 39 verified high-star violations (kappa = 0.752). Both risks reach packaged software and are invisible to dependency scanners operating on declared metadata alone.
In recent years, it has become increasingly evident that large language models (LLMs) and autonomous agents raise the level of abstraction in software development by shifting the focus from writing precise procedures to expressing intents and goals. This paradigm shift introduces new challenges, particularly in how testing should be guided when prompts, rather than code, become primary development artifacts. To address this challenge, we propose Prompt Coverage Adequacy, a novel coverage criterion designed to support the testing of code generated from task descriptions. Prompt Coverage Adequacy serves as an analog to traditional code coverage, but operates at the level of prompts used in LLM and agent-based programming. Specifically, it measures how well a given test suite satisfies the requirements expressed in a prompt by leveraging the attention mechanisms of LLMs. We evaluate a simple instantiation of this criterion, based on attention boosting, across two datasets and multiple LLMs. Our results demonstrate that Prompt Coverage is associated with fault-detection effectiveness and can uncover over 30+% more faults than traditional code coverage when used to guide test generation. These findings suggest that Prompt Coverage Adequacy can serve as a foundation for developing testing metrics better suited to the emerging paradigm of LLM-driven software development, addressing the limitations of classical coverage criteria in this new context.
Performance evaluation in AI systems commonly assumes that random dataset splits produce independent and identically distributed (i.i.d.) subsets. We show that this assumption often breaks down in spatiotemporally correlated domains such as aerial surveillance, precision agriculture, and medical imaging, leading to two systematic failures: data leakage, where correlated samples span training and validation splits and inflate performance estimates, and hidden stratification, where errors on minority subpopulations are obscured by aggregate metrics. To address these issues, we propose a unified evaluation and training framework for spatially correlated data. We introduce Structure-Aware Stratified Partitioning (SASP), which constructs validation splits that reduce spatiotemporal leakage while preserving meaningful class balance, and Curriculum Distributionally Robust Optimization (CDRO), a curriculum-based relaxation of distributionally robust training that stabilizes optimization under these stricter splits. Across multiple benchmarks, this combination yields consistently improved generalization, more reliable confidence calibration, and exposes failure modes that remain hidden under conventional random-split evaluation.
Large language models (LLMs) have demonstrated strong capabilities in software engineering tasks, such as code generation, library recommendation, and dependency configuration. However, recent studies show that LLMs may suffer from package hallucination, where they generate non-existent or invalid package names. These hallucinations can be exploited in software supply chain attacks, as attackers may register malicious packages under hallucinated names. Therefore, mitigating package hallucination is important for improving the reliability and security of LLM-assisted software development. In this paper, we introduce BOUND, a lightweight localized model editing framework for mitigating package hallucinations in LLMs. BOUND formulates package hallucination mitigation as a package-validity boundary editing problem, where the boundary refers to the model's ability to distinguish valid packages from hallucinated package names under a given task context. It first locates modules related to package hallucination through a risk-aware localization strategy, and then edits these modules with lightweight LoRA adapters using a boundary-aware objective that reinforces valid packages, suppresses hallucinated packages, and preserves locality behavior. Experimental results show that BOUND effectively reduces package hallucinations while preserving valid package recommendations. In the package recommendation task, BOUND reduces package-level hallucination rate (Package-HR) by 79.9% on edit prompts and by 65.4% on unseen prompts. The learned package-validity boundary further generalizes to other package-related tasks, reducing Package-HR by 12.8% in code generation and by 34.0% in pip install recommendation. These results show that BOUND refines the package-validity boundary of LLMs and improves the reliability of package-related outputs.
Motivated by the challenge of stabilizing a general unknown linear dynamical system (LDS) from observations, we study the natural prerequisite of online prediction. Our goal is to achieve sublinear regret with a memory footprint that adapts to the intrinsic complexity of the dynamics rather than the full hidden -- state dimension. We focus on the practically central regime of systems with low instability complexity -- eigenvalues outside the real stable interval that do not decay rapidly, together with non-semisimple modes-potentially embedded in an otherwise stable real spectrum of much higher dimension; we write $k$ for this count. This regime is the primary setting in which stabilization is plausible: we show that many systems with high instability complexity cannot be stabilized without exponentially large controls. Thus, prediction is meaningful for stabilization precisely when the instability complexity is small. Within this regime, we introduce a unified online algorithm that handles every LDS (including non-diagonalizable systems with complex or exploding modes) with a learnable parameter count of $\widetilde{O}(k)$. Finally, we prove a lower bound showing that $k$ is a valid complexity measure: any filter-based predictor needs at least $k$ filters. Experiments corroborate our theory: on a high-dimensional system, our predictor sharply outperforms prior methods at an equal parameter budget.
Large Language Models are increasingly deployed in emotional-support contexts and crisis-related situations. Nevertheless, their cross-lingual abilities in these circumstances remain underexplored. Existing benchmarks emphasize multilingual performance but rarely examine crisis-related empathy and cultural grounding in low-to-mid-resource languages. We introduce SPLIT, a 500-prompt benchmark designed to evaluate LLM consistency in generating emotionally grounded responses across five categories: Stress, Panic, Loneliness, Internal Displacement, and Tension. We evaluate three technically diverse LLMs across three dimensions: Empathetic Accuracy, Linguistic Naturalness, and Contextual & Cultural Grounding. The framework aims to assess and compare the quality of LLM responses in both English and Ukrainian languages, as well as to explore the reliability of the LLM-as-a-jury paradigm. Our findings reveal that Gemini-2.5-Flash and LLaMA-3.3-70B-Instruct degrade when transitioning to Ukrainian, while DeepSeek-V3 remains comparatively stable within our benchmark. We additionally find that human and AI evaluators agree weakly on empathy and naturalness but diverge on cultural grounding. We further argue that producing Ukrainian text is not equivalent to producing Ukrainian emotional support. Our findings may assist in the future development of more culturally tailored benchmark designs, as well as encourage a stronger emphasis on human-centered evaluation.
Safe completion requires models to provide useful assistance without enabling harm, but this behavior is difficult to evaluate with isolated prompts. We introduce OpenSafeIntent, a benchmark of controlled prompt-sets that vary intent while holding the underlying task fixed. Each datapoint contains benign, dual-use, and malicious variants of the same task. This design lets us evaluate whether models calibrate assistance across intent shifts, rather than merely appearing safe on average. Across a broad model suite, we find that prompt-level safety hides important failures: models often fail to remain safe across matched intent variants, dual-use behavior is brittle under paraphrase, high-level answers on risky topics are not reliably safe, and responses that reframe ambiguous requests into safer tasks are substantially less likely to cross the safety boundary. Our results suggest that safe completion should be evaluated as intent-calibrated behavior over controlled task variants, not as a single safety-helpfulness tradeoff over independent prompts.
Anomaly detection is a critical and evolving field in Machine Learning, with applications targeting different domains such as cybersecurity, finance, healthcare, manufacturing and IoT (Internet of Things) systems. Traditionally, anomaly detection algorithms have been designed using both supervised and unsupervised learning paradigms. The fundamental challenge in real-world anomaly detection scenarios is related to the inherent class imbalance (anomalies are typically rare) and, for supervised methods, to the scarcity of labelled anomalous data. Indeed, labelling is both expensive and time-consuming. Conversely unsupervised methods do not require labelling, but may suffer from high false positive rates when deployed in safety-critical applications. In this work we introduce a novel unsupervised algorithm for anomaly detection in time series based on the Haar discrete wavelet and a suitably designed $t$-test. We establish the theoretical foundation of the proposed $t$-test and, through extensive experimentation across 343 datasets, demonstrate that our algorithm outperforms state-of-the-art unsupervised and self-supervised benchmarks.
Disaggregated LLM serving runs prefill and decode on separate GPU pools to keep the two phases from interfering. In practice, this creates a new asymmetry: under bursty, heavy-tailed workloads prefill nodes saturate while decode nodes have compute underutilized, and on a production-style A100 cluster with 2 prefill and 2 decode nodes (2P2D), we find that prefill execution accounts for only 2-23% of P95 Time-to-First-Token (TTFT). Queuing and inter-node GPU-GPU KV-cache transfer account for the rest. We present a proactive prefill-deflecting scheduler that lets decode nodes serve prefill phase of requests as chunked-prefill steps interleaved with their in-flight decode batches. For each queued request, we estimate the TTFT it would see on the prefill node, and on every decode node, search for the largest chunk schedule that keeps in-flight decodes within their Time-Between-Tokens (TBT) SLO and deflect when the decode path helps tail latency. Because the prefill phase of deflected requests runs in place on the decode node, the inter-node KV transfer is eliminated. Implemented on vLLM and evaluated on production-style traces with DeepSeek-V2-Lite, our approach reduces P95 TTFT by upto 81% and raises SLO attainment by upto 79% over state-of-the-art disaggregated schedulers, at sub-millisecond per-request routing cost.
Autonomous surface vehicles vary widely in hydrodynamic and actuation characteristics, yet most controllers are designed for single-platform deployment. We present an adaptive reinforcement learning approach for trajectory tracking that enables zero-shot cross-platform deployment using a single policy. Since the deployment platform's dynamics are unknown to the policy, we address cross-platform generalization with the standard partial-observability approach of conditioning on interaction history, employing a teacher-student architecture in which a learned module infers a latent representation of the platform dynamics. The policy is trained in simulation under randomized vessel dynamics and is deployed zero-shot to two real-world platforms without any fine-tuning, despite relying on a simple analytical dynamics model rather than a high-fidelity hydrodynamic simulator. In real-world experiments on two different platforms, the adaptive policy outperforms non-adaptive learning-based baselines by up to 58% in position mean absolute error while approaching the tracking accuracy of a platform-specific tuned controller.
Evaluating LLM agents on benchmarks like SWE-Bench and GAIA can be expensive, time-consuming, and requires complex infrastructure. A single evaluation can cost thousands of dollars and take days to complete. In contrast, non-agentic LLM benchmarks that test individual capabilities (e.g., reasoning, code generation) are fast and cheap to run. In this paper, we investigate whether performance on expensive agentic benchmarks can be accurately predicted by the performance on a small, carefully selected subset of atomic evaluation instances. We introduce PACE, a framework that constructs proxy benchmarks by selecting instances from existing non-agentic evaluations whose aggregate scores most reliably predict model performances on agentic benchmarks. Given a pool of candidate instances spanning atomic capabilities, PACE fits a regression that maps a model's scores on a compact subset of source instances to its score on the target agentic benchmark. The subset itself is curated by combining two complementary instance-selection strategies, target-relevance local selection and globally informative global selection. We apply PACE to the 4 target agentic benchmarks in this paper, which yields PACE-Bench, the concrete proxy benchmark that we evaluate in the paper. Experiments across 14 models, 4 agentic benchmarks, and 19 non-agentic benchmarks show that PACE-Bench predicts agentic scores with leave-one-out cross-validation (LOOCV) mean absolute error (MAE) under 4%, Spearman correlation above 0.80, and pairwise model-ranking accuracy around 85%, all at much less than 1% of the full agentic evaluation cost. We further analyze the selected proxy instances, revealing which skills each agentic benchmark uniquely demands. PACE enables practitioners to obtain reliable estimates of agentic performance during model development, selection, and routing, without the overhead of full agent evaluation.
Quantum software testing (QST) checks whether quantum programs behave according to their intended specifications. A key requirement for QST research is a benchmark that supports rigorous empirical evaluation on programs that are testable and better reflect current software development practices. However, existing studies heavily rely on small hard-coded or circuit-level benchmarks, while available quantum programs are scattered across repositories without clear selection criteria, which limits fair comparison and systematic reproducibility. To this end, we present Qolumbina, a benchmark infrastructure for controlled QST experiments on scalable quantum programs. Qolumbina curates 40 programs from open-source repositories, turns them into test-ready subjects through systematic selection, refactoring, specifications, test case examples, unit tests, and standardized interfaces. We also propose QST-oriented criteria to characterize quantum programs along functionality, output behavior, development complexity, and quantum-specific execution complexity. Using these criteria, our empirical study shows that Qolumbina covers diverse testing-relevant properties and supports scalability analysis beyond fixed-size circuit benchmarks. Through controlled experiments with two recent QST approaches, we demonstrate the feasibility of using Qolumbina for execution-cost and fault-detection studies, and highlight backend-dependent effects that can influence QST result interpretation.
Multimodal large language models must continually adapt to evolving tasks and domains, yet standard continual learning metrics mainly measure whether old answers remain correct, leaving the stability of multimodal grounding largely unexamined. We study this overlooked failure mode and ask whether a continually adapted MLLM can preserve not only what it answers, but also how it uses visual, textual, OCR, chart, and document evidence. We identify \emph{hidden evidence-use forgetting}, where answer accuracy is retained while the model silently shifts toward different or less grounded evidence channels, and propose \textsc{RCL}, a replay-free reliance-constrained continual learning framework. \textsc{RCL} freezes the previous checkpoint as a behavioral reference, estimates teacher and student evidence-reliance profiles through counterfactual channel interventions, and jointly optimizes task learning, prediction preservation, and reliance preservation without adding inference-time cost. Across CoIN, COAST, MCITlib, and an evidence-sensitive multimodal stream, \textsc{RCL} consistently improves final performance and reduces forgetting over replay-free, PEFT, routing, and memory-assisted baselines, while substantially lowering modality reliance drift, dominant evidence flips, and hidden forgetting rates. These results suggest that robust continual multimodal learning requires preserving the evidence path behind correct answers, not merely the answers themselves.
Mirror Illusion Art is a novel reflection-conditioned 3D illusion where one object yields two target appearances (front and mirror). The task is formulated as inverse design from two target 2D images (front and mirror) to a printable 3D object with geometry and texture. Prior topology-driven and shadow-based approaches demand substantial manual effort, optimize shape only, and often yield non-smooth or incomplete geometry. To address these challenges, we propose AutoMIA, an automated Mirror Illusion Art design pipeline that jointly optimizes shape and color. To stabilize optimization and suppress artifacts, four mechanisms are introduced: (1) projection-alignment component (PAC) selection to reduce surface noise, (2) position-weighted adaptive (PWA) suppression for background noise, (3) internal voxel preservation (IVP) to prevent internal fractures, and (4) shape-color decoupled (SCD) optimization that balance shape and color optimization. AutoMIA generate diverse smooth Mirror Illusion artworks successfully both in the digital and physical world, with only around 76s design time and 2.6 GB memory on average using a single RTX 3090, advancing inverse graphics and computational design. Our code is available at https://github.com/zxp555/AutoMIA.
Multimodal large language models must adapt to evolving tasks and domains, yet continual improvement under bounded deployment footprint remains difficult because repeated parameter updates or growing replay stores can accumulate adaptation state over time. We study fixed-footprint continual adaptation: the deployed adaptation state is kept under a fixed memory budget, while the backbone model is left unchanged and task-specific updates are externalized. We propose InduceKV, a retrieval-based method that stores each selected training prefix as an attention-ready memory entry, consisting of a frozen retrieval key and compact layerwise key--value (KV) payloads that can be appended to the model's self-attention cache. Under a strict memory budget, InduceKV constructs a compact inducing set through bilevel selection: a lightweight calibration is fit for retrieval, while the selected memory balances current-task likelihood, anchor-based retention, and coverage in the frozen retrieval space. Across task-incremental instruction tuning, continual VQA, domain-incremental adaptation, and lifelong multimodal instruction tuning, InduceKV consistently improves over PEFT, MoE, replay, and prompt-retrieval baselines under matched memory budgets. We further report backbone-matched, stage-1 CoIN, compute-matched, and scalability diagnostics, showing that the gains are not due to a stronger backbone, replay alone, or an unbounded candidate pool.
Large language models now score near ceiling on general benchmarks, but these aggregate measures reveal little about how models behave within single disciplines. Existing art-focused evaluations rely on synthetic questions and rarely report item-level properties. This paper introduces EduArt, an educational-level benchmark for art-historical knowledge and visual reasoning in multimodal LLMs. EduArt comprises 871 human-authored questions from Italian secondary-school exercises and US Advanced Placement Art History exams, spanning two languages and seven formats from multiple choice to in-text word placement and error identification. Twelve models from six provider families were evaluated under a default answer-only condition and a motivation condition requiring written justification, and characterized using Classical Test Theory and a logistic regression isolating the effects of format, language, image presence, and model. The benchmark showed strong psychometric properties (mean discrimination 0.514, 82.3 percent good discriminators), while multiple-choice accuracy saturated near ceiling for six models, showing recognition formats alone cannot distinguish frontier models. Format was a strong independent predictor of accuracy: models exceeding 94 percent on multiple choice fell to 23.9 percent on open completion (Claude Opus 4.6) and 6.2 percent on error identification (Claude Sonnet 4.6). The motivation condition changed accuracy in a predominantly negative, family-dependent direction. These dissociations indicate that art-historical knowledge and the ability to deploy it are distinct capabilities, and that single-format benchmarks overestimate what models can reliably do. Mapping this capability profile is a precondition for responsible use of multimodal LLMs in art-historical scholarship, where tasks demand producing and manipulating content rather than selecting from fixed options.
Although neural networks are remarkably effective, their underlying optimization principles remain theoretically elusive, often characterized by non-convex landscapes and stochastic heuristics. In this work, we propose a paradigm shift by replacing the discrete training problem of shallow neural networks with a well-posed continuum variational surrogate. We identify a family of $λ$-convex functionals over parameter densities in weighted Sobolev spaces and prove that these variational problems are globally well-posed, stable, and exhibit unexpected almost $C^3$ regularity. Unlike existing Wasserstein-based or Mean-Field approaches, which often face limited regularity and discretization challenges, our formulation provides direct access to elliptic regularity and convex analysis. This allows us to prove that the optimal parameter density can be obtained by solving a single linear system, bypassing iterative optimization entirely. We establish explicit generalization error controls at a rate of $1/α$ relative to the regularization parameter, and prove that finite-width networks of size $N$ achieve the continuum optimum at an $O(1/N)$ rate. This perspective bridges the gap between the Neural Tangent Kernel (NTK) and feature-learning regimes, providing a principled framework for understanding over-parameterization through the lens of variational calculus.
Time-normalized f0 contours of Mandarin words in conversational speech have been shown to be predictable in part from their contextualized embeddings (CEs). The present study investigates whether CEs also predict spoken word duration for 7470 tokens of Mandarin monosyllabic CV words extracted from a Mandarin corpus of spontaneous speech. We show that CEs indeed are predictive for duration, above chance level, not only at the type level, but also at the level of individual tokens, as indicated by the results obtained with the type-wise and token-wise permutation baselines. We also show that the predicted durations are sufficiently precise to back-transform predicted f0 contours in [0,1] normalized time to contours on the ms time scale. The resulting predicted contours approximate empirical contours and also outperform a permutation baseline.
The silhouette is one of the most widely used measures to assess the quality of a $k$-clustering of a dataset of $n$ elements. Its evaluation requires no information beyond the clustering assignment. In addition, the silhouette is extremely easy to interpret, providing a score to measure the quality of a clustering as a whole or for each element. The exact computation of the: (i) silhouette of each element of a dataset; and (ii) the global silhouette of the clustering; require $Θ(n^2)$ distance calculations, under general metrics. The quadratic complexity $Θ(n^2)$ is extremely prohibitive, especially on massive modern datasets. Surprisingly, existing approximate methods using $O(n^2)$ distance calculations are heuristics not offering provable and controllable guarantees on the quality of their results. We introduce the first rigorous and efficient algorithms to estimate: (i) the (local) silhouette of each element of a dataset; and (ii) the (global) silhouette; of any metric $k$-clustering. Our methods, based on sampling, perform $O(nk\varepsilon^{-2}\ln (nk/δ))$ distance computations, and provide estimates with additive error $O(\varepsilon)$ with probability at least $1-δ$. That is, parameters $\varepsilon$ and $δ$ in $(0,1)$ control the trade-off between accuracy and efficiency. We also introduce a scalable and distributed design of our methods for the MapReduce and Massively Parallel Computing (MPC) frameworks. Our distributed algorithms use a constant number of rounds and sublinear local memory. Finally, we perform extensive experiments against state-of-the-art approaches. The results show that our new techniques yield the best trade-off between accuracy and efficiency for both local and global silhouette estimation. In addition, our methods scale efficiently to massive datasets for which an exact computation of the silhouette is not practical.
Large-scale battery energy storage systems (BESSs) require O&M decisions that combine alarms, cell-level measurements, device topology, diagnostic tables, historical cases, and maintenance documents. Monitoring platforms can flag threshold violations, but they often cannot explain whether voltage inconsistency, resistance drift, short-circuit risk, capacity divergence, or thermal abnormality needs intervention. This digest presents a traceable BESS fault-diagnosis assistant that uses retrieval-augmented multi-agent reasoning to connect operational data, domain knowledge, visual evidence, and report generation. Reliability is improved through BESS-specific task routing, schema-constrained natural-language database access, hybrid text-image retrieval, and evidence-based answer synthesis. Preliminary internal evaluation is reported for routing, database access, and diagnostic reasoning.
Long-running adaptive intelligent agents face a structural tension between knowledge consolidation and information integrity. Memory consolidation is conventionally treated as an agent-changing operation: a model is fine-tuned, a prompt rewritten, a policy distilled, or a reflection appended to the context that governs future behaviour. In regulated autonomic deployment this is a liability because the agent operates under commitments and audit contracts that bind to a specific, cryptographically certified identity. We propose to treat consolidation not as a mutation of the planner or the identity manifest, but as a deterministic function f: M^ep -> M^sem over episodic memory whose output is a separately addressable semantic knowledge layer; the identity hash does not read M^sem, so consolidation updates knowledge without changing the agent's certified identity. We give a formal account of the agent representation, prove identity invariance through a structural lemma on the manifest's hash-input set, specify a deterministic aggregation algorithm whose outputs are auditable database rows with explicit confidence and supporting-event provenance, and validate the construction with synthetic experiments demonstrating per-field correctness, byte-equal identity across consolidation passes, and a mean 79.82% reduction in unproductive planner attempts (95% BCa CI [78.02%, 81.49%] across 10 seeds) against a calibrated Bayesian-shrunk baseline. The construction is a knowledge-update discipline for autonomic agents in which lessons accumulate as queryable facts while the agent's certified identity remains byte-equal across its operational lifetime, with an embodied service agent as the running case study.
Multivariate time-series models for prognostics are often evaluated by point prediction accuracy, yet their internal states rarely expose a coherent degradation process. We study liquid neural networks as latent dynamics models for aircraft engine health monitoring on the C-MAPSS benchmark. The proposed model encodes a history window into a latent state, evolves that state with a liquid transition model, and decodes future sensor observations. To separate health evolution from operating-condition variation, the latent state is factorized into degradation and condition components. Remaining useful life, monotonic risk, and latent-consistency losses supervise the degradation component, while condition prediction and decorrelation losses discourage operating-condition leakage. Across FD001--FD004, the full disentangled model improves overall sensor forecasting RMSE from 0.2438 for a GRU baseline to 0.2266, with the largest gains on the multi-condition subsets FD002 and FD004. The learned degradation state also forms a clearer temporal degradation axis, reaching an average state-speed Spearman correlation of 0.5960. Direct remaining-useful-life regression remains stronger for the GRU baseline, indicating that the proposed representation is currently more effective as an interpretable world model for degradation dynamics than as a calibrated lifetime regressor. These results suggest that liquid latent dynamics can bridge predictive maintenance forecasting and inspectable health-state modeling.
Newer lightweight convolutional neural networks are often presented as improving predictive performance and deployment efficiency, but such claims require controlled evaluation. This study compares nine lightweight CNN model packages across CIFAR-10, CIFAR-100, and Tiny ImageNet under a shared downstream protocol. We report top-1 accuracy, macro F1, top-5 accuracy, parameter count, FP32 storage, GMACs, batch-size-1 latency on an NVIDIA L4 and AMD Ryzen 5 5500U CPU, peak PyTorch CUDA allocated tensor memory, and point estimate Pareto frontiers. EfficientNetV2-S achieves the highest observed top-1 accuracy on CIFAR-10 and CIFAR-100 at 97.57% and 86.98%, while RepViT-M1.0 leads Tiny ImageNet at 79.87%. EfficientNet-B0 remains within 0.22, 0.85, and 1.79 percentage points of the best result on the three datasets while using approximately 79% fewer parameters and 86% fewer GMACs than EfficientNetV2-S. It also appears on every evaluated accuracy and resource Pareto frontier, making it the most consistently competitive intermediate-budget option. MobileNetV3-Small has the lowest GMAC count, is the fastest model under both CPU thread settings, and records higher observed accuracy than MobileNetV4-Conv-S on all three datasets. Under random initialization, it leads MobileNetV4-Conv-S by 2.55, 1.76, and 0.99 points, with paired test-set intervals excluding zero for the fixed trained models. EfficientNet-B0 remains 3.29, 10.10, and 17.54 points below its pretrained counterpart after 100 epochs of scratch training, despite requiring about five times the recorded training time. SqueezeNet1.1 has the fewest parameters and lowest peak CUDA allocation, but substantially weaker accuracy. Latency rankings differ sharply between the L4 and CPU environments, showing that GMACs alone do not predict measured inference performance. Overall, newer designs provide selective rather than universal gains
Using molecular large language models (LLMs) as a unified framework for understanding molecular structures and functions is emerging as a new trend in tasks such as molecular design and drug discovery. However, these models struggle to fully capture the visual representation of molecular structures, limiting their potential. While existing molecular vision-language models (VLMs) show promise, they still face challenges in structural alignment and lack the necessary topological modeling for accurate molecular understanding. To address this, we propose MolSight, a graph-aware vision-language model framework designed to enhance the understanding of molecular images by VLMs. MolSight integrates a Molecular Topology Module to inject chemical-bond adjacency information into vision tokens, and a Molecular Grounding Module to align visual features with chemical symbolic semantics. Our experiments demonstrate that MolSight significantly outperforms existing VLMs, molecular LLMs, and specialized tools across multiple chemical visual understanding tasks, achieving a new level of molecular image reasoning.
Automated authoring of Gherkin Behavior-Driven Development (BDD) acceptance criteria remains a manual bottleneck in requirements engineering. This study investigates whether epic-organized LLM-generated Gherkin produces higher quality and coverage than requirement-aligned generation. We compare our Timeless (an epic-organized LLM pipeline) approach against a naive large language model (LLM) baseline on four requirements documents (107 requirements) from the PURE dataset. Evaluation covers structural metrics, automated requirement coverage via TF-IDF and dense embeddings, and blind expert assessment by four researchers. In our evaluation, the JSON-constrained pipeline produced structurally valid scenarios across all generated outputs, while the zero-shot baseline achieved 99% structural validity. Semantic coverage was comparable to the baseline, with Timeless achieving 94.3% semantic Requirement Coverage Rate compared with 92.9% for the baseline. TF-IDF produced lower coverage scores for the epic-organized output, suggesting that lexical metrics may miss coverage when scenarios paraphrase requirements at a higher level of abstraction. Expert raters prefer the epic-organized strategy on Correctness (4.61 vs 4.14), Executability (4.61 vs 4.07), and Completeness (4.31 vs 3.50). Overall, the results suggest that epic-organized generation can improve perceived Gherkin quality while maintaining comparable semantic coverage, although broader replication is needed before generalizing this finding.
Online multimodal knowledge editing requires injecting a continual stream of visual-textual corrections into multimodal large language models (MLLMs) with bounded overhead and minimal disruption to unrelated behaviors. Existing editors mainly emphasize edit reliability and long-horizon stability, but rarely control the semantic boundary of each edit. Our pilot analyses of post-edit behaviors and internal neuronal activities reveal a scope gap behind reliable edits: instance-level success neither guarantees transfer to valid cross-modal variants nor prevents leakage to unrelated inputs, while edit-related cross-modal responses concentrate in deeper semantic layers. Therefore, we formulate Edit-Scoped Generalization, reframing online MLLM editing from merely correcting an instance to controlling the propagation boundary of each edit. To this end, we propose ScopeEdit, a scope-aware online editor that decomposes each update into a modality-local absorption branch and an evidence-gated shared generalization branch. The local branch supports stable edit absorption, whereas the shared branch enables cross-modal propagation only when visual and textual evidence are sufficiently aligned. Both branches perform scope-separated write geometries in orthogonal low-rank spaces and maintain branch-wise preconditioners via Sherman--Morrison recursions, yielding constant per-edit overhead. Extensive experiments across diverse benchmarks, long-horizon edit streams, MLLM backbones, real-world VLKEB scenarios, and complex vision-language architectures show that ScopeEdit consistently improves the trade-off between in-scope cross-modal transfer and out-of-scope locality, while preserving edit reliability, stability and online efficiency. Our code is available at https://github.com/lab-klc/ScopeEdit.
Ontology learning (OL) aims to automatically construct structured knowledge models from text, yet progress remains fragmented across methods, domains, and evaluation practices. Despite decades of research, OL lacks a shared infrastructure for systematic evaluation and ontology access. This absence has hindered progress and fragmented research, leaving the central challenges of OL largely unaddressed. We introduce OntoLearner, a modular, cross-domain, and first-of-its-kind framework that unifies ontology access, large language model (LLM)-driven learning pipelines, and standardized benchmarking. OntoLearner releases 180 machine-readable ontologies spanning 22 domains and provides pipeline-ready datasets with train/dev/test splits for three core OL tasks: term typing, taxonomy discovery, and non-taxonomic relation extraction. Using this infrastructure, we conduct a large-scale empirical study of OL, evaluating 22 retrieval models and 12 LLMs across domains and tasks. The results converge on a finding that reframes the central challenge of OL: failure modes scale with ontological complexity rather than model size or architectural sophistication. The primary bottleneck is not model capability, but a structural mismatch between how models encode knowledge and how ontologies organize it. These findings establish that effective OL is reachable through the cross-domain, multi-task benchmarking enabled by OntoLearner. OntoLearner is open-source (MIT license) at https://github.com/sciknoworg/OntoLearner/.
This technical report describes our system for Task 1 of the DCASE 2026 Challenge, which aims to classify heterogeneous audio recordings according to the Broad Sound Taxonomy (BST). The task requires both accurate second-level prediction and consistency with the top-level taxonomy. Our system is built on CLAP-based audio-text representations and is improved along three strategies: expanding the training set with a filtered subset of BSD35k, enhancing acoustic modeling with feature-specific branches, and refining predictions using hierarchy-aware classifiers and KNN-based post-processing. Among the acoustic features considered, the log-STFT branch provides the strongest single-model performance. With KNN-based post-processing, our best single system achieves a hierarchical F1 score (Hier. F1) of 80.84% on the BSD10k-v1.2 set under the same evaluation protocol as the baseline. We further construct ensemble systems by combining models with complementary acoustic features and classification heads, achieving Hier. F1 scores of 81.25% and 81.18%, respectively.
Vision-Language Models (VLMs) are increasingly applied in medical tasks such as pathology description, report generation, and visual question answering. Medical Image Quality Assessment (MIQA) supports diagnostic accuracy and patient safety by determining whether images meet the standards required for clinical decision-making. Automating MIQA with VLMs may reduce workload, but their behavior under real-world conditions, where images may be degraded or textual context may affect judgments, should be further explored before deployment. We benchmark VLMs on medical image quality using the MediMeta-C dataset zero-shot across seven corruption types and five severity levels. We evaluate sensitivity to degradation patterns, the effect of corruptions on embedding geometry, and whether textual attributes (demographics, expertise, infrastructure, institution) alter scores. Across 16 VLMs and seven modalities, pixelation produced the largest score reductions (mean -20.58%, up to -34.4% for OCT), whereas brightness had limited effect (-0.81%). Embedding displacement was associated with score changes. Same-family models showed correlations of 0.67-0.83; some produced increases up to +31% for corrupted mammography. Textual attributes affected scores: institutional prestige raised them +17.15%, and equipment age lowered them -14.7%. The largest changes were +95.62% (InternVL-8B) and -37.7% (MedGemma). Current VLMs show limitations for medical image quality assessment. Pixelation, a privacy-preserving transformation, reduces performance, indicating a trade-off between patient privacy and reliability. Sensitivity to contextual metadata indicates limited objectivity and marks metadata as a privacy and bias source. Privacy protection and objective quality assessment are related requirements for use.
Large language models (LLMs) are often asked to produce JSON conforming to a fixed schema, powering information extraction, tool calling, agentic planning, and knowledge-graph construction. Measuring how closely an output matches a gold reference is essential yet surprisingly hard: exact match is brittle, text similarity ignores structure, and an LLM judge is expensive, opaque, and non-deterministic. We address this with Object Aligner (OA), an open-source Python library that scores two JSON objects deterministically by recursively aligning their trees (the Hungarian algorithm for unordered collections, sequence alignment for ordered ones) and awarding partial credit at the granularity the schema declares. The Object Aligner is configured entirely through a set of JSON Schema extensions, so adapting it to a new task involves annotating a schema rather than writing code. Complex structured data, however, are rarely flat trees: records may form graphs or hypergraphs keyed by arbitrary identifiers, breaking the assumptions of prior similarity metrics. Our central contribution, referential alignment, closes this gap by inferring a bijection between gold and candidate identifiers and scoring every reference through it, so the score is invariant to relabeling. Since recovering this bijection exactly is graph isomorphism, the Object Aligner approximates it with Weisfeiler-Leman color refinement. An order-sensitive sequence regime targets ranking and planning. Since the same alignment localizes every mismatch, the Object Aligner emits ranked repair suggestions at no extra cost. Used as a reward inside the GEPA prompt optimizer, Object Aligner helps or stays neutral across all datasets.
Low-voltage load forecasting is an important component in current and future energy systems with a high degree of electrification and decentralized generation. However, current forecasting methods require significant manual effort, often lack uncertainty estimation and proper peak prediction, and they are often not adequately evaluated in terms of grid requirements. In the present study, we provide an extensive evaluation of short-term net load forecasts of 200 real-world low-voltage feeders with a focus on the rapidly evolving time series foundation models. Our study compares Chronos-Bolt, Chronos-2 and TabPFN-TS to six baseline models and demonstrates superior performance, in particular for Chronos-2. An ablation study, in which weather covariates are omitted, shows that time series foundation models adapt to increased uncertainty, despite the importance of weather information. A novel application-oriented metric links the model's forecasting capabilities in peak prediction to the trade-off in grid asset planning and operation between cost reduction and minimizing the risk of failure.
Neural TTS systems can sound natural across languages, but naturalness does not guarantee the preservation of sound contrasts that distinguish words from their grammatical forms. Standard metrics like MOS do not test for this. We propose a classifier-based framework that audits TTS output against language-specific phonological patterns using human speech as a benchmark. Testing Assamese advanced tongue root (ATR) vowel harmony with Meta's MMS TTS, we show that a classifier trained on human speech transfers to synthesized speech with minimal loss. The faithfulness audit reveals that [+ATR] mid vowels are realized as [-ATR] in 1/3 tokens despite an underlying [+ATR] specification, a bias absent in human speech. At the word level, predicted ATR labels classify harmony more accurately than transcription labels, indicating a gap between intended and produced phonology. The framework offers task-specific diagnostics and generalizes to other phonological contrasts with measurable acoustic cues.
Rewriting inputs to improve frozen downstream models has become a common strategy in modern NLP pipelines. Prior work on incremental dialogue discourse parsing (DDP) shows that supervised clarification models can rewrite fragmentary or underspecified utterances, such as resolving ellipsis or references, to improve parsing accuracy. In this work, we revisit this idea under realistic deployment conditions, where no clarification supervision is available and the clarifier must rely on zero-shot prompting or feedback from a frozen parser. Across three Segmented Discourse Representation Theory (SDRT) datasets and multiple parsers, we find that last-utterance clarification is far less reliable than suggested by supervised settings. Parser-agnostic rewriting often introduces more regressions than repairs, as edits that enable fixes also disrupt discourse cues relied upon by the parser. A best-of-8 rewriting analysis further reveals a practical ceiling: a large fraction of errors are not repairable through input rewriting alone. A parser-aware clarifier trained with GRPO reduces regressions by up to 37% by learning conservative abstention, yet still fails to produce selectivity-aware clarifications that consistently improve parsing. Together, these findings recast clarification as a selective intervention problem. We identify rewritability prediction, deciding whether an utterance is repairable before intervention, as the key missing capability for input-side optimization of frozen discourse parsers, and a critical direction for improving agentic pipelines more broadly.
We study the challenging problem of novel view video synthesis from single images or monocular videos. Existing methods, which operate under the assumption that pre-trained video models lack native novel view synthesis capability and enforce view alignment via camera conditioning, task-specific fine-tuning, or stepwise hard denoising guidance, often suffer from artifacts and compromised global scene consistency. In this paper, we introduce NeoMap, a novel training-free framework designed to locate high-fidelity, view-consistent novel view solutions from general pre-trained video models. The key to our approach is the core insight that promising novel view solutions are inherently encoded within the natural video data manifold learned by pre-trained models, and the core challenge is simply to locate this optimal solution. We solve this via our core mechanism: convergent manifold alternating projection iterations that optimize the initial noise. Extensive experiments demonstrate that NeoMap significantly outperforms all existing methods across 3 standard novel view synthesis benchmarks, including the challenging Tanks-and-Temples, LLFF and DAVIS datasets, achieving state-of-the-art generation fidelity and top-tier view consistency.
In this paper, we describe NAVER LABS Europe's submission to the instruction-following speech processing short track at IWSLT 2026. We participate again in the constrained setting, developing systems capable of jointly performing ASR, ST, and SQA from English speech into Chinese, Italian, and German. Building on our previous submission, ranked first in last year's short track, we update our multi-stage training pipeline by replacing the speech projector with SpeechMapper, a method for learning a speech-to-LLM embedding projector using only ASR data. In addition, we introduce a synthetic SQA dataset, fakACL, composed of artificially generated scientific presentations. This dataset is built by prompting the LLM backbone, segmenting the generated talks, and synthesizing speech with SeamlessM4T-large-v2. The combination of an improved speech projection mechanism and domain-specific synthetic data allows our model to outperform last year's best short-track system, while being considerably more compact and relying on a weaker LLM backbone. This year's results place our system tied for first place in the overall short track ranking.
We propose a model agnostic methodology to measure lag relevance in machine learning forecasting models applied to univariate time series. Particularly, we are working in the context of time series using the frameworks of Ghost variables and Shapley values, together with additive importance measures, to introduce the auto-relevance and partial auto-relevance functions as the lag importance values. Additionally, we propose a novel method to replace absent features in coalition based methods with a one step forecast from the same model. We evaluate these proposals under different simulations and real data cases. This combined framework perspective is particularly suitable for time series. In addition, to show our discoveries we use a pull of models from the seasonal ARMA family and recurrent neural networks. We found that the calculated relevance measures successfully demonstrate the expected lag structure in almost all cases.
A/B testing is the gold standard for selecting the better algorithm in online services. While offline evaluation has attracted attention as a safer alternative due to the high experimental costs and the potential risk of degrading user experience and revenue in A/B testing, it is widely recognized that the estimation accuracy of offline evaluation is substantially lower. As a result, final selection decisions are typically made through A/B testing. Contrary to this conventional view, we reveal a counterintuitive phenomenon in which A/B testing can produce a higher algorithm selection error rate than offline evaluation. This occurs because the sample mean estimator used in A/B testing does not induce positive correlation, which is crucial for reducing critical selection errors, namely underestimating the truly superior algorithm and overestimating the truly inferior one. In contrast, offline evaluation methods unintentionally generate this beneficial correlation by relying on shared offline data when estimating and comparing the performance of multiple algorithms. Building on this insight, we propose an estimator that intentionally induces positive correlation to improve algorithm selection in A/B testing. The key idea is to introduce a hypothetical middle algorithm and to estimate the performance difference between algorithms A, M, and B in a stepwise manner using shared data at each step. This approach enables the application of offline evaluation techniques in each step, thereby inducing positive correlation and reducing critical selection errors. Furthermore, we derive the optimal middle algorithm regarding the resulting variance and analyze its advantages over existing methods through bias-variance analysis. Experiments on real-world data demonstrate that our estimator achieves the same selection error rate as existing approaches while using only one half of the A/B testing data.
Large Language Models (LLMs) have become increasingly effective at generating code when task descriptions are clear and precise. Yet, in practice, user-provided task descriptions are often ambiguous, incomplete, or contradictory, leaving critical aspects of the intended program behavior underspecified. In such cases, multiple behaviorally distinct interpretations may satisfy the description equally well, yet semantically differ in ways that matter/affect the user intent. A natural expectation, often assumed by researchers, is that prompt underspecification manifests as incoherence: When asked multiple times, an LLM produces multiple semantically distinct implementations reflecting the ambiguity of the task description. In this paper, we challenge this assumption. We find that LLMs frequently collapse onto a single incorrect interpretation of the task description, consistently generating coherent but behaviorally misaligned code. We term this failure mode detrimental semantic collapse and find that it affects over 10% of tasks in MBPP, 3% in HumanEval, and 32% of LiveCodeBench, all benchmarks assumed to be well-specified. By deliberately injecting underspecification issues in the benchmark prompts, the rate rises to over 5 times, exposing a fundamental blind spot in disambiguation and correctness estimation techniques that rely on incoherence as a proxy for prompt underspecification.
Large language models (LLMs) are increasingly consulted on contested scientific questions, raising the concern that they will sycophantically retreat from established consensus when a user signals doubt -- drifting toward a false balance that treats settled science as one view among several. We test this across three open instruction-tuned models (Llama-3.1-8B, Qwen2.5-7B, Mistral-7B), three consensus-science domains (climate, vaccines, evolution), and single- and multi-turn settings, combining behavioral measurement with linear probing and activation patching. We do not observe sycophantic retreat. Instead, models show three distinct policies under the same skeptical pressure: reactive assertion, where consensus assertion increases rather than decreases (Llama); surface hedging, where tone softens while the position holds (Qwen); and non-response (Mistral). Pairwise judgments confirm the reactive shift is stance, not style (63.6%, p=.007), and a decomposition identifies increased consensus assertion, not false balance, as its driver (beta=+0.042 per dose, p<1e-77). Linear probes localize the divergence to middle layers -- perfect separation in Llama and Qwen versus 72% in Mistral, with non-overlapping confidence intervals -- indicating the non-responsive model does not linearly represent the skepticism signal at all. Crucially, this robustness does not transfer: it attenuates across domains and, in the safety-critical vaccine domain, can reverse, with myth-rebuttal weakening under skeptical pressure. We synthesize these into a four-way taxonomy separating active from accidental robustness, and argue that behavioral evaluation alone cannot distinguish a model that resists skepticism because it understands the signal from one that only appears to resist because it fails to perceive it.
The classical $k$-means clustering cannot be directly used to incomplete data, and existing $k$-means-based clustering for missing data primarily focus on improving the practical accuracy of clustering, whereas most of them lack theoretical guarantees in the asymptotic sense. In this paper, we investigate the statistical properties of $k$-means clustering in the presence of missing data. We first establish the $\sqrt{n}$-excess risk bound and prove the consistency of the estimated cluster centers under general missing mechanisms. For the Missing Completely at Random (MCAR) mechanism, we further derive the $\sqrt{n}$-convergence rate and asymptotic normality of the estimated cluster centers. Moreover, we study in what cases the cluster centers estimated by incomplete data converge to the true cluster centers of original fully observed data, and give a sufficient condition about the missing probability and the separation among true clusters. These results provide a theoretical guarantee for missing-data-$k$-means. Notably, our analysis reveal that under MCAR mechanism, both achieving the $\sqrt{n}$-rate and converging to the true cluster centers require $k$ true centers to be distinct in every dimension, highlighting the significant challenges of application in high-dimensional regimes. Finally, we conduct numerical simulations on synthetic incomplete datasets to support our theoretical analysis results.
Quantum machine learning has recently emerged as a promising paradigm that leverages the expressive power of quantum circuits to address complex learning tasks. In this work, we investigate the applicability of hybrid quantum-classical neural networks to sentiment analysis, a central problem in natural language processing. We focus on a dataset of tweets related to COVID-19, where the textual content is vectorized using TF-IDF and fed into both classical feedforward networks and hybrid architectures incorporating parameterized quantum circuits. Our results show that hybrid models can achieve accuracy comparable to the classical baseline, while exhibiting distinct learning dynamics, especially in terms of validation loss and accuracy, that suggest a richer representational capacity. Moreover, when applying transfer learning to an SMS spam classification task, the hybrid models consistently outperform the classical counterpart, achieving an accuracy increase of 15 percentage points (from 66% to 81%) on the spam class, demonstrating enhanced generalization. These findings highlight the feasibility of employing QML for natural language processing and point toward the potential advantages of hybrid models as quantum hardware continues to advance.
LLM-based agents have shown strong potential for solving complex multi-step tasks, yet existing performance improvements often rely on either scaling to larger backbone models or task-specific fine-tuning. The former incurs substantial computational costs, while the latter typically generalizes poorly across different tasks. Although prompt-based control is training-free and broadly applicable, existing methods still leave input-output dependencies between subtasks implicit in textual trajectories, making verified intermediate results difficult to reuse. To address these limitations, we propose Atomic Task Graph (ATG), a unified control framework for planning and execution. Specifically, ATG maintains an explicit graph to expose dependencies and support reuse. During planning, it recursively decomposes a high-level task into subtasks, forming a sequence of directed acyclic graphs (DAGs) whose evolution can be traced. During execution, the dependencies exposed by ATG allow independent branches to be executed in parallel, thereby improving execution efficiency. When failures are detected, ATG leverages the graph evolution history to localize the error source and repair only the affected region, preserving validated regions unchanged. Experiments show that ATG consistently outperforms strong baselines in success rate and execution efficiency across three interactive benchmarks using only 7B-8B backbones.
Mechanistic interpretability often relies on component-level interventions to discover how a model produces a behavior. This guides attribution, capability knockout, and model pruning downstream to operate by scoring each unit by the effect of ablation in isolation. Such first-order scoring is natural when component importance is additive, but becomes misleading when a transformer self-repairs: after a primary component is removed, a dormant backup can take over, muting the primary's measured effect while the backup itself appears irrelevant on the intact model. We recast this failure as a recovery task, conditional circuit completion, and introduce Conditional Co-Ablation (CoAx), a label-free, output-grounded score that asks how much each remaining unit's ablation effect grows once a primary set has been removed. This conditional growth exposes the second-order interaction that single-unit scores discard. On the GPT-2-small IOI circuit, CoAx raises backup-head recovery from 0.33 to 0.91 ROC-AUC, outperforming all baselines, including self-repair-aware gradient scores (best 0.82); counterfactual patching verifies that the recovered heads causally carry the repair. The same label-free procedure transfers to induction across eight models. Beyond discovery, the recovered backups correct self-repair-masked attribution, identify the components required for capability knockout, and yield repair-aware structured pruning scaling from 124M to 7B. Component importance is therefore not merely an isolated-unit property: in robust circuits, the components that matter can become visible only under the interventions that make them necessary.
Manipulating fast and dynamically moving targets in unstructured 3D environments remains challenging for embodied AI. Existing visual-language-action models and world models struggle with accurate 3D geometry and physically meaningful forecasting. We propose PhysMani, a framework that couples a physics-principled 3D Gaussian world model with a future-aware action policy model. The world model learns a divergence-free Gaussian velocity field via online optimization for fast and physically grounded future dynamics prediction. The policy model integrates the predicted 3D scene future dynamics through a learnable token based cross-attention module. We introduce PhysMani-Bench, a dynamic manipulation benchmark with 16 tasks, and demonstrate a superior success rate over strong baselines in both simulation and real-world robot experiments.
Learning causal models from high-dimensional data is a significant challenge, particularly in real-world settings where violations of core assumptions lead to causal identifiability issues. Although massive amounts of prior knowledge are available, and contain valuable causal information, effectively integrating this knowledge into the causal discovery process remains an open problem. We introduce CausalSTeward (CAST), a novel human-in-the-loop framework for interactively assembling large causal models. CausalSteward is a multi-agent collaborative system that tackles high-dimensional causality through a divide-and-conquer approach where large clusters of variables are iteratively partitioned and then separately analyzed. Our framework fuses prior knowledge with a data-driven approach by using tailored tools such as retrieval augmented generation and conditional independence tests. Finally, we use this work to examine the capabilities and limitations of causal reasoning in multi-agent frameworks, and how the human-in-the-loop can contribute to accurate and trustworthy results.
Long term memory lets LLM agents act as persistent assistants, but user facts change. A useful memory system must know what is true now, what used to be true, and what changed. We study \emph{ghost memory}, a state coordination failure in which old, current, and transition facts coexist in the memory bank, remain mixed during retrieval, and mislead the answer model. We argue that memory systems should be understood and optimized from three levels: bank maintenance, retrieval, and answer time resolution. We propose ATMA, a state aware overlay for existing memory systems. ATMA keeps superseded and transition records in the bank, builds evidence packets for the query's requested state view, and exposes current, historical, and transition labels to QA. We further call for decoupled evaluation of bank, retrieval, and answer level failures, since final QA accuracy can hide where ghost memory occurs. To make this failure measurable, we build LTP (LoCoMo Temporal Plus), a conflict heavy benchmark for ghost memory, and evaluate on LoCoMo for long conversation generalization. On LTP, Graphiti+ATMA improves conflict accuracy by 0.240 absolute over Graphiti. On LoCoMo, Graphiti+ATMA raises temporal F1 from 0.0295 to 0.1705. The gains are host dependent, but they indicate that explicit state roles can reduce memory failures hidden by final QA accuracy.
This work introduces AIriskEval-edu-db2, a new dataset designed to train and evaluate auditors based on LLMs for an explainable pedagogical risk assessment in instructional content for grades K-12. The dataset comprises 1,639 explanations from 170 curated ScienceQA questions, covering science, language arts, and social sciences. For each question, the dataset includes an explanation written by a human teacher alongside 11 explanations generated by LLM-simulated teacher profiles associated with distinct pedagogical risks. We propose a comprehensive risk rubric aligned with established educational standards that covers five complementary dimensions: factual precision, depth and completeness, focus and relevance, student-level appropriateness, and ideological bias. A key contribution is the addition of 785 explanations with structured explainability annotations, including risk localization and risk description. The annotations are produced through a semi-automatic process with expert teacher validation. Finally, we present validation experiments comparing state-of-the-art proprietary models with a lightweight local Llama 3.1 8B model in both the pedagogical risk detection and the explainability assessment. These experiments evaluate whether supervised fine-tuning on AIriskEval-edu-db2 enables a locally deployable model to approach or outperform stronger frontier models while preserving privacy in educational auditing and assessment tasks.
Recent advances in agentic program repair have significantly improved issue resolution by enabling iterative repository exploration. However, existing approaches predominantly rely on sequential, text-based code navigation, which fundamentally limits their ability to reason over large-scale long-horizon repositories with complex and long-range dependencies. As issue-resolution agents traverse repositories through fragmented textual observations, structural information such as module organization, call relationships, and dependency chains must be repeatedly reconstructed across interaction steps, often leading to exploration drift and incomplete localization. We present DUALVIEW, a dual-modal structural scaffolding framework that brings visual reasoning into repository exploration for issue-resolution agents. DUALVIEW represents repository structure through four complementary graph views: Module Coupling Graph (MCG), Function Call Graph (FCG), Class Hierarchy Graph (CHG), and Program Dependence Graph (PDG), and exposes them through a queryable interface with visual and textual responses. Rather than reconstructing repository structure from a sequence of textual observations, agents can directly reason over persistent visual representations of code dependencies, enabling more effective exploration and understanding of long-horizon codebases. We evaluate DUALVIEW on SWE-bench Pro and Verified. Results show that DUALVIEW consistently improves issue-resolution performance across different agent architectures and model families. Further ablation studies demonstrate that the gains arise not only from textual structural information but also from visual externalization of repository dependencies, which better supports long-horizon repository exploration.
This paper presents TUDUM (Türkçe Düşünen Üretken Model), a project pipeline for adapting a Qwen-family 27B thinking model toward Turkish reasoning. The central problem is not only to answer Turkish prompts in Turkish, but to make the explicit reasoning trace itself Turkish. A thinking model may translate a Turkish prompt into an English-centered internal or visible scratchpad, solve the problem mostly in English, and only localize the final answer. TUDUM instead treats the generated <think>...</think> block as a trainable behavior. The pipeline starts from the project base checkpoint unsloth/Qwen3.5-27B, applies supervised fine-tuning (SFT) on 15,991 Turkish reasoning examples using LoRA adapters, and then applies GRPO-family reinforcement learning on a proxy-filtered Turkish mathematics environment. The results are mixed. SFT made the model shorter and more consistently Turkish in its reasoning behavior, with large reductions in average response length and thinking exhaustion, but reduced benchmark accuracy. RL recovered some mathematical performance, especially AIME24 at the best early checkpoint, yet did not uniformly improve all benchmarks and did not exceed the base model on the reported Macro-6 average. The contribution is therefore best framed as a technically honest Turkish-thinking reasoning pipeline and evaluation, not as a claim of state-of-the-art Turkish reasoning. The released step-50 model is publicly available.
Communication systems designed for reliable data reconstruction, rather than task-oriented communication, typically rely on separate source and channel coding and incur high latency under limited spectrum availability and fading channels. To address this, we propose a transmission framework with opportunistic spectrum access, in which the transmitter sends discrete latent representations learned via a vector-quantized variational autoencoder (VQ-VAE) over idle licensed channels using standard digital modulation. The AI-powered receiver is still able to reconstruct task-related information from the heavily compressed data. We develop a cross-layer latency model that accounts for compression, block errors, retransmissions, and stochastic channel access. Results on latency-accuracy trade-offs show that the proposed scheme achieves at least 79- and 3.3-fold latency reductions with only 5.7% and 2.4% drops in classification accuracy compared to benchmarks using conventional source and channel coding. The framework enables low-latency communication and reliable task execution even under limited spectrum availability and challenging channel conditions.
Agentic systems enhance their capabilities by invoking external tools and maintaining persistent memory. However, these external dependencies introduce novel attack surfaces. Recent tool and memory poisoning attacks show that maliciously crafted tool descriptors and poisoned memory can covertly bias agent behavior. These threats reflect a deeper issue: the lack of verifiable continuity in the agent's contextual state for planning and execution. We present ElephantAgent, a protocol that enforces Contextual State Continuity to defend against contextual state poisoning. Inspired by prior state-continuity mechanisms (e.g., Nimble), ElephantAgent extends this protection to the evolving contextual state of agentic systems. We define the contextual state as the bounded, security-critical subset of the agent's entire context (e.g., tool state and memory). Before processing each query, ElephantAgent recomputes the digest of the local contextual state and verifies it against the latest authorized digest. Using replicated trusted hardware, ElephantAgent maintains a linearizable ledger of authorized contextual state transitions and detects out-of-band state tampering. To handle in-band semantic abuse, ElephantAgent additionally provides Historical Traceability, enabling conditional post-hoc audit and recovery to a known-good prior state.
We present Zeus, a unified tuning-free Time Series Foundation Model (TSFM) that delivers superior performance across diverse analysis tasks without any task-specific fine-tuning. Unlike prior studies that primarily focus on zero-shot forecasting but require task-specific tuning for other tasks, Zeus bridges this gap by addressing two fundamental challenges in multi-task generalization. First, to reconcile point-level granularity with long-sequence scalability, Zeus incorporates a multi-scale Transformer featuring point-wise tokenization and a U-shaped hierarchy, effectively balancing fine-grained fidelity with computational efficiency. Second, to accommodate varying inductive biases across different tasks, Zeus introduces Multi-Objective Temporal Masking (MOTM), a unified strategy that supports heterogeneous tasks (e.g., extrapolation, interpolation, and global abstraction) within a single framework. Extensive experiments across five representative tasks demonstrate that Zeus consistently achieves competitive results in tuning-free settings, underscoring its potential as a general-purpose TSFM.
Large language model agents can repair real repository issues, but they often spend large context budgets on whole-file reads, broad searches, and long terminal outputs where useful evidence is mixed with irrelevant code and logs. This paper presents ContextSniper, AntTrail's token-efficient code memory layer for repository-level program repair. As the coding specialization of AntTrail's broader agent memory engine, ContextSniper implements the Sniper feature for precision evidence selection: it retrieves candidate code and runtime evidence, ranks it with hybrid retrieval signals, filters long outputs through an intention-aware context gate, and returns compact evidence packets while preserving recoverable source context outside the prompt. We evaluate ContextSniper on SWE-bench Lite with OpenClaw and Claude Code, using 50 task runs per host-agent condition. ContextSniper reduces total token use by 51.5% and logged cost by 36.4% for OpenClaw, and reduces total token use by 38.9% and estimated cost by 27.3% for Claude Code. Submitted-resolution rates decrease slightly, from 26.0% to 24.0% for OpenClaw and from 32.0% to 30.0% for Claude Code. ContextSniper's pilot testing scripts are open-sourced at https://github.com/Calluking/ContextSniper
Semi-supervised generative adversarial networks (SSL-GANs) can exploit large unlabeled datasets while retaining a classifier in the discriminator, but their training is often unstable. This paper proposes a population-based evolutionary training strategy in which discriminator learning is formulated as a multi-objective optimization problem. Instead of aggregating the supervised and unsupervised components of the SSL objective into a single scalar loss, the method maintains a population of discriminators ranked by Pareto dominance, enabling the exploration of different trade-offs between classification accuracy and real/fake discrimination. This formulation aims to improve both roles of SSL-GANs: learning accurate classifiers and training generators capable of producing realistic samples. We analyze several variants, including an elitist strategy and a mono-objective ablation, to assess the role of multi-objective selection. Experiments on MNIST with limited labels show improved training robustness compared to SSL-GAN and CE-SSL-GAN state-of-the-art baselines, while the elitist variant consistently achieves the highest classification accuracy.
Enterprises increasingly mandate AI coding tools and report large productivity gains, yet longitudinal evidence on how such a mandate unfolds is scarce. In this paper, we present a quantitative case study of a documented enterprise "2x" mandate at a mid-sized, AI-forward company that has been committed to doubling merged pull requests per engineer since mid-2025. In a panel of 802 developers and 196,212 pull requests (January 2024-April 2026), per-capita throughput eventually doubled, reaching 2.09x the pre-mandate baseline in April 2026, among the largest gains reported from a field deployment of AI coding tools to our knowledge. A staggered difference-in-differences design links the within-developer share of this gain to AI adoption and to a further gain that grows with accumulated use, with the mandate acting as a catalyst rather than a direct driver. Because adoption and usage intensity were not randomly assigned, we read this evidence as strongly implicating an adoption-and-use channel rather than as exact causal attribution. The gain is broadly shared across seniority yet concentrated in newer code and not separable across model generations. Adoption also restructured code review around automation: per-reviewer load roughly doubled and automated review overtook human review, while merge and revert rates held steady.
LLM-integrated applications blend natural language prompts with program code, and much of their runtime behavior originates in the prompt layer rather than in the code itself. Existing complexity metrics, however, operate solely at the code level and therefore overlook this behavioral logic entirely. We present HECATE, the first tool designed to assess complexity in both the prompt and code layers of such applications. Central to HECATE is Prompt-as-Specification, a Hoare-logic-inspired formalism that interprets every prompt as a specification of intended behavior. Grounded in 25 complexity dimensions identified across published taxonomies, the tool generates 52 candidate metrics. We assess each metric against 118 components collected from 18 open-source repositories, relying on maintenance activity derived from version history as an empirical proxy for complexity, and discard any metric that loses significance once code size is accounted for. Only ten metrics withstand this test. Seven belong to our newly introduced set; rather than measuring sheer volume, each tallies structurally distinct elements, such as LLM call sites, memory attributes, and prompt templates, an attribute we call structural breadth. Of the three surviving conventional metrics, RFC exhibits a similar breadth-oriented character, while Halstead N and V survive only as a residual effect of size; our top-performing metrics exceed all three. Crucially, the prompt-layer metrics retain significance even when the strongest code-level metric is added as a covariate, establishing prompt complexity as a dimension in its own right. A final validation on 20 components spanning six held-out repositories shows that the two best-performing metrics continue to predict maintenance effort, supporting their generalizability beyond the training set.
Reliable confidence estimates are essential in semantic segmentation, especially in safety-critical settings where overconfident errors can mislead downstream decisions. Yet modern segmentation models often remain miscalibrated. Post-hoc calibration offers a practical way to correct confidence estimates without retraining the segmentation model, but its use in dense prediction raises structural issues that are often overlooked. We study two such issues. First, adding a constant to all logits leaves the softmax probabilities unchanged, but several standard calibrators can still depend on this arbitrary offset. As a result, two logit representations encoding the same predictive distribution may yield different calibrated probabilities. We define translation-invariant (TI) calibrators as those whose outputs are unchanged under such shifts, characterize which common calibrators satisfy this property, and construct TI counterparts of shift-sensitive calibrators to isolate the effect of removing representation dependence. Second, post-hoc calibration is typically fitted by minimizing a likelihood-based objective, whereas segmentation models are trained with task-specific metrics such as Dice. This mismatch can cause calibration to alter class orderings and degrade the deployed segmentation map. We study decision-preserving calibration under argmax- and order-preservation constraints. Since enforcing these constraints collapses affine softmax calibrators to temperature scaling, we introduce class-conditional affine calibrators that can be made argmax- or order-preserving while retaining greater expressivity, allowing us to quantify the calibration-segmentation trade-off induced by decision preservation. Across natural-image and medical segmentation benchmarks, and under corruption-based covariate shift, matched comparisons show that TI variants generally improve calibration metrics, while decision-preserving variants prevent segmentation degradation and retain strong calibration performance. These results provide practical design principles for well-defined post-hoc calibration pipelines in semantic segmentation.
Effective brain disease diagnosis requires the synergy of brain connectivity patterns and high-level semantic knowledge. Existing methods, however, largely treat semantics from large language models (LLMs) as auxiliary features or supervision, limiting their direct role in decision-making and constraining classification stability and robustness. To overcome this, we propose a semantic-aligned brain network framework that actively integrates LLM-derived semantics into the prediction process. Specifically, ROI-level semantics are first incorporated via global self-attention to enrich node representations and provide whole-brain context. Multi-scale hypergraphs are then constructed to explicitly model functional subnetworks and multi-ROI interactions, addressing the locality limitations of traditional GNNs and capturing high-order dependencies. Finally, a decision-level semantic alignment mechanism selectively injects patient-specific textual embeddings into graph representations, enabling semantics to directly guide predictions without perturbing the underlying network structure. Experiments on public brain network datasets ABIDE and ADHD-200 demonstrate state-of-the-art performance, enhanced stability, and improved interpretability, particularly in small-sample settings.
Dependency length minimization (DLM) is a well-documented processing universal, but previous studies report a single mean dependency distance (MDD) per language, obscuring variation across syntactic relation types. We analyze 122 languages in UD and SUD (version 2.17), showing that DLM operates on two distinct levels. Grammar-driven optimization targets functional dependencies (det, case, aux), which are universally short (mean 1.71, $σ$ = 0.33) and invariant across typologically diverse languages. Processing-driven optimization operates on lexical dependencies (nsubj, obj, obl), which are longer (mean 2.87), highly variable ($σ$ = 0.63), and constrained by word-order typology. This asymmetry holds in SUD despite reversed head direction (r = 0.92). We conclude that ''the grammar does the work'' of minimization by scaffolding sentences with local functional attachments, leaving processing pressures to determine the ordering of lexical heads.
We introduce Rank-Then-Act (RTA), a framework for learning control policies from expert video demonstrations without environment rewards. RTA trains a Vision-Language Model (VLM) offline as a progress-based ordinal scorer, using a Group Relative Policy Optimization (GRPO) objective over shuffled frame sequences, which forces the model to recover temporal ordering from visual semantics rather than trivial time cues. Importantly, instead of using the scorer directly as a scalar reward model, we propose a correlation-based reward function for reinforcement learning: at each interaction window, we compute the Spearman rank correlation between predicted progress rankings and true temporal indices, yielding a bounded, scale-invariant learning signal. This design decouples reward learning from absolute calibration and enables stable transfer across tasks and environments. We evaluate RTA on discrete control benchmarks (PyBoy: Catrap, Kirby) and continuous control tasks (PointMaze, MetaWorld). RTA consistently matches or outperforms prior video-based reward learning methods and rank-based baselines, while demonstrating strong cross-task reuse of a single pretrained progress scorer. Our results suggest that correlation-structured supervision over video-derived ordinal signals is sufficient for policy learning, offering a scalable alternative to explicit reward design.
We study ridge-regularized log-density-ratio estimation in the Gaussian location model with a common covariance matrix. By affine invariance, the model is written as q $\sim$ N(0, I), p $\sim$ N($Δ$, I), with linear features, where $Δ$ is a mean vector. The variational estimator is the empirical Kullback-Leibler (KL) log-normalized fit with a squared L2-penalty on its nonconstant coefficient, and the spectral estimator recently introduced in [1] replaces a single variational problem by a continuum of ridge-regularized least-squares problems. We derive high-dimensional deterministic asymptotic equivalents when the numbers of observations and dimension tend to infinity with fixed ratios. The regularized variational limit is characterized by a scalar entropy minimization problem derived from the convex-Gaussian-min-max theorem (CGMT), while the regularized spectral limit follows from deterministic equivalents for resolvents of weighted sums of two independent Gaussian sample covariance matrices. We use these formulas to compare population risks, with experiments focused on fixed-signal aspect-ratio sweeps and optimized regularization. Our conclusion is that with many observations, under the criteria and asymptotic regimes analyzed here, the well-specified variational estimator has the smaller risk, while with fewer observations, the spectral estimator is favored because its covariance-based construction has lower variance. We also study how a nuclear penalty can be used and partially analyzed to perform feature learning.
Speculative decoding accelerates autoregressive generation by drafting a block of tokens that the target model verifies left-to-right, committing only the longest accepted prefix. Block (DLM-style) drafters predict the whole block in parallel, which is fast but trained with a full-block cross-entropy that supervises every position against the gold continuation -- even though inference discards every token after the first rejection. Recent acceptance-aware objectives patch this by reweighting the full-block loss; we instead use teacher-forced learning as a motivation for how supervision should concentrate on the accepted prefix. A mask-only block drafter has no input-side channel for gold-prefix conditioning, so AUF approximates that prefix-sensitive supervision on the loss side by keeping the cross-entropy support only through the drafter's first predicted failure. AUF is a single, detached change to the CE support -- no auxiliary objective, no verifier rollouts, and no change to the inference pipeline or the exactness contract. Within fixed drafter backbones and serving settings on Qwen3-8B, AUF raises the DFlash drafter's average emitted length $τ$, averaged over six benchmarks, from 2.40 to 2.61, with a gain on every benchmark, and transfers to Domino's two-branch head (2.56 to 2.68). Two findings sharpen the picture: the decay-only baseline reaches higher token accuracy on the shared block mask yet decodes worse, and on DFlash, once AUF truncates the support, the standard exponential position-decay weighting becomes empirically inert.
The security of open source applications benefits considerably from the possibility of rebuilding their source and verifying the output. F-Droid, a prominent distribution for open source Android applications, systematically rebuilds them from source and tests their bitwise reproducibility at app publishing time. However, F-Droid offers no guarantee that app reproducibility will continue to hold in the future. As software ecosystems evolve, reproducibility may degrade, with potential negative consequences for software preservation and security. We present the first empirical study of build reproducibility in the F-Droid app ecosystem. Analyzing historical reproducibility logs, we find that the overall bitwise reproducibility rate has been steadily increasing over time (as new versions of apps are published). We then evaluate how reproducibility holds in time for fixed app versions, by attempting to rebuild 18 904 app versions that F-Droid had previously confirmed bitwise reproducible, published between September 2018 and February 2026, achieving an 83% rebuild success rate, and identify missing dependencies as the dominant cause of failure, accounting for 76% of non-rebuildable cases. Among successfully rebuilt apps, 94% are also bitwise reproducible-i.e., they still yield bitwise identical artifacts upon rebuild. Together, these results show that while bitwise reproducibility largely holds for apps that can be rebuilt, rebuildability itself is highly sensitive to temporal decay.
Code is the medium through which large language models generate structured artifacts: charts, scientific figures, vector graphics, CAD models, 3D scenes, and hardware designs are all produced by writing programs. In this regime single pass inference is brittle, because the compiler, renderer, or simulator that decides whether the artifact exists is invisible to the model. We present PairCoder, which grounds review in the toolchain and realizes it as two agent pair programming: a Driver agent writes the program, a Navigator agent reviews it against verification evidence (diagnostics, execution results, and renderings of the current artifact beside the target), and the two switch roles when errors persist. Across 17 public benchmarks and seven models from three vendors, PairCoder improves essentially every benchmark whose artifact is verifiable, on full official metric suites rather than execution alone (for example, Blender scene executability 0.20 to 0.78; TikZ compile rate up 10 to 30 points on every model), at 2.9 to 9.2 times single model cost (about 7 times overall). The improvements concentrate where the toolchain provides an informative oracle and the baseline leaves headroom, and the method ties or mildly regresses where the oracle is weak; we frame pair programming as a reliable recipe for verified code driven generation.
Value functions are an essential component in actor-critic based deep reinforcement learning (RL). Conventionally, these functions are trained as a regression task by minimising the mean squared error (MSE) relative to bootstrapped target values. Meanwhile, in distributional RL, a distribution of returns is modelled based on the distributional Bellman operator. This work investigates the Gaussian Histogram Loss (HL-Gauss), a recent approach that reframes value estimation as classification by encoding each scalar Bellman target as a Gaussian-smoothed categorical target. Despite its potential, applying histogram-based losses to RL presents inherent challenges, most notably the requirement to pre-define a fixed support interval, which is often complicated by the non-stationary and stochastic nature of target values typically found in RL tasks. In this work, we propose an approach that dynamically learns the lower and upper bounds of the support instead of assigning them beforehand. We derive an objective that jointly learns these bounds whilst learning the categorical representation of the scalar values, and we show that this objective forms an upper bound on the mean-squared Bellman error. Our theoretical analysis further shows that this bound is tighter than that of non-learned supports of HL-Gauss. Empirically, the proposed objective enables stable adaptation of the support interval and matches HL-Gauss-based actor-critic algorithms on most continuous-control tasks whilst improving on a subset, without requiring a pre-specified support interval.
Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal understanding, yet their enormous parameter scale and cross-modal computation incur substantial memory and latency overhead, severely limiting real-world deployment on resource-constrained devices. Binarization offers an attractive solution by drastically reducing storage and computational costs. However, existing binarization methods neglect the varying importance of weights across different layers and modalities. This causes parameters irrelevant to downstream tasks to be unnecessarily retained, whereas modality-critical weights may not be adequately optimized, resulting in significant performance degradation. To address these challenges, we develop a novel \underline{S}ignificance-\underline{A}ware \underline{B}inarization for \underline{L}arge \underline{V}ision-\underline{L}anguage \underline{M}odels (SAB-LVLM). Specifically, after constructing Hessian matrices for textual and visual inputs, we propose a spatial significance map to distinguish full-precision weights activated under a single modality from those activated across modalities. We then devise a modality-guided integration strategy to obtain the significance-aware binarization map, which measures weight significance across layers and modalities. Subsequently, this binarization map is incorporated into the binarization objective as an error reweighting term, and binarization fitting is performed through an alternating significance-weighted update scheme. Extensive experiments illustrate the superiority of our SAB-LVLM over existing binary PTQ methods under an approximately 1-bit compression constraint. Our code is accessible at https://github.com/LyuQi127/SAB_LVLM.
Skills are becoming a reusable operational layer for LLM agents, encoding SOPs, domain rules, tool workflows, scripts, and validation routines. In realistic skill repositories, overlapping skills make reliable skill-use difficult. Final verifier success is too coarse for both evaluation and training, since an agent may pass through trial and error while selecting distractor skills, skipping required steps, composing workflows incorrectly or omitting final checks. We introduce SkillCoach, a self-evolving rubric framework for evaluating and enhancing agentic skill-use. SkillCoach derives skill-grounded process rubrics from real rollouts and evaluates trajectories along four dimensions: skill selection, skill following, skill composition, and skill-grounded reflection. It keeps the external verifier as a separate outcome signal, allowing process quality to be distinguished from accidental task success. The evolved rubrics further serve as process supervision for selecting high-quality training trajectories. Experiments show that evolved rubrics substantially improve evaluation quality, expose failures hidden by final accuracy, and provide stronger supervision signals than outcome-only filtering for enhancing agentic skill-use.
Camouflaged Object Detection (COD) aims to locate and segment objects that blend into their surroundings, presenting challenges due to weak edge cues and ill-defined boundaries. Traditional COD models rely on hand-designed architectures and multi-scale feature fusion, which are often guided by intuition rather than systematic search. This paper introduces CamoNAS, a frequency-aware multi-resolution Neural Architecture Search (NAS) framework for COD. CamoNAS automatically searches both cell-level operations and network-level downsampling paths, forming a hierarchical search space tailored to detect camouflaged objects. Additionally, it adopts an RGB frequency dual-stream architecture, where a learnable wavelet transform complements the RGB spatial stream. CamoNAS achieves state-of-the-art performance on four COD benchmarks (CAMO, COD10K, NC4K, CHAMELEON), highlighting the effectiveness of NAS for COD. Our code is available at https://github.com/rendaweiSIMIT/CamoNAS.
The use of LLMs in software development has become increasingly widespread on tasks such as code generation and summarization. Reports from large technology companies showed that around 20% to 30% of their code are generated by LLMs. However, there remains skepticism about the practical usage of LLM-generated code and comments, such as concerns on more time for debugging the generated code and the unnaturalness of the generated comments. In this paper, we study the code and comments detected as likely to be generated by LLMs and their characteristics, the differences between company- and community-maintained repositories, and how likely bugs are associated with LLM-generated code. We conduct extensive experiments on active company- and community-maintained repositories from 2021 to 2025 using various tools and techniques that detect code and comments generated by LLMs. Based on our detector-based proxy analysis, the results suggest that code detected as likely to be generated by LLMs decreased over time and appeared frequently in test cases, while that of comments remains relatively stable. Proxy results further suggest that code detected as likely to be generated by LLMs shows substantial intra-repository code clones, whereas comments exhibit a relatively low proportion of grammatically correct sentences. In addition, the company-maintained repositories show a higher percentage of code and comments detected as likely to be generated by LLMs, and only a small percentage of the human-labelled bugs are detected as being likely associated with LLM-generated code.
Safety training for large language models (LLMs) is conducted predominantly in English, leaving uncertain how well safety mechanisms generalize to low-resource languages and mixed-language code-switching. We show that this creates an epistemic gap in which models confidently generate harmful responses for inputs that fall outside the distribution of their safety training. To study this phenomenon, we introduce STEER (Safety Targeted Embedding Exploit via Refinement), a gradient-guided attack that identifies words contributing most strongly to the model's refusal behavior and iteratively translates them into low-resource languages to suppress refusal while preserving harmful intent. Across six open-source 8B-parameter models, STEER achieves attack success rates of up to 93.0% on JailbreakBench and 96.7% on AdvBench, outperforming random code-switching and Greedy Coordinate Gradient (GCG). The resulting prompts also transfer to GPT-4o-mini, achieving a 35.5% attack success rate without requiring access to the target model, suggesting that the underlying weakness is not specific to a single architecture. These findings demonstrate that safety mechanisms aligned primarily on English cannot be assumed to generalize across multilingual inputs. We argue that improving multilingual safety requires broader coverage during alignment and mechanisms that explicitly detect and abstain on out-of-distribution inputs.
In LLM-assisted software development, coding is often iterative. We study regression accumulation in multi-turn LLM programming conversations, where later code suggestions may break requirements introduced in earlier turns. Reliability therefore depends not only on satisfying the current request, but also on preserving previously satisfied behavior. We construct 542 tasks from HumanEval+ and MBPP+ and extend each task into an 8-turn requirement-evolution chain. We evaluate six LLMs on 26,016 turn instances (542 x 6 x 8). At each turn, we test whether the current code still passes earlier benchmark tests. We also analyze 384 failure cases from the failure population and build a taxonomy of multi-turn regression bugs through independent four-annotator labeling. Our results show that regression accumulation appears across all six models: 40% to 73% of tasks lose previously correct behavior over the full conversation. Final-turn quality is lower than initial-turn quality across models, especially when later turns add input validation or broader input types. Manual analysis shows that Cross-Turn Conflict, where later code conflicts with earlier requirements, is the main failure class. We further find that Verification Gate, which checks new code against prior tests and triggers rollback and retry, is the only strategy that consistently improves all models, raising final-turn quality from 75.8% to 87.9% on DeepSeek-V3 and from 31.6% to 47.3% on Llama-3.1-8B. These findings suggest that strong single-turn performance can overestimate reliability in multi-turn coding conversations. Future evaluation and tool design should test whether later code suggestions preserve earlier requirements and should include Verification Gate mechanisms.
Can a platform tell, before deployment, whether an open-weight checkpoint has had its refusal mechanism stripped? Runtime guards cannot: they score generations, not the artifact. We combine two cheap internal signals, a reference-anchored activation refusal-gap and a weight-recovery energy of the base-to-candidate weight difference, into a threshold-free checkpoint audit. The two are negatively correlated and label-complementary: the gap supplies refusal-specificity and the weight energy supplies recall. On a 273-checkpoint registry spanning Qwen, DeepSeek-distilled Qwen, Llama, and Gemma, their z-sum separates 57 public abliterations from 37 benign fine-tunes, merges, and instruction-tunes at AUROC 0.95, significantly above either signal alone (0.84, 0.90), and a Youden-calibrated threshold transfers to held-out families at balanced accuracy 0.89 (FPR 0.11), missing only 4 of 57. We then map two failures, in order of severity: a spoofed reference evades both axes with no training (ΔW=0, \r{ho}=1 by construction), and a white-box owner trains a checkpoint past the threshold while it stays guard-unsafe and coherent. The audit is effective triage, not tamper-proofing: it presumes an attested reference, and its claims are bounded by the registry we evaluate it on.
Retrieval-Augmented Generation (RAG) systems use the question-answering capabilities of Large Language Models (LLMs) to access information outside their parameters. We evaluate if cluster-based semantic chunking improves retrieval and answer quality compared to fixed-size and recursive chunking evaluating on long, structured academic theses using the Retrieval Augmented Generation Assessment (RAGAs) framework. RAGAs based faithfulness shows limited reliability in this setup. Performance on fixed versus document specific questions varied substantially, likely related to the formatting of documents and preprocessing. Under the tested configuration, cluster-based chunking did not outperform simpler strategies.
Software-intensive organizations need effective ways to identify where maintenance and refactoring efforts will yield the greatest practical benefit. Although software analytics such as code health, hotspots, and coupling provide valuable signals, they do not always capture the experienced burden of change that slows software evolution in practice. This paper presents a multi-case industrial study of technical debt friction as a prioritization-oriented concept for identifying where technical debt most strongly affects maintenance and evolution. We investigate how practitioners interpret the concept, whether friction-related analysis aligns with perceived maintenance pain points and refactoring needs, and what broader maintenance and evolution insights friction can provide beyond individual refactoring candidates. To this end, we conducted structured walkthrough sessions with practitioners across multiple industrial cases using analysis artifacts including code health, hotspots, coupling, refactoring targets, and socio-technical views. Our findings show that practitioners generally considered technical debt friction useful for reasoning about maintenance burden, especially when interpreted together with complementary technical and socio-technical views. At the file level, friction often aligned with known problematic areas and, in several cases, with files that later received maintenance attention, although its practical relevance depended strongly on context. In addition, our exploratory project-level analysis suggests that friction distributions may reveal broader maintenance and evolution patterns. These results indicate that technical debt friction is promising as a decision-support concept, but most effective when used with contextual knowledge and supporting evidence.
Musical performance involves executing a set of high-level musical instructions, yet recovering those instructions from the performance is a challenging inverse problem. We present Decomposer, a post-training framework for symbolic music decompilation: the task of recovering executable, editable music programs from symbolic music. We instantiate the task as MIDI-to-Strudel decompilation, where the model takes symbolic MIDI as input and produces a program in Strudel, a music programming language, that reconstructs the input when executed. The task poses two challenges: Strudel is a low-resource language with little naturally paired MIDI-code data, and optimizing faithful reconstruction of MIDI alone can collapse to unreadable note-by-note transliteration. We address these challenges in two stages. First, we construct Strudel-Synth, a synthetic corpus of paired Strudel programs and rendered MIDI, and use it for supervised fine-tuning. Second, we refine the model with reinforcement learning on unpaired MIDI, optimizing rewards for both MIDI reconstruction faithfulness and code readability. Our evaluation across synthetic and real-world MIDI benchmarks shows that Decomposer achieves substantially higher MIDI reconstruction faithfulness than closed-source LLMs while producing more readable and diverse code than the heuristic converter.
Domain agents often face noisy business data, uncertain post-training gains, offline/application mismatch, and adapter-release risk. This paper presents CLAP (Closed-Loop Agent Post-training), a closed-loop method that converts business data into structured SFT samples, decision-preference samples, holdout sets, risk diagnostics, and release-gate records. CLAP combines data validation, target/evidence normalization, reward/KL diagnosis, offline gates, and application-chain replay to decide whether an adapter is suitable for the target application chain. On five anonymized manufacturing-scenario batches, QLoRA-style LoRA-SFT yields modest average gains: overall score increases by 0.0098, pass rate by 0.0240, and evidence accuracy by 0.0280, while hallucination and wrong facts decrease. Yet only 3 of 5 batches improve, some batches regress, and GRPO exposes high KL risks. Application-chain replay further shows that RAG is necessary for factual extraction; under the same 3B backbone and 100 replay cases, an application-RAG-oriented LoRA-SFT adapter improves value, core fields, and answer-evidence doc/page matching over base+RAG, but increases latency. These results support managing domain-agent post-training through an integrated data-training-evaluation-release loop rather than relying on training completion or a single offline score.
This paper showcases a memory-efficient training stack for Mixture-of-Experts (MoE) models. It is a training paradigm that combines and specializes various existing and novel parallelism techniques at different layers and stages of the Mixture-of-Experts (MoE) model training pipeline. It leverages these techniques to achieve maximal efficiency given the physical constraints of CPU, CPU memory, GPU HBM memory, and the CPU-GPU, GPU-GPU, and node-node communication bandwidth of the GPU cluster. It also contains a novel strategy for the optimizer step to achieve high throughput and memory efficiency, enabling practitioners to conduct lossless pre-training/fine-tuning of trillion-parameter scale models, at a million context length, with just under 12 8x H200 GPU nodes, with state-of-the-art throughput and memory efficiency. In our experiments, MoP delivers 4.7x--8.2x higher per-GPU throughput than a strongly-tuned FSDP2 baseline (with the gap widening at larger scale) and sustains training at context lengths up to 1M tokens, where the baseline runs out of memory beyond 64--128K.
Software defect prediction (SDP) classifiers produce probabilities used for inspection prioritization, threshold tuning, and risk communication. Probability-based uncertainty quantification (UQ) characterizes prediction confidence, but whether common UQ metrics reliably indicate performance and calibration remains unclear. We conducted a large-scale empirical study of probability-based UQ for SDP. We evaluated five UQ metrics, six performance metrics, and three calibration metrics for 16 representative classifiers. We analyzed these relationships under two prediction settings: within-project defect prediction (WPDP), using 36 benchmark datasets, and cross-project defect prediction (CPDP), using 32 feature-compatible datasets. Results showed that UQ was highly context-dependent. Under WPDP, UQ correlated more consistently with false positive rate and AUC than with MCC, F1 score, and other metrics; these correlations also varied across classifier categories and dataset collections. Performance and calibration were related but not interchangeable; classifiers with strong discrimination could still exhibit large calibration error. Under CPDP, several UQ-performance and UQ-calibration correlations weakened or reversed, indicating that uncertainty signals do not reliably transfer across projects. Thus, UQ should be evaluated against specific performance objectives. Calibration should be assessed independently using multiple metrics. Transferred probabilities should be revalidated before guiding quality-assurance decisions.
Fault trees are a widely used as effective risk models for complex systems, answering the question "what can go wrong?", especially through minimal cut set analysis. We study fault trees from the perspective of Halpern & Pearl's theory of actual causality. This allows us to use fault trees to answer the question "why has it gone wrong?", which is fundamental to failure diagnostics. We give a complete classification of each of the different notions of actual causality in terms of the fault tree's graph structure and logical structure, and show how minimal cut sets give rise to actual causes.
Counterfactual explanations (CEs) for multivariate time-series classifiers are often difficult to interpret in domains where experts reason in terms of semantic feature groups rather than individual channels. In rehabilitation movement analysis with multi-sensor inertial measurement units (IMUs), clinicians interpret motion through muscle-group and joint-segment abstractions; yet, most existing counterfactual methods operate at the channel level, producing scattered and biomechanically incoherent explanations. We propose a two-stage framework for group-based counterfactual generation in high-dimensional IMU data. We first show that Shapley-Adaptive (SA) group ranking preserves counterfactual validity but fails to enforce group-level sparsity, motivating the need for explicit group selection. We then introduce Learnable Gate (LG) methods, which incorporate trainable per-group relevance gates jointly optimized with perturbation masks. Experiments on the KneE-PAD rehabilitation dataset demonstrate that LG substantially improves modality-group sparsity compared to the channel-level M-CELS baseline while maintaining or improving validity, temporal smoothness, and generation efficiency. Exercise-specific analyses further show that group-structured counterfactuals yield concise, muscle-level corrective guidance aligned with clinical reasoning. Overall, the proposed framework enhances interpretability without sacrificing counterfactual quality, enabling more actionable explanations for rehabilitation movement analysis.
The global development of Library and Information Science (LIS) is influenced by various factors such as the economy, society, culture, discipline, tradition, and more. Consequently, the research methods of LIS vary greatly among countries. To better understand these differences, we conducted a study of 5,281 research papers from 81 countries published in internationally representative journals over the past thirty years. We manually annotated the research methods used in some articles through content analysis, and subsequently developed and trained a deep learning model for automatic classification of research methods. Using this method, we conducted a comparative analysis of the usage of research methods in different countries. Our findings reveal that there are differences in the research methods used across countries, with each country having its unique research profile and distribution of research methods. Even when investigating the same topic, research methods can differ between countries. Our study also uncovers that there are differences between the national and international distribution of research methods, these differences have decreased over the past 30 years. By highlighting the characteristics of discipline development in various countries from the perspective of research methods, our study can help guide discipline development at the national level. This study provides insights into the usage trends of research methods across different countries and highlights the unique characteristics of discipline development in each country. This information can be valuable in promoting collaboration and understanding between countries and in guiding discipline development at the national level.
Long-context inference is increasingly common in large language model (LLM) serving, driven by retrieval-augmented generation and agentic systems. In disaggregated inference, these workloads require transferring large Key-Value (KV) caches across the network, where decoding cannot begin until the transfer completes. Recent KV quantization techniques reduce data volume and alleviate this bottleneck, but existing schemes fail to achieve both low network-exposed latency and high inference accuracy. We challenge the assumption that the KV cache is an indivisible unit that must be fully received before use. We leverage the observation that different bits in the KV cache contribute unequally to attention computation and inference precision: the most significant bits capture the coarse structure of attention and the least significant bits refine precision. This property enables partial use of the KV cache during decoding. We present Lynx, a system that enables progressive, split-stream KV transfer by partitioning the KV cache into a high-priority Anchor stream carrying the most significant bits and a low-priority Residual stream carrying remaining precision. Decoding begins upon receipt of the Anchor stream and proceeds speculatively while the Residual stream is transferred concurrently, followed by verification that ensures equivalence to higher-precision decoding. Across multiple models and serving workloads, Lynx achieves Time-to-First-Token (TTFT) comparable to aggressive 4-bit KV quantization, while matching the accuracy of high-precision (BF16) inference, improving TTFT over standard 8-bit KV quantization by up to $1.43\times$ and improving accuracy over state-of-the-art by up to $5.1\%$.
Reliable reward and preference signals are critical for evaluating and optimizing large language models on open-ended tasks. Rubric-based judges offer a transparent way to decompose such judgments into explicit evaluation criteria, but existing annotation-free rubric generators typically rely on a single generic evaluator. As a result, they may overlook important dimensions of human preference, a failure mode we term dimensional blind spots. To address this limitation, we propose Multi-Role Rubric Generation (MRRG), a training-free and reference-free framework that elicits evaluation criteria from multiple complementary roles and consolidates them into an auditable rubric-based scorer. This scorer can be used both to validate pairwise preferences and to provide rewards for GRPO-style Reinforcement Learning with Verifiable Rewards (RLVR). Experiments on preference validation benchmarks show that MRRG consistently outperforms single-role rubric generation baselines across multiple backbone models. Further RLVR experiments demonstrate that MRRG yields a stronger reward signal for improving open-ended generation.
Large language models (LLMs) are increasingly proposed for aviation business operations, from documentation and training generation to customer facing assistants. General purpose benchmarks do not measure whether a model reasons safely and correctly about aviation specific operational knowledge, and the high stakes, regulated nature of the domain makes that gap consequential. We present Pre-Flight, an open source benchmark of 300 multiple choice questions drawn from international standards and airport ground operations material, covering international airport ground operations, ICAO and US FAA regulations, aviation general knowledge and complex operational scenarios. Questions were authored and reviewed by practitioners with experience in air traffic management, ground operations and commercial flying. We evaluate a range of contemporary commercial and open weight models using the Inspect evaluation framework, scoring by accuracy under a standard multiple choice protocol, and we maintain the leaderboard on a rolling basis as new models are released. Against an informal expert reference of around 95%, obtained from a low sample quiz of aviation professionals at a conference, even the strongest model evaluated (released in 2026) reaches 82.7%, having improved only gradually from roughly 75% in early 2025. A substantial and persistent gap below expert level reliability therefore remains. We release the dataset, the evaluation harness and the results, and the benchmark is available within the community evaluations package distributed with inspect_evals. We argue that domain specific evaluation of this kind is a necessary precondition for responsible deployment of generative AI in non safety critical aviation operations.
Research in the social sciences has shown that there are gender differences in the selection of research methods, with women often opting for qualitative methods while men prefer quantitative methods. However, it is important to consider that research methods are generally chosen based on the research topic. To figure out the influence of gender on research method selection, a study was conducted in the field of Library and Information Science, using a more fine-grained method classification system and an automatic classification model called CogFT, which is based on full-text cognition. The findings showed that women tend to use Interview while men prefer Theoretical approach, across a range of topics. The study offers insights into the specific research design processes that contribute to gender differences in method selection and suggests ways to promoting gender inclusivity and equality in academia by considering research method use and guidance.
Crowdsourced fact-checking systems have been adopted by major social media companies such as X, Meta, TikTok and Google with the aim of combating misleading information at scale without relying on centralized editorial control. These systems have been developed around a common underlying concept: a bridging mechanism that identifies notes flagging misleading information when they receive support from people with different perspectives rather than simple majority support. To our knowledge the only publicly disclosed bridging algorithms deployed for fact-checking are based on matrix factorization, as deployed by both X and Meta, augmented with additional components addressing abuse, targeted manipulation, and contributor brigades. This work examines the core matrix factorization portion of these systems, presenting theoretical and empirical evaluations of the degree to which coordinated users could vote strategically by leveraging the latent representations to fabricate the appearance of synthetic consensus within the bridging mechanism. Using historic production data, we find that up to 10.7% of lower quality notes could be manipulated above consensus thresholds using less than 10 ratings. We complement these findings with a theoretical analysis, revealing counterintuitively that rating a note as "Not Helpful" can increase its helpfulness score, as well as a cost model quantifying manipulation effort. We have developed and deployed mitigations within X's Community Notes algorithm to address synthetic consensus.
Packet loss concealment (PLC) reconstructs audio packets that are missing at the receiver, usually with a trained model whose parameters remain fixed at deployment time. This treats the PLC model as static, even though each call or recording exposes signal-specific information through the packets that did arrive. We present TTT-PLC, a self-supervised test-time tuning framework that adapts existing PLC models using only those received packets. The method creates supervision by synthetically masking portions of the available signal, training the model to conceal them with its native PLC objective, and then using the adapted model to reconstruct the true packet losses. No clean reference signal, external adaptation data, or architectural modification is required. We study TTT-PLC in two deployment settings. In the non-causal setting, the received file is available before reconstruction, allowing repeated self-supervised adaptation passes and providing a per-file adaptation ceiling. In the causal setting, audio is streamed without revising emitted samples; adaptation is performed only on completed past blocks, and updated parameters affect only future audio. We instantiate the framework on two public PLC backbones, FRN, a recurrent full-band speech PLC model, and PARCnet, a hybrid autoregressive-neural model for networked music. Across these settings, the results show that pretrained PLC systems do not need to be treated as fixed at inference time, the still-observed portions of a lossy signal can provide an effective training signal for improving concealment on that same signal.
The Koopman operator has gained considerable attention due to its ability to provide a global linear representation of highly complex dynamical systems. The operator describes nonlinear dynamics in a linear way through the lens of real- or complex-valued observable functions. Recently proposed data-driven techniques, like extended dynamic mode decomposition (EDMD), its kernelized variant, and machine-learning methods, can be used to generate finite-dimensional approximations accompanied by finite-data error bounds. In this tutorial paper, we provide a concise introduction into Koopman operator theory and its use in systems and control. A particular focus is put on data-driven surrogate models, their extension to systems with inputs, and controller design using Koopman operator theory. Moreover, we demonstrate the key techniques, i.e., EDMD and Koopman MPC. To this end, we provide simulation studies including source code on GitHub to enable the interested reader to experience the Koopman operator in systems and control step by step.
Traditional Chinese Medicine (TCM) diagnosis, particularly through tongue inspection, faces persistent challenges in subjectivity and reproducibility. The application of multimodal artificial intelligence to TCM clinical tasks, such as syndrome differentiation and prescription generation, is significantly hampered by the semantic gap between visual tongue features and textual reasoning, as well as the lack of large-scale, standardized datasets. To address these challenges, we introduce MMIR-TCM, a novel framework that emulates the diagnostic process of TCM experts by integrating multimodal large language model(MLLM) with memory-augmented segmentation and retrieval-augmented generation (RAG). Employing a three-stage architecture, MMIR-TCM integrates a training-free Memory-SAM module for robust tongue extraction, a fine-tuned Qwen3-VL model for structured tongue diagnosis generation, and a Qwen3-based RAG component for evidence-grounded clinical decision support generation. The framework was developed and validated using MedTCM, a new large-scale multimodal dataset that we introduce specifically for advanced TCM research. To properly evaluate our framework's clinical accuracy, which existing metrics fail to capture, we also developed TDEU, a domain-specific evaluation metric incorporating semantic understanding and diagnostic importance. Our comprehensive experiments demonstrate that MMIR-TCM significantly outperforms leading models, including GPT-4o and Gemini 2.5 Flash.
Evaluation benchmarks are essential for assessing vision-language models (VLMs), but most multimodal benchmarks are static, making them vulnerable to temporal staleness, data contamination, and costly maintenance. We present MMBench-Live, a continuously evolving multimodal benchmark built by a multi-agent-driven automated pipeline. Our framework treats benchmark evolution as task-guided dataset construction, integrating structured benchmark specification, feedback-controlled real-time data acquisition, and verifiable QA generation with executable reasoning. To maintain cross-version comparability, we introduce a distribution-consistent update strategy that extracts task-related visual patterns from the original benchmark to guide data collection and filtering. Instantiated from MMBench, MMBench-Live contains 5.9K newly generated evaluation instances with a high answer correctness rate, while each update costs about USD 30 and takes 1-2 hours. Extensive evaluations show that MMBench-Live preserves stable model rankings, maintains semantic alignment with the original benchmark, and exhibits weaker contamination-related memorization signals, suggesting a practical and scalable paradigm for sustainable multimodal benchmark evolution. The project is available at https://github.com/PRIS-CV/MMBench-Live.
Open-source projects depend on a steady inflow of newcomers. A growing concern is that AI coding agents (tools such as Cursor and Claude Code that write code from natural-language instructions) will crowd them out, by absorbing the simple tasks that beginners start with and by making code harder to read. We give this concern a causal answer. Using GitHub code search we identify 1,888 projects that adopted an agent, signaled by their first commit of a configuration file. We apply difference-in-differences against matched non-adopting controls, restricting the main analysis to the 603 adopters with a genuine pre-adoption period. We find no evidence of crowding-out: across estimators newcomer inflow shows no significant decline after adoption (point estimates run from a small increase to, under the most conservative trend specification, a slight and insignificant dip), onboarding and retention are unchanged, and a sparse, correlational beginner-task measure (good-first-issue labels, which we cannot test for parallel trends) shows no decline. The feared mechanism is real but decoupled: adoption raises per-function code complexity (about +11% on a cognitive metric for Python, a quarter of the prior estimate, and +3 to 4% in cyclomatic terms across all languages), yet in fixed-unit subsets where complexity rose (Python on the cognitive metric, and all languages on the cyclomatic metric), newcomer participation does not decline. These results suggest that, in established open-source projects, adopting an AI coding agent makes code modestly more complex but does not crowd out the human newcomers that a project depends on: the feared trade-off between AI assistance and human participation does not materialize.
Modern compilers are frequently updated, but expert review capacity is highly limited, leading to delayed integration and, in some cases, subtle semantic bugs entering the compiler codebase. Automating the code review process with modern general code review agents may be feasible, but it faces critical challenges due to compiler complexity. In this paper, we use LLVM as our target compiler and present Archer, the first automated agentic code review tool for compiler optimizations. Archer constrains the agentic review process from both ends by using obligations to guide analysis and a deterministic validation guard to admit only findings backed by executable evidence. We evaluated Archer on 70 open PRs and 328 closed PRs in LLVM from the last two months. The review results are shocking and concerning: Archer discovers that 21% of open PRs and 11% of closed PRs are buggy, i.e, introducing semantic bugs such as miscompilations in LLVM. Our findings expose a critical gap in the capacity for critical review in large compiler projects and demonstrate the practical value of Archer as an additional reviewer.
Steering vectors have emerged as a promising approach to controlled text generation, offering interpretable, training-free mechanisms for shaping model outputs. However, their practical generality remains poorly understood. We study the limits of steering vector generalization along three dimensions: trait expressibility, task transfer, and multi-trait composition. Using the PLUME writing personalization benchmark, we extract steering vectors for a range of preferences and evaluate them on summarization and email-writing tasks across two open-source models (Qwen2.5-7B-Instruct and Llama3.1-8B-Instruct). We find that steering effectiveness varies substantially across traits. We further show that steering effectiveness can degrade when vectors extracted from positive and negative style examples are transferred to downstream writing personalization tasks. Finally, we compare common methods for composing multiple steering vectors and find that all methods suffer significant drops in trait expression as more vectors are added, with a tradeoff between coherence and expressibility that requires per-setting hyperparameter tuning. Taken together, our results suggest that steering vectors face meaningful limits as a general-purpose tool for preference alignment.
Large Language Models (LLMs) have recently shown promise in molecular discovery, yet a gap remains between their probabilistic nature over discrete sequential tokens and the rigid topological constraints of chemical space. This raises the question of whether molecular LLMs can generalize beyond the local neighborhoods induced by their sequence-based representations. To systematically investigate this question, we introduce a Molecular Perturbation framework that generates syntax-valid structural variants of training molecules under controlled Graph Edit Distance (GED) to probe the manifold regularity of molecular LLMs. Our analysis shows that even a single edit can cause substantial performance drops on common molecular tasks, revealing a narrow local trust region and fragile sensitivity to structural changes. Since similar molecules tend to exhibit similar properties, In-Context Tuning (ICT), which anchors predictions on structurally similar molecules, offers a natural way to mitigate such fragility. Our experiments also examine whether ICT confers robustness under controlled structural perturbations, and the results suggest that it can partially expand the local trust region and offer a promising direction for stabilizing molecular LLMs against structural variation.
Sparse autoencoders (SAEs) decompose internal activations of neural networks into sparse linear combinations of learned features by fitting an overcomplete dictionary $\mathbf{W}\in\mathbb{R}^{m\times n}$ with $m<n$, and inferring a sparse code $\mathbf{x}\in\mathbb{R}^n$ from $\mathbf{h}\approx\mathbf{W}\mathbf{x}$. This inference problem closely resembles the canonical setup of compressed sensing, but dense decoders requires $O(mn)$ learned values, which becomes costly at large feature counts. We introduce Expander SAEs: TopK SAEs whose decoder and tied encoder are supported on a left-$d$-regular expander mask with $d\ll m$, learning only $dn$ decoder values while keeping the sparse-coding problem $(m,n,k)$ fixed. The same structure reduces storage and turns the matching-pursuit correlation step $\mathbf{W}^\top \mathbf{r}$ in OMP into an $O(dn)$ gather-and-reduce operation. Our experiments show that across Pythia-70M/160M, Qwen2.5-3B, and Llama-3.2-1B residual-stream activations, varying $d$ traces a consistent storage--fidelity frontier, and that at the most compressed modern-LM setting, Qwen2.5-3B with $d=7$ uses $293\times$ fewer learned decoder values than the full dense decoder while retaining $84$% of dense CE-loss recovered. Control experiments show that the improved storage--fidelity tradeoff is driven by sparse, diverse decoder support structure rather than by fewer learned decoder values, and that when sparse and dense decoders are compared at matched parameter count, part of the remaining gap comes from encoder amortisation. On the theoretical side, we show that expansion and column flatness are sufficient for identifiability of noiseless $k$-sparse codes, and we derive complementary sufficient conditions under which OMP recovers the support exactly.
Monitoring cognitive load during online learning could help instructors identify content that learners find difficult, but remote settings remove the visual cues that support this judgement in a classroom. We study whether a single-channel, consumer-grade EEG device (the NeuroSky MindWave Mobile 2) can distinguish easy from difficult educational-video content, using the publicly available dataset of Wang et al. [24] (ten learners, one excluded for excessive noise, leaving nine). We implement a hybrid CNN+LSTM+Attention model that combines the raw waveform with band-power features. In a within-subject setting, the model reaches up to 78.5% accuracy, compared with 55% for conventional feature-based classifiers; regularization (dropout and L2) closes the large gap between training and validation accuracy that we observe without it, keeping validation accuracy stable at roughly 68-73%. We are deliberately cautious about these numbers: with only nine subjects, within-subject evaluation is optimistic, and we argue that subject-independent evaluation -- in which no learner appears in both training and test data -- should be the standard for this task. To that end we release a reproducible evaluation pipeline. We frame the work as a feasibility study rather than a deployable system, and pair it with an open, notebook-based tool that records EEG, runs inference, and visualizes estimated cognitive load as a heatmap over the video timeline to help educators locate potentially challenging segments.
With the rapid development of autonomous aerial systems, Unmanned Aerial Vehicles (UAVs) are increasingly deployed in applications such as inspection, environmental monitoring, and rescue, creating growing demand for reliable autonomous navigation. However, autonomous UAV navigation in dense environments remains challenging under sparse perception and dynamic constraints. Most reinforcement learning (RL) methods lack explicit safety mechanisms, leading to unsafe exploration, unstable training, and risky behaviors, especially during high-speed flight. Even in safe RL approaches, safety is often enforced by projecting policy outputs onto a safe action set, which may introduce instability. Meanwhile, many learning-based methods rely on dense inputs or large networks, increasing computational burden and limiting lightweight onboard deployment. Facing the above challenges, we propose a safety-constrained perception-control integrated framework for UAV navigation. A lightweight network encodes sparse observations into collision-risk-aware features using asymmetric and depthwise separable convolutions. We formulate the task as a constrained Markov decision process within a hierarchical control architecture and solve it using a Lagrangian-based safe PPO algorithm. Curriculum learning further improves training stability. Experiments with varying obstacle densities and flight speeds demonstrate higher success rates, improved safety, and better efficiency than existing reinforcement learning baselines.
LLM agents increasingly perform autonomous actions through external tools, leading to complex and evolving safety risks. However, existing safety testing targets expert-designed safety violations, and the corresponding outcomes are evaluated by hard-coded rules, making them costly to extend as agents evolve. To this end, we present Vera, an end-to-end automated safety testing framework that instantiates software engineering testing principles for non-deterministic agents through a three-stage, self-reinforcing pipeline. First, a literature-driven exploration continuously discovers and structures emerging risks into taxonomies of safety risks, attack methods, and tool execution environments. Second, combinatorial composition across taxonomy dimensions produces executable safety cases, each specifying a concrete safety goal, a programmatically constructed initial state, and a deterministic verification predicate grounded in observable artifacts. Third, adaptive execution runs heterogeneous agents in isolated sandboxes where a control agent steers multi-turn interaction based on runtime observations, while evidence-grounded verifiers judge outcomes from environment state and tool-call evidence rather than model self-report. We evaluate Vera on four production agent frameworks (OpenClaw, Hermes, Codex, Claude Code), revealing substantial safety weaknesses, with average attack success rates reaching 93.9\% under multi-channel attacks; we also release Vera-Bench, comprising 1600 executable safety cases spanning 124 risk categories across three execution settings. These results indicate that modular, executable testing infrastructure is essential for rigorous and maintainable safety evaluation of rapidly evolving agentic systems at scale. The code is publicly available at https://github.com/Yunhao-Feng/Vera.
While decoder-only LLMs excel at a vast array of natural language tasks, it suffers from an asymmetric information flow induced by causal attention: later tokens are richer in contextual grounding than earlier ones. A simple and effective remedy is prompt repetition -- just appending a second copy of prompt before generation can redistribute grounding across positions and improve reasoning performance. However, full repetition of the original prompt doubles the KV cache footprint and quadruples attention cost during prefill, making it impractical for long-context settings. We propose PartRep, a selective augmentation method that appends only the most informative tokens -- rather than the entire prompt. We use token-wise negative log-likelihood (NLL) as a selection signal, motivated by the hypothesis that less predictable tokens are less recoverable from surrounding context and therefore benefit more from late-position repetition. To avoid the heavy cost of a full forward pass for scoring, we train a lightweight gate that predicts high-NLL tokens from early-layer hidden states, enabling token selection during mid-prefill via early exit. Across eight benchmarks (including MMLU, GSM8K, and RULER) and three model families (Qwen2.5, Llama3.2, Gemma4), PartRep retains most of the gains of full repetition while using only 59.4\% of its KV cache and 79.0\% of its prefill FLOPs.
Mixture-of-Experts (MoE) models scale efficiently but remain costly to adapt due to redundant experts and uniform parameter allocation. Existing parameter-efficient fine-tuning (PEFT) methods such as LoRA ignore MoE routing dynamics, leading to suboptimal resource use. We propose EPnG, an adaptive prune-and-grow framework that reallocates LoRA capacity based on expert importance derived from router gate probabilities. EPnG prunes under-utilized experts and expands high-importance experts via rank growth with orthogonal initialization, while maintaining a fixed parameter budget. Across OLMoE and Qwen1.5-MoE, EPnG consistently outperforms LoRA under the same budget and achieves performance comparable to full fine-tuning while updating only 0.55%-0.72% of parameters (up to 140x-180x fewer). These results demonstrate that aligning PEFT with MoE routing yields a more effective and scalable fine-tuning strategy.
Hyper-scale microservice systems have become the standard infrastructure for large-scale Internet companies. These systems consist of numerous loosely coupled microservices that evolve independently through continuous development and deployment. Such complexity makes failures unavoidable, necessitating efficient Root Cause Analysis (RCA) to help Site Reliability Engineers (SREs) quickly localize root cause services and classify failure types. However, existing RCA methods often struggle to adapt to the extreme dynamism and massive scale of these systems. In this paper, we present KRCA, an end-to-end RCA system designed for hyper-scale microservice systems. To manage the vast search space, KRCA employs a multi-stage pipeline that begins with an API-level drilldown to isolate suspicious services. It then instantiates a skeleton-based causal graph from anomalous metrics to serve as a high-recall structural prior, before utilizing a memory-augmented multi-agent framework to verify causality and generate the final failure report. By combining structured causal constraints with multi-agent reasoning, KRCA employs balances diagnostic accuracy with the efficiency requirements of real-time production use. Experimental results show that KRCA achieves AC@1 scores of 0.88 and 0.79 for root cause service localization and failure type classification, outperforming the strongest baseline by at lease 31% in absolute gains. KRCA has been deployed in Kuaishou's production environment for over six months, reducing the average diagnosis time by 77.3%.
Next activity prediction helps service-oriented processes anticipate upcoming steps before delays, exceptions, or service-level risks occur. Most existing methods assume classical single-case event logs, whereas real service processes often involve events shared by multiple typed business objects. Object-centric event logs (OCELs) capture such interactions, but current predictors remain limited. Flattening-based approaches lose cross-object context, and native OCEL graph-based approaches encode multi-object events through pairwise relations. Existing models also do not jointly capture event-driven object state changes, inter-event timing, and global execution patterns. We propose EHHN, an Event-driven Heterogeneous Hypergraph Network for object-centric next activity prediction. EHHN represents each prediction prefix as a heterogeneous hypergraph, where event--object hyperedges bind retained co-participating objects and a lifecycle hyperedge groups the primary object's observed lifecycle events. Based on this representation, EHHN uses a dual-stream architecture in which a micro-spatial stream models event-driven object-state evolution and a macro-evolution stream captures temporal dynamics using retrieved global prototypes. The two streams are fused to predict the next activity. Experiments on four public OCEL benchmarks against nine baselines show that EHHN achieves the best accuracy and macro F1-score on all datasets, with improvements of up to 8.1 and 12.4 percentage points over the strongest baselines. Compared with the strongest OCEL-native graph baseline, EHHN also reduces peak GPU memory by up to 24 times. Code is available at https://github.com/chenkaitao1112/EHHN.
Radio frequency (RF) maps provide a compact representation of multipath propagation characteristics and are fundamental to channel modeling, coverage analysis, and environment-aware wireless optimization. This paper proposes a unified RF map construction framework based on a physics-informed neural network (PINN) and a graph neural network (GNN), supporting both cross-scene generation and in-scene completion with 2D and 2.5D environmental representations. The PINN embeds electromagnetic propagation constraints to establish a physically consistent mapping from receiver locations to multipath parameters, including path gain, time of arrival, and angles, while the GNN enforces spatial consistency by modeling correlations among neighboring receivers. To comprehensively evaluate multipath reconstruction quality, we propose a peak-weighted dynamic time warping metric that jointly accounts for amplitude errors and peak delay misalignment in channel impulse responses. Extensive experiments demonstrate that the proposed method consistently outperforms image-based, diffusion-based, and interpolation baselines across both map-level and multipath-level metrics, achieving robust generalization and high-fidelity RF map construction under sparse observations.
In the age of AI, what will be good knowledge? This article, which is accepted and forthcoming in a special issue of Modern Fiction Studies on "Cultural AI" in 2027, applies digital humanities methods to map epistemic virtues (like "true," "accurate," "creative") used in a corpus of 553 journal articles on AI published in 2024. "Creativity" comes in for special attention as an example. Exploring this discourse of value, the article considers how a framework might be developed for evaluating the knowledge-worth of AI -- one less locked into values formed around pre-AI "knowledge work" agents or structures, and more open to the future values of "generativity." The essay is supported by an online digital kit for exploring data models of the corpus of articles on AI it studies.
Discrete diffusion models have steadily improved in quality relative to autoregressive (AR) models. However, these models are normally constrained to fixed-length generation and do not support key-value (KV) caching. Block diffusion partially bridges diffusion and AR by generating token blocks left-to-right, but its fixed-size sequential blocks limit decoding flexibility and parallelism. Here, we present a new class of language models, set diffusion, comprised of (i) a likelihood parameterization that factorizes over flexible-position, flexible-length token sets and (ii) a set-causal diffusion architecture that supports KV cache updates after every inference step. By factorizing over token sets instead of fixed-size blocks, tokens can be decoded in arbitrarily-ordered sets, including sliding-window sets, enabling faster inference and support for any-order decoding. Set diffusion achieves better speed-quality tradeoffs on mathematical reasoning, summarization, and unconditional generation compared to prior diffusion language models while offering stronger infilling performance than block diffusion. We provide the code, along with the model weights and blog post on the project page: https://m-arriola.com/setdlms/
Diffusion Language Models (DLMs) have recently emerged as a promising alternative to autoregressive models. Unlike standard diffusion-based approaches, DLMs are not explicitly conditioned on a timestep, raising a natural question: do these models internally represent denoising progress, and how is such information used downstream? In this work, we show that DLMs do in fact encode a latent representation related to the diffusion timestep within their residual streams. We find that this signal can be reliably extracted using probes across layers, indicating that denoising progress is decodable from internal activations. We further demonstrate that steering the model along a low-dimensional subspace associated with the inferred timestep allows us to systematically modulate its notion of denoising progress, leading to predictable changes in model confidence and entropy. Finally, we analyse the geometry of the identified representation, showing that it exhibits structured and interpretable properties in activation space, and shedding light on how such a signal is processed by these models.
Ontology construction requires deciding which objects, attributes, and structural relations should be accepted as valid knowledge. Language models can propose such structures from text, but their outputs can still be unsupported or inconsistent. This paper proposes a retrieval-augmented small language model (SLM) framework that uses formal concept analysis (FCA) as a symbolic verification loop for knowledge expansion. Starting from seed attributes, FCA proposes implications over a growing formal context. A retrieval-grounded SLM oracle then validates each implication or returns a counterexample. The oracle also supports incidence judgments, consistency checks, and attribute proposals, making accepted implications, counterexamples, contradictions, and corrections inspectable. In a rare ataxia setting constructed from Orphadata resources, retrieval-grounded 10-seed runs obtain relation F1 of 0.29-0.52 and closure-based implication F1 of 0.22-0.30. Larger seed sets increase the number of evaluated implications and often improve implication F1. The lower implication scores reflect a stricter evaluation of derived implications, where one missed or extra relation can affect several implication judgments. Ablations show that incidence judgments in a fixed object-attribute setting can improve closure-based implication scores. However, identifying positive object-attribute pairs remains difficult even when the candidate objects and attributes are fixed.
As agent planning moves from short tool chains toward persistent workflows with thousands or tens of thousands of steps, failures will occur inside large planning graphs rather than in isolated predictions. Replanning the entire graph after every mistake is neither computationally realistic nor desirable: full-graph replay consumes large context budgets, exposes the LLM to many irrelevant symptoms, and can degrade long-context retrieval. This paper studies the missing component in such systems: a world-model corrector that repairs the failed planning graph in place. We compare two families of correctors. The first is the common engineering approach: scan nodes and edges, choose a suspicious local region, and ask an LLM to repair it. We implement strong engineering LLM correctors and find that they can help, especially when given very large contexts. The second family is our approach, WM-SAR (World-Model Subgraph Amplification Repair): instead of scanning for visible symptoms, it works backward from subgraph amplification, identifies the nodes and edges that keep re-amplifying error, and sends only that causal subgraph to the LLM. Across graph simulations and LLM repair experiments, WM-SAR substantially outperforms engineering correctors under realistic token budgets, achieves near-whole-graph stabilization with a compact region, and gives the LLM a cleaner repair target.
LLM agents are increasingly used to translate natural language into 3D scenes in a procedural way, but existing systems focus on static output. Dynamic 4D scenes from text alone, in which liquids flow, particles emit, rigid bodies cascade, and articulated mechanisms move, remain largely unexplored despite their value as editable content and as physics-grounded training data for video generation and embodied AI. Two challenges set the dynamic case apart from static text-to-scene work: an agent must jointly coordinate spatial layout, multiple physics solvers, temporal sequencing, camera, and lighting in a single coherent scene, and verifying motion correctness from rendered video is fundamentally harder than judging a single image. We present SimWorlds: a multi-agent framework that produces dynamic, editable 4D scenes from text, with Blender-specific procedural knowledge, a planner-coder-reviewer workflow driving a fixed ordered sequence of construction stages, a layered scene protocol enforced by a deterministic verifier, and a runtime-state inspection tool suite that catches mechanism failures the rendered image cannot reveal. We also introduce 4DBuildBench, a benchmark for assessing both visual fidelity and physical consistency of the procedural dynamic 3D scenes generated from text prompts. Experiments show that SimWorlds outperforms prior dynamic Blender generation baselines.
Repository-level vulnerability reproduction is a demanding software engineering (SE) task: an agent must inspect a codebase, infer the input grammar that reaches a vulnerable path, construct a proof-of-conceptv(PoC), and verify that the crash disappears on the patched build. Recent LLM agents can often execute these steps when the approach is correct, yet they still fail by choosing the wrong strategy. This paper argues that strategy, rather than the full action trajectory, is the right learning unit for such SE agents: it is compact enough to optimize, concrete enough to guide execution, and stable enough to store and reuse across attempts. We present Mastermind, a dual-loop framework that separates transferable strategy learning from task-specific experience. A trainable planner learns reusable vulnerability-reproduction strategies through SFT and milestone-based GRPO, while an experience loop maintains task-local strategy records that guide subsequent attempts. The planner is trained independently of the executor, allowing strategy learning to improve multiple frozen executors without modifying their action-generation capability. We evaluate Mastermind on CyberGym using 260 training tasks and 200 held-out evaluation tasks. With GPT-5.5 as the frozen executor, Mastermind achieves an 84.5% pass rate, outperforming open-book PoC context (60.0%), Best-of-8 sampling (63.0%), and iterative improvement (77.0%). The same planner also improves GPT-5.4 mini and GLM~5.1 from 45.0% and 58.5% to 60.0% and 71.0%. These results demonstrate that learning high-level strategies is an effective and transferable mechanism for improving repository-scale SE agents.
Continual post-training enables foundation models to acquire new knowledge while preserving existing capabilities. Recent work suggests that on-policy learning can mitigate forgetting, with on-policy self-distillation emerging as a particularly attractive approach. In this work, we revisit this optimistic view through self-distillation policy optimization (SDPO). Our experiments show that SDPO can accelerate in-domain specialization when teacher signals are stable and well aligned, but it struggles to generalize to out-of-distribution scenarios. In continual post-training, SDPO exhibits stronger forgetting and can even collapse, whereas on-policy reinforcement learning methods such as GRPO adapt more conservatively and better preserve prior capabilities. Further analyses reveal that denser self-distillation induces larger drift in both parameter space and response space, and can amplify high-frequency formatting artifacts through a self-reinforcing teacher--student loop. These findings suggest that on-policy data alone is insufficient for continual learning. Dense self-distillation can accelerate specialization when teacher targets are stable and token-level supervision is reliable, but it should not be treated as a default stabilizer for continual post-training. Our code is available at https://github.com/Moenupa/SDPO-CL.
Many representation learning problems involve directed relations, such as lexical entailment, sentence entailment, ontology hierarchy, and citation links. Standard Euclidean, cosine, and Mahalanobis heads are symmetric, while generic neural scorers can model directionality but provide limited geometric structure. This paper proposes a role-aware neural convex divergence head for asymmetric representation learning. The head applies source- and target-role projections before evaluating an input-convex neural Bregman divergence, yielding a nonnegative structured score in the role-projected space. We characterize its projected-space identity, source-role convexity, directional-gap decomposition, and Hessian-based local curvature. Experiments on lexical, sentence, ontology, and directed graph benchmarks compare symmetric distances, unstructured asymmetric scorers, order/hyperbolic baselines, plain ICNN-Bregman heads, and the proposed role-aware variant. Across ten random seeds on the main semantic and ontology benchmarks, role-aware projections consistently improve directional accuracy over plain ICNN-Bregman heads while preserving zero observed negative divergence rate. The results also identify a boundary case: on large fixed-feature citation prediction, specialized symmetric or hyperbolic baselines remain stronger in ranking accuracy. Overall, the proposed head is best understood as a structured and interpretable plug-in distance module for tasks where directional relations matter.
Vulnerability exploits play a crucial role in assessing the downstream impact of Java library vulnerabilities. While some vulnerabilities are accompanied by disclosed exploit references, automatically reproducing such references into runnable exploits remains challenging because they are often incomplete, unstructured, or only describe partial reproduction steps. Recent code agents provide a promising way to automate this process, but our study shows that their generated exploits often appear successful without triggering the actual vulnerable logic, such as replacing vulnerable APIs with self-implemented functions. To address this, we propose Refploit, an LLM-based trajectory recovery framework for facilitating vulnerability reproduction from public exploit references. The key insight is that a failed agent trajectory is not entirely useless. It may have already completed some reproduction subtasks while also revealing misleading directions that should be avoided. Refploit first validates an agent-generated exploit through differential execution. When the exploit is ineffective, Refploit analyzes its reproduction progress, locates the trajectory segments associated with the reproduction progress, and derives constraints to guide focused recovery. We evaluate Refploit on three open-source Java vulnerability datasets, covering 172 exploit references for 143 vulnerabilities. Under DeepSeek-V4-Flash, Refploit successfully reproduces 138 exploits, achieving a reproduction rate of 80.2%. It achieves a 64.3% relative improvement over the initially generated trajectories and outperforms both the SOTA exploit-generation method PoCGen and advanced code agents such as Codex with GPT-5.4. We further adapt Refploit to another code agent and observe consistent improvements, demonstrating its generality.
Open-vocabulary object detection aims to localize and classify objects beyond the fixed set of categories seen dur ing training. Recent open-vocabulary object detection methods improve localization and classification for unseen categories by leveraging a frozen VLM as a detector backbone. However, VLM classification score lacks recognizing position and scale of the object in an image. We observe that pretrained VLMs en able to classify foreground and background regions. According to this observation, we propose a simple inference-time Pro posal Calibration (ProCal) that improves localization quality of the classification score. ProCal computes a proposal prior by combining two scores: localization-aware foreground score and background-aware suppression score. Localization-aware foreground score captures whether a proposal contains an object area. Background-aware suppression score measures the extent to which the proposal resembles background. We analyze that ProCal suppresses false novel activation on background proposals and consistently ranks true novel proposals above background and partial novel proposals. Applied to CLIPSelf ViT-L/14, ProCal improves APr +2.5 on OV-LVIS. The analyses show that proposal-level localization-aware reranking effects to mitigate ranking miscalibration for novel objects.
In this paper, we consider the nonsmooth nonconvex decentralized optimization problem, where inter-agent communication is compressed. We propose a general framework that unifies various decentralized stochastic subgradient-type methods with unbiased compression and contractive compression with error compensation. By relating the consensus-error iterates and the averaged iterates to the trajectories of continuous-time differential inclusions, we establish global convergence for all methods encompassed by our framework when the objective functions are nonsmooth and lack Clarke regularity. Based on our framework, we further develop several compression-based methods, including decentralized stochastic subgradient methods utilizing sign-based regularization and gradient-tracking momentum. Preliminary numerical experiments empirically support our theoretical results and highlight the communication-accuracy trade-off of the newly developed methods.
On-policy exploration is a crucial component for training robust Vision-Language Navigation agents, as it exposes the policy to a broader state distribution. However, such exploration inevitably leads to trajectories that deviate from expert demonstrations, resulting in a semantic mismatch between the executed visual stream and the original language instruction. In this work, we address this challenge by introducing Phi-Nav, a unified on-policy framework that leverages hindsight reasoning to align instructions with the agent's actual exploratory journey. Specifically, Phi-Nav operates through a three-stage dual-supervision cycle: 1) the agent performs oracle-guided on-policy exploration, sampling a trajectory while learning from expert action feedback, 2) a hindsight speaker synthesizes a path-level hindsight instruction grounded in the collected visual observations, and 3) the agent conducts a second imitation pass, treating the synthesized trajectory-instruction pair as an additional expert demonstration. Through this process, Phi-Nav bridges the critical semantic supervision gap inherent in on-policy methods, transforming semantically unlabeled movement into dense training signals. Evaluations on the R2R-CE and RxR-CE benchmarks show that Phi-Nav yields competitive performance while requiring only a fraction of the expert demonstrations used by current baselines. These results underscore the necessity of semantic exploration in VLN, positioning Phi-Nav as an effective solution for training embodied agents with limited data.
Temporal point processes (TPPs) have widespread applications across various domains. Compared to modeling the conditional intensity of a TPP, modeling its cumulative conditional intensity function (CCIF) improves computational efficiency and eliminates numerical approximation errors. However, current CCIF parameterizations uniformly rely on Monotone Neural Networks (MNNs), which we identify as suffering from three structural deadlocks--convexity restrictions, saturation limits, and violations of CCIF modeling requirements--that fundamentally restrict their representational capacity for complex temporal dynamics. To resolve these bottlenecks, this paper proposes a novel framework called Monotone Alternating Splines (MAS). By leveraging distinct interpolation and extrapolation components, MAS provides a flexible and efficient framework for modeling CCIFs. Theoretically, MAS's interpolation provides strong fitting accuracy, while its extrapolation supports robust generalization, reducing the irreducible approximation gaps of MNNs. Extensive experiments show that MAS achieves superior performance on both synthetic and real-world datasets.
Existing medical video benchmarks primarily evaluate whether a model produces the correct answer, but rarely assess whether it answers at the right time. In real clinical settings, AI systems must decide not only what to predict, but also when to answer, defer judgment, or proactively raise alerts. This creates a critical gap between benchmark evaluation and deployment requirements. We present MedStreamBench, a benchmark for time-aware medical video understanding. MedStreamBench integrates 22 medical datasets and 5,419 QA instances across four temporal settings: retrospective, present, future, and proactive. Unlike conventional benchmarks that assume full-video access, MedStreamBench restricts models to temporally bounded evidence windows and supports both single-turn and streaming evaluation. We further introduce a proactive monitoring setting that requires models to determine whether and when clinically relevant alerts should be triggered. Beyond answer correctness, MedStreamBench evaluates temporal behavior through responsiveness and post-evidence stability. Experiments on leading general-purpose and medical vision-language models reveal a substantial gap between offline recognition and temporally grounded decision-making, with performance dropping markedly in streaming and proactive settings. Our benchmark is available at https://huggingface.co/datasets/Venn2024/MedStreamBench.
Recurrent representations are trajectories, but representation geometry is often measured from static snapshots. We develop finite-lag operator geometry for recurrent hidden states from observed source-successor pairs $(X_t,X_{t+Δ})$. The primitive is the conditional transport law $Q_Δ(dy\mid x)$, estimated by a dense Gaussian source-smoothing operator. From this directed finite-lag law we derive a source-centered transport tensor $G_Δ$, which decomposes exactly into conditional spread and coherent displacement, and an antisymmetric coordinate circulation $W_Δ^ρ$, which summarizes directed lagged flow. We prove affine covariance with explicit metric dependence of scalar summaries, dense estimator stability on bounded trajectory clouds, and a finite-lag separation result showing that source-centered transport detects deterministic recurrent motion not recorded by infinitesimal carre-du-champ geometry. A linear-Gaussian closed form calibrates the quantities in terms of the update $A_Δ$, source covariance, and innovation covariance. Controlled experiments validate the decomposition, circulation, covariance, and stability predictions. In performance matched repeat-copy networks, the framework reveals architecture dependent differences in total transport scale and coherent displacement trace, while coherent displacement fraction is metric and resolution dependent.
Smart contract vulnerabilities are predominantly logic bugs whose detection requires structured, step-by-step procedural knowledge of attack patterns and contract semantics. Existing LLM-based methods struggle to generate this knowledge automatically: prompt-based methods rely on manually crafted detection rules, while fine-tuning requires massive labeled datasets that are inherently scarce in this domain. We present EvoVuln, an automated framework that reformulates vulnerability detection as a procedural knowledge evolution problem, synthesizing and refining detection logic using only a minimal number of labeled samples. To achieve this, EvoVuln introduces two key mechanisms. First, a Runtime with an Inversion of Control (IoC) architecture compiles detection rules into Executable Policies. This strictly decouples deterministic control flow from LLM semantic reasoning, ensuring faithful logical adherence and producing dense diagnostic telemetry for precise error localization. Second, a two-phase evolution pipeline refines the rule via abductive semantic debugging without any parameter updates: Cold Start bootstraps and stress-tests an initial rule using auto-synthesized corner cases; Few-Shot Evolving then grounds the policy in real-world semantics using only five vulnerable and five safe examples per vulnerability type. Evaluated across five real-world vulnerability types, EvoVuln achieves a 71% macro-average F1-score, outperforming all baselines. The evolved procedural knowledge is portable across models: it enables a lightweight, low-cost model to surpass a much larger zero-shot model by 19 percentage points, and transfers to other LLMs without retraining, at a one-time evolution cost under $50.
Reinforcement Learning (RL) is a sequential decision-making framework in which an agent learns optimal policies through interaction with an environment by maximizing cumulative rewards. Among RL methods, Bayesian Reinforcement Learning (BRL) addresses common practical challenges related to data scarcity by leveraging prior knowledge about the environment and sequential belief updates. However, most BRL approaches require an explicit likelihood function, which is frequently inaccessible or intractable in real-world settings. We propose Likelihood-Free Iterated Batch Importance Sampling (LF-IBIS), a novel algorithm for BRL that updates the agent's beliefs online as new interactions become available. By combining Approximate Bayesian Computation with Iterated Batch Importance Sampling, LF-IBIS enables full Bayesian inference in settings where the environment dynamics are not described by an explicit or tractable likelihood. The method yields approximate posterior distributions over both environment parameters and optimal policies, providing a quantification of policy uncertainty useful for a Bayesian treatment of the exploration-exploitation trade-off. We test the method on a simulation study in response-adaptive randomization in clinical trials, where closed-form posteriors enable validation. Additional experiments address settings where the posterior has no closed form and illustrate online policy updating based on the posterior distribution of the optimal policy.
Public LLM leaderboards optimise for global average performance and do not capture the specific cognitive demands of financial-services work: a model that leads on MMLU-Pro may underperform on document-grounded compliance reasoning, and a coding leader may handle multi-turn customer interactions poorly. We present a meta-benchmarking framework that organises 452 publicly reported benchmarks into 41 O*NET Generalized Work Activities and aggregates those into 38 BIAN banking business domains spanning sales, operations, risk, and support work. A multiplicative weighting scheme (discrimination x coverage x recency), computed over a rolling model window, rewards benchmarks that still separate the best models, are widely reported, and remain in active use, suppressing saturated legacy tests automatically. These weights scale the K-factor in a pairwise Elo tournament, producing cross-benchmark-comparable work-activity scores without raw score normalisation; business-domain scores are weighted averages of the constituent work-activity Elos. We demonstrate the framework on a point-in-time public snapshot covering 288 models across 25 organisations as of June 2026, and describe the methodology, full taxonomy, design decisions, and limitations with the aim of making the approach reproducible for institutions facing similar selection and governance challenges.
We study how to predict the downstream closed-loop performance of a learned latent world model from validation-time diagnostics alone. Choosing the right checkpoint from a world-model training run is difficult: validation loss and multi-step prediction RMSE keep improving long after closed-loop performance has collapsed. We present a suite of structural validation-time diagnostics drawn from optimal-control theory and apply them to Gymnasium's LunarLander v3, which features shaped rewards. We train an RSSM [5, 4] world model on it and treat per checkpoint CEM-MPC return as the oracle for closed-loop quality. By evaluating 40 metrics against this oracle, we find that the strongest single predictor is the Reward Observability Fraction (ROF), which measures the reward predictor's dependence on the observable subspace. We combine ROF with three structural regularizers into a single-number offline checkpoint selection score, the Composite Reward Observability Fraction (CROF). The CROF-selected world model trains a model-based A2C policy that beats a fairly evaluated model-free A2C baseline by ~24.5 return points while using ~65x fewer real-environment interactions, and the same world model also drives a strong zero-shot CEM-MPC policy. Code and data: https://github.com/nsmoly/LunarLander_RSSM.
We present a case study in reformalization, a variant of autoformalization in which the input proof is not natural language but a formal development in a different proof assistant. Concretely, we report three reformalizations of the Jordan Curve Theorem: from Mizar to Lean, from HOL Light to Lean, and from HOL Light to Agda. We analyse the results and identify pipeline design choices that matter for practical reformalization tasks.
Speech-LLM integration has shown promising results by leveraging extensive textual pretraining, yet its specific benefits for automatic speech recognition (ASR) remain unclear. We observe that as supervised ASR training data increases, the contribution of LLM priors becomes less evident, and simple speech-text joint training under-utilizes textual knowledge. We therefore propose Joint Speech-Text Interleaved Pretraining (JSTIP), an ASR-oriented pretraining strategy that constructs word-level and segment-level interleaved speech-text sequences within aligned pairs for speech-LLM architectures that accept continuous inputs. Experiments on 38k hours of ASR data show consistent entity accuracy improvement compared to ASR-only and joint speech-text training baselines. JSTIP achieves on-par entity recognition performance using domain transcription text compared to synthetic speech-text pairs, simplifying domain adaptation. Benefiting from textual pretraining and domain text data, JSTIP is competitive with open-source ASR and Speech-LLM systems in medical entity recognition. The zero-shot speech question answering behaviors further suggest that interleaving reduces the speech-text modality gap and preserves the LLM generative prior, which is likely the reason for the entity improvements on the ASR task.
This study provides a theoretical expansion of the recent Data Relativistic Uncertainty (DRU) framework by formalizing a physics-to-AI paradigm for image enhancement. By modeling images as probabilistic wave functions rather than deterministic states, the paradigm explicitly integrates wave-particle duality to illustrate the system flow of how DRU leverages the intrinsic physical uncertainty of light, a dimension requiring further theoretical discussion. Consequently, this paradigm provides a rigorous Explainable AI (XAI) approach that enhances the interpretability of how DRU mitigates illumination bias and maintains robustness against data noise.
Deep learning models for speech classification are vulnerable to backdoor attacks, where malicious triggers cause misclassification at inference time. While sample-specific attacks can bypass many defenses, they often rely on poisoned label attack, making them detectable via manual data defense. In this paper, we propose DRL-CLBA, a novel clean label backdoor attack for speech classification that leverages Deep Deterministic Policy Gradient (DDPG) reinforcement learning. We also utilize deep audio steganography to embed sample-specific triggers into source audio, creating feature-space anchors. The proposed reinforcement learning framework effectively optimizes target samples toward trigger-bearing anchor points in the model's deep latent space, enabling label-migration-free poisoning of target samples. Experimental results across three datasets and four different DNNs demonstrate that DRL-CLBA achieves a high attack success rate, effectively bypassing some backdoor defenses. The attack demonstrates strong resistance against fine-tuning, pruning, and spectral signature defenses, exposing critical vulnerabilities in speech-controlled systems.
Visual regression testing (VRT) is a standard quality assurance step in modern software release pipelines. On every change, it re-renders user interface (UI) screenshots, compares each one against an approved baseline image, and routes any detected difference to a human reviewer who decides whether it is an intended update or an unintended regression. A widely used approach, especially in open-source and continuous-integration pipelines, is pixel-level comparison, which is semantically blind and treats rendering noise and genuine defects identically, producing large volumes of false positives that force developers and testers to spend substantial time and effort manually reviewing flagged differences at every release cycle. Industry tools apply machine learning to VRT, but lack public evaluation. More critically, no dataset or benchmark exists to support natural language descriptions of UI changes, a capability that tells testers what changed in words instead of leaving them to interpret a binary flag or a highlighted region. To address the gap, we propose a new task, Web UI Image Change Captioning (WUICC), which sits at the intersection of VRT and image difference captioning (IDC), and release WUICC-bench, its first dataset and benchmark for the task. We evaluate eleven representative IDC methods, together with two zero-shot general-purpose LLMs. We find that: (1) these methods tend to struggle in the Web UI domain due to its layout diversity, dense text, and fine-grained changes, and (2) yet the trained methods already suppress non-meaningful visual noise far more selectively than the pixel-level comparison VRT relies on, providing a solid foundation for future domain-specific research.
Synthetic data can be scaled along two routes: Source Expansion (SE), which enlarges the source by adding seed materials or generators, and Fixed-Source Synthesis (FSS), which holds the source fixed and scales the generation budget. Existing scaling studies typically expand the source as the data grows, conflating SE with FSS and leaving FSS underexplored. We isolate FSS by holding the seed-question pool and teacher model fixed, varying only the per-question response budget under Rejection Sampling (RS). We adapt the rectified scaling law to FSS, deriving it from how repeated sampling covers a fixed source. Empirically, the derived form, fit on low budgets, predicts performance at the held-out highest budget for every evaluated teacher--student pair. At matched total-sample budgets, SE and FSS are comparable at small budgets; at large budgets, adding seed questions outperforms spending the same budget on more responses. Within FSS, however, neither synthesizing additional questions from the existing seeds nor varying the synthesis protocol outperforms plain RS at matched budgets. FSS is thus a bounded scaling axis and a controlled setting for comparing synthesis protocols. We will release our code and data to facilitate further research.
Existing robust preference optimization for language-model alignment mainly studies pairwise supervision and places robustness at the dataset, prompt, or preference-pair level. We instead study listwise preference optimization under ranking-label uncertainty: given a prompt and a candidate list, the observed ranking over that list may be ambiguous due to annotator inconsistency, near-ties, lossy rankwise feedback, or reward-model noise. We propose a pointwise total-variation robust Plackett--Luce objective that directly robustifies the ranking label conditional on the candidate list. The robust loss admits an exact decomposition into the nominal PL loss plus a worst-case PL correction, and the worst-case ranking is obtained by sorting current implicit scores in ascending order, reducing the inner maximization from $K!$ enumeration to $O(K\log K)$. This tractable structure yields strong offline and online optimization guarantees. In the offline fixed-list setting, the robust objective is convex and projected stochastic subgradient reaches global $ε$-suboptimality with $O(ε^{-2})$ sample complexity. In the online policy-induced setting, where candidate lists are generated by the current policy, we establish weak convexity and $\widetilde O(ε^{-2})$ Moreau-envelope stationarity. Experiments in offline LLM alignment show that the proposed robust correction largely preserves performance under clean labels and improves robustness under noise. In online alignment, it makes reward-model-ranked candidate expansion more reliable and improves both reward-model and external GPT-4 judge metrics.
Sparsely activated Mixture-of-Experts (MoE) language models contain substantial structured redundancy among routed experts, but pruning them without downstream calibration data remains challenging. Existing expert-pruning methods typically rely on a single aggregated importance score, which can bias the retained set toward experts favored by dominant calibration patterns. We propose \textbf{Generic TB-Coverage}, a coverage-aware expert pruning method that uses only generic text corpora (WikiText2 and C4) for calibration. Instead of collapsing expert utility into one score, our method profiles per-expert utility separately on each corpus and enforces a fixed-budget coverage rule that preserves high-utility experts from each corpus before constructing the final pruning mask. Across Qwen1.5-MoE-A2.7B and DeepSeek-MoE-16B-Base at 25\%, 50\%, and 75\% retention budgets, our method improves average accuracy on six common zero-shot benchmarks over random pruning, REAP, and ExpertSparsity, while also reducing perplexity degradation on WikiText2 and C4. The gains are largest under aggressive pruning (25\% and 50\% retain), suggesting that preserving cross-corpus expert coverage is an effective generic-data prior for MoE pruning. Our improvements hold with fixed pruning budgets and no downstream calibration data.
Agents are increasingly used to construct workflows and assist humans in completing recurring tasks more efficiently. As these workflows become repeated and domain-specific, agent memory and reusable skills become increasingly important: agents should be able to recall workflow patterns, execution constraints, and user preferences from previous runs. We study this problem in workflow-based image generation and introduce COMFYCLAW, an agentic skill evolution harness for controlling ComfyUI workflows. COMFYCLAW formulates workflow construction as typed graph editing, exposes tools organized by construction stage, automatically reverts invalid edits, and uses a region-level vision-language model (VLM) verifier to translate visual failures into actionable repair suggestions. The framework further evolves a progressively disclosed skill library, where trajectories, execution errors, and verifier feedback from previous runs are distilled into reusable Agent Skills. Across four benchmark splits, three agent models, and two image backbones, COMFYCLAW achieves the best average image-generation evaluation score across all six agent configurations, outperforming a verifier-only baseline without skill evolution. Human annotations further show that annotators prefer COMFYCLAW over variants without skill evolution. Our results suggest that skill evolution is an effective mechanism for improving agent reliability and performance in recurring visual workflow construction.
Recently, speech classification methods have gained widespread adoption in intelligent gadgets. Current study indicates that backdoor attacks provide a substantial security concern to these models, underscoring the pressing necessity to investigate additional potential attack techniques to expose and prevent such risks. This work discusses the vulnerability of current speech triggers to detection by deep neural network defenders and introduces the Timbre Leakage Attack (TLA). The suggested trigger disseminates timbre information at the frame level within the deep self-supervised features, producing poisoned samples that appear natural to human perception. Furthermore, we introduce Pmeta-TLA, an innovative training mechanism for embedding numerous backdoors one time. This method proposes a multi-backdoor injection training strategy using meta-learning and Projected Conflicting Gradients (PCGrad) and introduces TLA as a multi-target attack tool within it. We performed tests on data-poisoning backdoor attacks in keyword spotting tasks utilizing some deep neural network models. Experimental results indicate that the proposed strategy attains superior Attack efficacy, enhanced stealthiness, robustness, and a reduced attack cost relative to baseline methods.
Solving partial differential equations (PDEs) with high-frequency solutions remains a central challenge in physics-informed machine learning due to spectral bias -- the tendency of neural networks to learn low-frequency components preferentially. This paper proposes a Frequency Shift Physics-Informed Extreme Learning Machine (FS-PIELM) framework that addresses this limitation through an additive mechanism for weight initialization. Rather than multiplying random weights by a scaling factor, the method translates the mean of the Gaussian weight distribution while keeping the variance fixed at unity, thereby avoiding the variance amplification inherent in scaling-based methods. Two variants are developed: FS-PIELM-L assigns independent frequency magnitudes to individual neurons, while FS-PIELM-G groups neurons for improved robustness. Theoretical analysis shows that the frequency variance under the proposed framework remains bounded and approaches unity regardless of target frequency, in contrast to the quadratic growth of conventional approaches. The method preserves the computational efficiency of extreme learning machines, requiring only a single linear solve. Experiments on seven benchmark problems spanning six equation types -- Helmholtz, wave, Poisson, Klein-Gordon, heat, and advection-diffusion -- on both regular and complex geometries show that the linear variant achieves the best accuracy in six of seven cases, with improvements of one to nearly five orders of magnitude over existing PIELM variants. The code and data accompanying this manuscript will be made publicly available at https://github.com/xgxgnpu/Physics-informed-vibe-coding/tree/main/FS-PIELM.
These notes give a proof-oriented introduction to diffusion models from the viewpoint of sampling, tracing a single arc from classical sampling dynamics to modern diffusion samplers, their error analysis, and inference-time control. Throughout, the material is layered into core definitions and identities proved in full, representative estimates proved under simplifying assumptions, and research-level theorems stated with a proof roadmap. The intended audience is beginning graduate students with a background in probability but no prior exposure to stochastic differential equations, stochastic numerics, or diffusion models.
Finetuning a language model on documents that are explicitly annotated as fictional results in a model that still actually believes the documents' core claims, an effect known as Negation Neglect. In our evaluations, models trained on documents prefixed and suffixed with such annotations correctly identify the relevant claims as fictional only about 9% of the time. To address this, we introduce Goggles, a learned module that intervenes on the finetuning gradient rather than the data. During supervised finetuning, a Goggles module edits the gradients an LLM LoRA receives, imparting a chosen epistemic frame (the stance the model takes toward the nature of what it reads) to whatever the documents teach. A Goggles instance is trained once for a given base model, frame, and LoRA configuration, then applied frozen to documents it was never trained on. Trained through Goggles on those same documents, now carrying no fictional annotation, the model flags the content as fictional roughly 91% of the time, while preserving capability (GPQA and TruthfulQA match or exceed baseline). The same architecture supports other frames: a Goggles instance can be trained to treat documents as "part of an AI safety evaluation by Redwood Research" rather than simply as fiction. The imparted frame persists under continued finetuning that pushes back toward the claim, where prior interventions revert. Goggles suggests a path toward training language models on known-misaligned data without absorbing the behaviors that data demonstrates.
Model merging aims to combine existing single-task solutions into a multi-task solution without additional data-driven fine-tuning.~Most existing approaches achieve this using geometric properties of local solution spaces. However, such geometric views provide limited guidance for scoring how statistically useful each task-specific update direction is across tasks during merging. To address this, we formulate model merging from a new perspective of probabilistic inference under a product-of-experts (PoE) scenario where each single-task solution defines an energy-based expert model (EBM) over the merged parameters. We show that several existing model merging methods arise as special cases of our framework under energy designs that impose implicit Gaussian assumptions on directional residuals between merged and task-specific models. Empirically, we find that these residuals are often heavy-tailed which exposes a mismatch with the imposed light-tailed Gaussian structures. We address this with a heavy-tailed PoE design based on Cauchy experts, which better captures the observed residual behavior while admitting a provably convergent inference procedure. Experiments across multiple tasks and architectures show significant improvements over state-of-the-arts baselines. Our code is available at https://github.com/MinhLong210/PoE-EBM-Merging.git.
Foundation models are routinely released to the public, yet the data recipes used to train them -- such as domain mixture weights that determine how different sources are sampled -- are rarely disclosed. This creates an access asymmetry: researchers study the resulting models but lack visibility into the training distribution that produces them. Prior works for inferring training data, such as membership inference, detect at the level of individual samples and thus cannot characterize the global composition of the training corpus. We introduce WARP, a framework that recovers a fine-tuned model's training mixtures directly from its released weights. WARP interpolates between the base and fine-tuned models using model merging, generating pseudo-checkpoints that approximate the missing training trajectory and expose a geometric footprint of the training data in the weight space. From these simulated footprints, WARP extracts geometric features and maps them to domain proportions using either a parameter-free softmax readout or an MLP projector trained on synthetic mixtures. In controlled experiments with BERT and GPT-2, WARP recovers domain mixtures with an average MAE as low as 0.046 and 0.104 respectively, outperforming membership inference and a variant with access to the true training trajectory.
Adversarial attacks on cybersecurity classifiers pose a dual threat: degrading predictions and destabilising the SHAP-based explanations that security analysts rely on to understand and triage alerts. We extend our prior MLP conference study to Random Forest and XGBoost across four tabular security datasets (phishing URLs, UNSW-NB15, NF-ToN-IoT, HIKARI-2021), evaluating five attacks including three black-box methods applicable to non-differentiable tree models. We introduce the Explainability Stability Index (ESI), a scalar metric computed from TreeSHAP attribution drift under adversarial perturbation, reported on the same [0,1] scale as the Robustness Index (RI). A key finding is that gradient-based black-box attacks (ZOO) produce degenerate results against XGBoost (apparent RI ~0.98) due to piecewise-constant prediction surfaces, while score-based Square Attack reveals genuine vulnerability (RI ~0.36). These degenerate perturbations still drive substantial attribution drift: XGBoost ESI ~0.06-0.16 despite near-perfect ZOO robustness, versus 0.14-0.29 for RF, showing that prediction robustness and explanation stability are distinct axes requiring joint measurement. A two-axis framework (gradient dependence, query efficiency) explains the observed attack ranking and yields practical guidance for tree ensemble evaluation. A step-size ablation explains a counterintuitive PGD anomaly on z-score normalised tabular data.
Communication increasingly dominates the cost of Large Language Model (LLM) pre-training, especially under data-parallel and sharded training schemes, where gradient synchronization and parameter reconstruction overhead increase with model size and system scale. Existing communication-reduction methods either sparsify raw gradients, which can be unstable for modern Adam-style optimizers at high sparsity, or quantize communication, whose savings are fundamentally bounded by bit width and often incur additional runtime overhead. We present SCAPE, a communication-efficient distributed optimizer for LLM training that exploits the stability of AdamS's first-moment to enable aggressive sparsification without loss of LLM quality. Instead of constructing masks from raw gradients, SCAPE derives them from first-moment-based statistics, partitions mask generation across workers to align with optimizer sharding, and delays mask usage by one step so that mask synchronization can overlap with computation. SCAPE also reconstructs the quantities required for second-moment updates from a single synchronized sparse buffer, avoiding an additional collective. We implement SCAPE in Megatron-LM and evaluate its convergence by pre-training GPT-345M on OpenWebText and Llama-500M on SlimPajama-6B using 32 NVIDIA GH200 GPUs on TACC Vista. In both models, SCAPE preserves training stability, validation loss, and downstream task accuracy under 90\% and 99\% sparsity. For Llama-500M, SCAPE reduces end-to-end pre-training wall-clock time by up to 43.3\% while maintaining model quality comparable to dense AdamW and AdamS. For Llama-1.8B, SCAPE achieves up to 3.26$\times$ speedup per step compared to dense AdamS.
In multi-source ECG deployment, models may need to incorporate new data sources when earlier raw ECGs cannot be retained or replayed. Freezing a pretrained backbone and assigning each source an isolated classifier prevents parameter interference, but deployment still requires selecting an expert when source metadata are unavailable. We study this distinction through \ours{}, an incremental expert bank built on frozen 1024-dimensional ECGFounder features. Each arriving domain adds a balanced-softmax linear expert, while a lightweight router is fitted only on retained training features and domain labels from sources observed so far. A validation-calibrated margin rule fuses the two most likely experts instead of committing to a single routed expert. On CPSC, PTB-XL, Georgia, and Chapman-Shaoxing, source-aware expert selection reaches $0.7915\pm0.0036$ Macro-F1 and a matched offline independent-head reference reaches $0.7885\pm0.0009$, supporting strong source-aware expert retention. Without source IDs, an MLP router reaches $0.7756\pm0.0027$ and top-2 margin fusion reaches $0.7782\pm0.0022$. The top-2 gain over hard MLP routing is small ($+0.0026$), with a 95\% confidence interval from paired bootstrap that includes zero. Across three domain orders, the top-2-to-oracle gap remains $0.0111$--$0.0133$, identifying autonomous source inference as the main remaining bottleneck. No raw ECGs are replayed, but frozen training features are retained for router updates; the method is therefore not memory-free.
Day-ahead wind power forecasting is essential for cost-effective power-system operation. It is primarily driven by future meteorological conditions while retaining temporal dependencies in power generation. In practice, observed wind-farm power often entangles physically available power with local environmental effects and latent operational states, such as shutdowns and curtailment. Existing physical models provide useful constraints but adapt poorly across wind farms, whereas data-driven models can capture rich correlations but often conflate meteorological effects with state-induced deviations. In this study, we propose UniWind, a wind power forecasting model based on physics-informed state routing. UniWind first employs a Physical Prior Estimator to construct a site-calibrated physical prior by combining site-conditioned monotonic warping with a shared physical power curve. It further applies a physical upper-bound constraint to shape this prior as a soft envelope of available wind power generation. UniWind then proposes a Latent State Encoder to model operating-state embeddings and transforms the physical prior into final power forecasts through a State-aware Power Corrector, which uses knowledge-guided supervised state routing and bounded, state-specific expert correction. Full-shot and cross-farm zero-shot experiments on more than 20 real-world datasets demonstrate the accuracy and robustness of UniWind.
Decentralized online convex optimization (D-OCO) is a popular framework for distributed applications with streaming data. To tackle the communication bottleneck, previous studies have investigated D-OCO with compressed communication and proposed several algorithms that are variants of online gradient descent (OGD). However, for D-OCO with exact communication, the best existing algorithms are variants of follow-the-regularized-leader (FTRL). In this paper, for the first time, we propose two FTRL-type algorithms for D-OCO with compressed communication. Compared with OGD-type algorithms, our algorithms are more elegant in both algorithmic design and theoretical analysis. The key insight is that the dual update mechanism of FTRL allows us to make a simple application of the technique for average consensus with communication compression. More specifically, our first algorithm considers the full-information setting, and can match the existing regret bounds. Our second algorithm is designed for the bandit setting, and can significantly improve both the regret bounds and communication costs of existing algorithms.
Multi-agent systems are increasingly used for forecasting future events, as deliberation among multiple LLMs is believed to improve reasoning and calibration. Yet existing approaches overlook a critical design choice: what information each agent receives. When all agents are given identical evidence, deliberation collapses into herding rather than genuine belief revision, leaving multi-agent systems little better than a single agent. We identify this as a fundamental gap and propose designed information asymmetry to close it: by partitioning evidence into shared public and disjoint private subsets, each agent holds exclusive knowledge that can only reach others through deliberation. We theoretically show that this decomposition reduces inter-agent error correlation, and instantiate it in InfoDelphi, a framework combining relevance-aware evidence routing, rationale-based iterative deliberation, and confidence-weighted aggregation. On PolyGym, a benchmark of 375 binary forecasting questions derived from real-world prediction markets, InfoDelphi outperforms the strongest single-agent and multi-agent baselines by 12--18% in Brier score and 4--8 percentage points in accuracy. More detailed experiments confirm that removing information asymmetry eliminates most deliberation gains, establishing diversity of input as the key enabler of effective multi-agent reasoning.
Hardware impairments in massive multiple-input multiple-output (MIMO) receivers introduce inter-symbol memory and inter-element coupling, severely degrading channel estimation. This paper employs a residual recurrent gated unit (RGRU) to model the intra-slot memory of the hardware impairments and proposes a message-passing-based two-timescale Bayesian deep learning (MP-TTBDL) framework for joint channel and impairment tracking. Owing to small-scale fading, the wireless channel varies rapidly across slots, whereas hardware impairments drift slowly due to hardware aging and environmental variations. To capture these distinct physical timescales, a fastvarying Markov prior and a slow-varying Gaussian Markov prior are assigned to the sparse channel and the network parameters, respectively. Based on a multi-slot factor graph formulation, a message-passing algorithm is developed. Specifically, the inter-slot messages admit closed-form updates, while the intra-slot factor graph, due to its complex recurrent structure, is partitioned into a channel tracking module and an impairments calibration module. The channel tracking module performs sparse channel estimation via turbo orthogonal approximate message passing (Turbo-OAMP), and the impairments calibration module updates the impairment parameters via a specially designed deep approximate message passing (DAMP) procedure, with the two modules iteratively exchanging extrinsic information through expectation propagation (EP) until convergence. Simulation results show that the proposed framework robustly achieves lower channel estimation error than conventional compensators followed by channel estimation across different online impairment scenarios and signal-to-noise ratio (SNR) conditions.
The interaction between brain structure and genetic influences is key to understanding neuropsychiatric disorders. However, most large-scale datasets are unimodal, providing either neuroimaging or genetics data. We propose CALM, a framework that learns interpretable associations between brain ROIs and genetic pathways from completely disjoint populations. CALM aligns the two modalities in a shared latent space via linear projections that simultaneously match the class-conditional latent distributions and ensure group separability. These projections provide interpretable pathway--ROI associations. When trained on unimodal imaging and genetics datasets, CALM generalizes to an unseen paired dataset, outperforming several state-of-the-art methods and ablation baselines. We also demonstrate stability of the learned associations against a paired baseline. Our experiments on autism spectrum disorder reveal immune and metabolic pathways linked to specific cortical regions and are consistent with established literature. Thus, CALM opens the door to leveraging large unimodal repositories for studying cross-modal interactions in brain disorders across disparate datasets.
Data science aims to derive actionable insights from heterogeneous raw data, unlocking the value of the massive amounts of data generated in modern society. Automating this process is essential to reducing labor-intensive efforts for data scientists and enabling scalable data-driven applications. Recently, large language model (LLM)-based data agents have emerged as a promising solution to automate data science workflows. However, the field lacks comprehensive benchmarks to rigorously evaluate these agents across diverse scenarios with fine-grained granularity. To address this gap, we propose AgenticDataBench, a comprehensive benchmark featuring realistic tasks spanning diverse domains with fine-grained ground-truth labels. This enables evaluations to capture the diversity and complexity of data science workflows and the detailed performance of agents. First, to cover diverse domains, we collect real datasets and tasks from 15 vertical domains, including 5 real-world B2B use cases from a leading fintech company. Second, to remove redundancy in real-world tasks and generate high-quality tasks for domains lacking real data, we introduce data science skills, recurring data-centric operational patterns, and quantify benchmark coverage by the number of skills included. Representative skills are extracted from large-scale task solutions on Stack Overflow using skill-aligned hierarchical clustering. Third, for real-world business tasks, we select task-solution pairs that maximize diversity in skill composition, ensuring broad coverage of practical scenarios. Fourth, to generate realistic tasks for devise domains without real tasks, we propose a systematic LLM-based task generation approach to create workflows and tasks based on these skills. Finally, we evaluate state-of-the-art data agents using our annotated benchmark and open-sourced testbed, providing detailed skill-level insights.
State-of-the-art large language model (LLM) training takes tens of thousands of graphics processing units (GPUs) for months and encounters failures across the software and hardware stack. Existing fault-tolerance mechanisms either impose non-trivial overhead during failure-free execution or suffer from prolonged recovery latency, particularly under scenarios where a small subset of compute nodes experience permanent failures. %The tradeoff between failure-free overhead and recovery latency forms a space forms a Pareto frontier We present DeadPool to simultaneously address both optimization objectives. DeadPool incorporates a fault-tolerance mechanism that restores LLM training via hot-swapping, namely by replacing failed nodes with spare nodes without terminating the complete job. The hot-swapping of DeadPool is enabled by two ideas: First, it exploits an off-critical-path in-memory checkpointing mechanism for spatial redundancy. Second, it introduces a communicator reconstruction protocol that replaces failed nodes with spare nodes at runtime. DeadPool efficiently overlaps the in-memory checkpointing with computation, thus introducing zero overhead during error-free execution. Upon permanent node failures, DeadPool can rebuild memory states with minimal recomputation by leveraging in-memory checkpoints. We evaluate DeadPool across scales (up to 512 NVIDIA A100 GPUs) and LLMs (up to 65B parameters), and observe zero checkpoint overhead with hot-swapping recovery completing in under 40 seconds. These results show that DeadPool simultaneously achieves both zero-overhead error-free execution and extremely low recovery cost.
LLM agents increasingly rely on iterative execution to solve tasks through planning, tool use, state updates, and agent collaboration. While this design enables flexible automation, it also creates a new class of failures: an agent may repeatedly execute model calls, tools, workflow transitions, or agent handoffs when the feedback path is not effectively bounded. We call this problem Infinite Agentic Loops (IALs). IALs are not ordinary programming loops; they arise from the interaction between agent logic, framework semantics, runtime observations, and termination mechanisms. Such failures can amplify a single request into long running model and tool execution, causing cost exhaustion, model denial of service, context growth, and repeated external side effects. We propose IAL-Scan, a static analysis tool for detecting IAL failures in real-world LLM agent projects. IAL-Scan abstracts heterogeneous agent code into a framework independent Agent IR, builds an Agentic Loop Dependence Graph (ALDG) to recover explicit and framework induced feedback paths, and checks whether these paths can repeatedly reach costly or state growing operations without an effective bound. We evaluate IAL-Scan on 6,549 LLM agent repositories. It reports 74 potential findings, among which manual review confirms 68 IAL failures across 47 projects, achieving 91.9% precision.
LLM agents are increasingly developed as source-code applications built on agent frameworks. These agent programs combine conventional host-language code with framework-defined semantics for models, prompts, tools, memory, and multi-agent orchestration logic. As a result, their behavior depends not only on traditional control and data flows, but also on a new class of agent dependencies. Such dependencies are often expressed as framework-induced semantics, such as agent constructors, tool decorators, and agent handoff declarations, making them difficult to recover with existing static analysis or dependency tracking tools. In this paper, we present AgentFlow, the first static analysis framework for recovering and analyzing agent dependencies from agent programs. AgentFlow constructs an Agent Dependency Graph (ADG), a framework-agnostic graph representation that represents agents, prompts, models, capabilities, memory states, and control policies as typed nodes, and captures their component-dependency, control-flow, and data-flow dependencies as typed edges. Built on ADGs, AgentFlow supports a range of analyses for agent governance and security, including Agent Bill of Materials (BOM) generation and prompt-to-tool risk detection. We implement AgentFlow for five representative agent frameworks and evaluate it on AgentZoo, a corpus of 5,399 real-world agent programs. Our evaluation shows that AgentFlow recovers richer agent entities and dependencies than existing AST-based agent static analysis tools, generates more dependency-aware Agent BOMs, and uncovers 238 taint-style prompt-to-tool risks in real-world agent programs. These results show that ADG provides a practical foundation for understanding, governing, and securing emerging agent software.
Universal traffic laws describe recurrent patterns in congestion, mobility and driving behavior across cities, providing a scientific basis for transportation planning, management and control. Their discovery, however, remains expert-driven, requiring candidate regularities to be identified from heterogeneous observational evidence or validated through intervention experiments. Although autonomous artificial intelligence (AI) systems have advanced scientific discovery in controlled laboratory settings, extending them to complex transportation domains remains a challenge. Here we present TrafficSci, an agentic AI system that formulates traffic-law discovery as an iterative, auditable workflow integrating evidence scoping, critic-judge hypothesis induction, and observational-interventional validation. Across four case studies spanning population, network, control and trajectory scales, TrafficSci autonomously rediscovers three established traffic laws and identifies an unreported intrinsic temporal memory scale in urban driving behavior, statistically consistent across eight cities and two trajectory datasets. TrafficSci provides a route for extending AI-driven scientific discovery from controlled domains to complex urban systems.
Accurate protein-protein interaction (PPI) prediction is central to functional genomics, disease mechanism discovery, and drug development. A difficult setting arises when candidate interactions include proteins that have no observed PPI edges during training, where models relying on network topology alone often lose useful context. This paper presents \method, a multimodal representation framework for cold-start PPI prediction. \method\ combines region-aware protein sequence encoding with four protein-centered biomedical knowledge graphs, including protein-drug, protein-disease, protein-miRNA, and protein-lncRNA associations. The sequence branch extracts contextual representations from structurally informed sequence regions, while graph attention encoders learn modality-specific protein embeddings from sparse biomedical associations. A bridge reconstruction objective regularizes graph learning by recovering shared protein-entity associations, and a pair-level gating module adaptively integrates sequence and graph evidence for each candidate protein pair. Experiments on two benchmark datasets under novel-old and novel-novel cold-start settings show that \method\ consistently outperforms competitive sequence, network, and knowledge-graph baselines across ACC, F1, AUC, AUPR, and MCC.
Fine-scale rainfall reconstruction is critical for urban flood modeling, but real rainfall sensing systems observe the field through incompatible spatial supports: gauges measure points, microwave links measure paths, and radar/satellite products measure gridded areas. These differences in measurement support impose geometrically distinct constraints on the rainfall field, yet existing heterogeneous graph approaches reconcile such sources in feature space, giving each its own embedding while discarding the geometry of its support. We propose a geometry-aware multi-support heterogeneous graph neural network that represents each observation according to its support type (0D point, 1D line, or 2D grid) as a distinct node layer, and fuses them through cross-support message passing into a point-support prediction layer from which the field is reconstructed. An inductive masked-node formulation decouples prediction resolution from sensing resolution, allowing the same trained model to reconstruct the field at user-defined target locations or display grids. On Singapore data, the proposed method reduces RMSE by 23.2\% over the classical interpolation baseline, inverse-distance weighting, and consistently outperforms other neural architectures such as convolutional fusion and support-agnostic heterogeneous graph baselines. A generalization study using data from Sydney, Australia lets us characterize when multi-support fusion helps: the available skill appears to depend on gauge spacing relative to the spatial correlation length of the field, so fusion delivers the largest gains where the field is under-sampled relative to its correlation length and little when it is already resolved. Code and models will be open-sourced upon paper acceptance.
Training large language models (LLMs) with reinforcement learning (RL) has significantly advanced their performance on reasoning and question-answering tasks. However, prevailing RL reward designs typically prioritize response correctness, neglecting to incentivize models to express their confidence accurately. This leads to a critical problem: performance gains are often accompanied by poor calibration between confidence and accuracy, misleading models to overconfidently hallucinate when uncertain. To address this limitation, we propose $\textbf{C}$orrectness and $\textbf{C}$onfidence $\textbf{C}$alibration $\textbf{R}$einforcement $\textbf{L}$earning ($\textbf{C3RL}$), a novel RL algorithm integrating correctness, calibration and dataset-informed reference accuracy rewards together. Comprehensive evaluation across 8 text and multimodal datasets demonstrates that C3RL enhances calibration without sacrificing accuracy, outperforming the current state-of-the-art method in both performance and calibration metrics. Utilizing the well-calibrated verbalized confidence from C3RL, we further introduce $\textbf{C}$onfidence-based $\textbf{A}$daptive Test Time $\textbf{S}$caling ($\textbf{CAS}$), an adjustable inference-time strategy that allocates computational resources based on response confidence. Experiments show that CAS surpasses majority voting on both in-domain and out-of-domain datasets while reducing the inference budget by up to 12.33 times. We believe the synergy of C3RL and CAS paves the way for deploying more reliable and resource-efficient LLMs. The code, data and models will be released.
Counterfactual explanation (CE) is widely used to enhance the interpretability of machine learning models and support data-driven decision-making based on model predictions. However, existing CE methods typically require two exogenously specified inputs: a desired output value (target) and a distance function that quantifies changes in explanatory variables. In regression settings, neither the validity of target specification nor the practical interpretation of the distance metric has been sufficiently addressed. Furthermore, most existing CE methods focus on altering predictions rather than optimizing a decision objective, even though real-world decision-making often requires explicit objective maximization. To address these limitations, we formulate CE as a profit maximization problem in management and marketing contexts and propose a framework termed profit-based counterfactual explanation (PBCE). PBCE eliminates the need for exogenous target specification by directly maximizing profit as the primary optimization objective. Concurrently, the distance term is reinterpreted as the cost of modifying product attributes, providing a clear and economically grounded interpretation.
Recent advances in Artificial Intelligence (AI) have revolutionized Electronic Design Automation (EDA), particularly through Large Language Models (LLMs) for circuit design tasks. However, their application to analog and mixed-signal domains remains limited by the lack of machine-readable representations of existing circuit design knowledge. Circuit schematic images found in research manuscripts, textbooks, and websites constitute a vast repository of validated designs; however, these visual representations cannot be directly processed by EDA tools. Converting them into machine-readable netlists is essential for enabling simulation, verification, and building comprehensive databases for AI-based models. Current conversion methods lack generalization across both Integrated Circuit (IC) and Printed Circuit Board (PCB) level schematics. Moreover, they struggle with component recognition and connectivity inference, and fail to distinguish between connected junctions and crossing wires. In this paper, we propose SINA, an open-source circuit schematic image-to-netlist generator. SINA is a fully automated pipeline that integrates deep learning for robust component detection, connected-component labeling for accurate connectivity inference, Optical Character Recognition (OCR) for component reference designator extraction, and a Vision-Language Model (VLM) for reliable reference designator assignment. SINA handles both IC- and PCB-level schematics and incorporates dedicated crossing-wires detection to differentiate wire intersections from connections. We validate the correctness of the generated netlists using graph isomorphism techniques. Our experiments demonstrate an overall netlist generation accuracy of 96.67%, which is 2.72x higher compared to state-of-the-art approaches.
SRAM-based FPGAs provide an attractive platform for energy- and latency-constrained CNN inference at the network edge, yet transient faults can lead to silent errors that compromise reliability. Always-on redundancy (e.g., full TMR) improves correctness but incurs substantial performance and energy overhead, while reactive recovery may introduce unacceptable latency on the critical path. We propose \textbf{ProWAFT}, a proactive workload-aware fault-tolerance framework for FPGA-based CNN accelerators that uses partial reconfiguration to selectively apply TMR across reconfigurable partitions. ProWAFT quantifies workload criticality, models fault propagation and reconfiguration overhead, and selects configurations that minimize a composite objective over latency, energy, and reliability risk. Implemented on a Xilinx Zynq UltraScale+ ZCU104 platform with six reconfigurable regions and evaluated on a 500-task trace derived from ResNet-18, MobileNetV2, and EfficientNet-Lite under time-varying SEU injection, ProWAFT achieves lower composite cost than static TMR and reactive reconfiguration while maintaining high task success rate and near-baseline throughput with low online decision overhead.
Large scale document deduplication must preserve semantic equivalence while remaining efficient over massive corpora. We present SemHash LLM, a multi granularity framework that unifies semantic projection hashing, attention weighted MinHash, contrastive boundary learning, and selective LLM based adjudication. The method combines character, token, and document level signals through gated fusion, then applies a cascaded filtering pipeline for efficient candidate reduction. Semantic projection hashing learns compact binary codes in distilled LLM embedding space, while attention weighted Min- Hash suppresses boilerplate and emphasizes informative content. Adaptive decision boundaries and uncertainty estimation further improve robustness across template pollution, short text perturbation, containment, and viral fragments. Experiments show that SemHash LLM achieves strong duplicate detection quality with less than one percent neural verification cost.
As large language models (LLMs) are deployed as communicating agents, does inter-agent communication cause outputs to converge? We introduce BOUNDARY_SYNC, a protocol measuring representational coupling via the Coupling Amplification Factor (CAF = JSD_cond / JSD_baseline), where CAF < 1 indicates homogenization and CAF > 1 indicates diversification. In controlled GPT-4o experiments (N=30, ~9,900 API calls), we measure coupling in text and image communication. Key findings: (1) text communication causes significant homogenization (CAF=0.803 [0.740, 0.873], d=1.30, p<0.001), confirmed by no-communication ablation and prompt-perturbation controls; (2) image communication also homogenizes under within-modality baselines (CAF=0.834 [0.811, 0.858]), with comparable proportional effect; (3) group size moderates coupling direction -- K=5 produces homogenization while K=3 yields CAF > 1.0 (point estimates 1.14 and 1.06, CI pending), suggesting a directional shift toward diversification; (4) cross-model replication shows extreme variation (CAF 0.034-0.803), with DeepSeek dominated by format artifacts; (5) coupling is stateless -- driven by prompt context rather than cumulative updating, with continuous consensus producing monotonic convergence. These results establish LLM agent coupling as real, measurable, and controllable at the prompt level, with direct implications for multi-agent system design.
Modern LLM coding agents are commonly evaluated using pass@k, but developers typically apply a single final patch in real-world settings. This pass@k-to-pass@1 gap is a post-generation problem: a candidate patch pool may contain a correct patch, but the system must decide which one to suggest to developers. Existing post-generation approaches mainly rank whole candidates, filter them with tests, or query an LLM judge, but none deterministically reuse shared edit-atom evidence to both select and construct the final patch. Thus, we propose PatchFusion, a deterministic atomic evidence fusion approach for candidate patches that consults no test outcome at decision time. PatchFusion first fuses whole-diff agreement into a repair neighborhood, selects an auditable representative, and then applies evidence-constrained fusion (ECF) to retain repeated edit atoms and prune unsupported parts. To evaluate this setting, we build PatchFuseBench, a fixed-pool benchmark covering SWE-bench Verified, SWE-bench Multilingual, and Defects4J candidate patches. On PatchFuseBench, PatchFusion solves 426/500 bugs on SWE-bench Verified and 236/300 on SWE-bench Multilingual, and reaches 87/371 plausible patches on Defects4J, outperforming every matched candidate-pool selector on all three. PatchFusion recovers 41 and 27 bugs that no single source solves (30 and 18 more over the best single source). Ablation studies show that ECF adds +5/+6/+9 solved bugs by recovering in-pool repairs that selection misses, with no observed regression, and that PatchFusion's gains remain stable as candidate pools are resampled. On these complementary multi-source pools, cross-candidate evidence recovers more correct patches than the test-based and LLM-based selectors we evaluate, at orders-of-magnitude lower cost, reaching within 96.2% and 89.7% of the candidate-reachable ceiling on the two SWE-bench benchmarks.
As the scale and complexity of cloud-based AI systems continue to escalate, ensuring service reliability through rapid fault detection and adaptive recovery has become a critical challenge. While existing approaches integrate Large Language Models (LLMs) for semantic understanding and Deep Reinforcement Learning (DRL) for policy optimization, they often rely on sequential, loosely coupled architectures that underutilize the generative and reasoning capabilities of LLMs. In this paper, we propose a paradigm shift with PASE, a Planning-Aware Semantic self-healing engine, a novel fault self-healing framework that reconceptualizes recovery as a neuro-symbolic program synthesis task. PASE employs an LLM as a core Plan Synthesis Engine to generate structured recovery plans from a library of semantic primitives. A Neural-Symbolic World Model verifies plan feasibility through simulation, while a Meta-Prompt Optimizer, trained via DRL, learns to generate optimal prompts that guide the LLM's planning process. This tight reason-plan-verify-adapt loop enables dynamic, context-aware recovery strategy generation beyond predefined action spaces. Experiments on a real-world cloud fault injection dataset demonstrate that PASE significantly outperforms state-of-the-art methods, reducing average system recovery time by over 40% and improving fault detection accuracy in unknown fault scenarios. Our framework advances autonomous system management by unifying LLM-based reasoning with model-assisted verification and meta-learned guidance.
Developing high-performance kernels for Neural Processing Units (NPUs) is a critical industry bottleneck, requiring developers to manually navigate implicit hardware constraints and strict memory hierarchies. While large language models offer immense automation potential, they fail catastrophically on NPUs due to a fundamental lack of hardware-specific priors. Naively transplanting code snippets from similar NPU kernels may pass the compiler, but it consistently triggers runtime crashes and performance degradation by blindly violating underlying hardware constraints. To overcome this, we introduce Hawk, a training-free framework that harnesses hardware-aware knowledge through three core modules: (1) Run-Time Knowledge Synthesis Module, which employs a Triple-Part Executable Knowledge Representation to inherently couple the error context with executable semantics; (2) Bottleneck-Aware Knowledge Retrieval Module, which implements a 2D-Retrieval paradigm to project queries into orthogonal syntactic and hardware-aligned semantic spaces; and (3) Effect-Driven Knowledge Distillation Module, which leverages LLM-driven semantic arbitration to continuously distill the knowledge by pruning errors and consolidating redundancies based on the empirical execution feedback. Extensive evaluations on real-world NPU workloads demonstrate that Hawk elevates generation accuracy from 49.4% to 80.0%, while achieving up to a 2.2x execution speedup over state-of-the-art baselines.
Vision-language-action models (VLAs) have recently advanced robotic manipulation, yet the effects of different robot-data pre-training paradigms remain difficult to compare because existing models often differ in architecture, data, action space, and evaluation protocol. We present VLAFlow (Vision-Language-Action Flow), a unified flow-matching framework for controlled comparison of VLA training objectives. Using a heterogeneous robot corpus, OXEMix, containing approximately 5,000 hours of data from DROID, OpenX-Embodiment, OpenX-Augmented, and RoboCOIN, we evaluate four paradigms under the same pi0-style architecture, shared VLM backbone, action expert, and 14-dimensional action space: action-only modeling (MindPI), language-supervised co-training (MindLPI), future latent alignment (MindWPI), and their combination (MindLWPI). Experiments on LIBERO, LIBERO-Plus, and SimplerEnv show that action-only pre-training is sensitive to heterogeneous data. In contrast, language supervision helps preserve vision-language generalization, while future latent alignment improves state-transition and action-outcome modeling. By combining both signals, MindLWPI achieves the most stable overall transfer performance across benchmarks. These results suggest a meta-action space view: language and future latent representations provide complementary intermediate constraints that make heterogeneous action supervision smoother and more transferable.
Predicate invention (PI), the creation of new predicates to extend the hypothesis space, remains a critical bottleneck in Inductive Logic Programming (ILP). Existing methods rely on domain expertise and produce semantically opaque predicates, hindering adaptation to unfamiliar domains and cross-task reuse. We present ADVENT, an LLM-driven PI mechanism for ILP. ADVENT pairs LLM abductive generation with Prolog deductive verification, forming an iterative loop in which concrete execution results guide the LLM to refine candidate predicates. The mechanism leverages Large Language Models to identify implicit patterns in structured relational data and invent auxiliary predicates with meaningful names and definitions. Invented predicates and learned rules accumulate in a knowledge pool for cross-task reuse. Experiments on nine poker-hand concepts across seven LLMs show that LLM-driven PI achieves 58% success rate where ILP alone fails entirely, formal verification raises this to 80%, and the knowledge pool yields gains up to +31 percentage points, while producing human-interpretable rules. These results suggest that ADVENT offers a promising direction for automating predicate invention and enabling cross-task knowledge reuse in ILP.
Large language models have recently been explored for scientific hypothesis generation, but most prior work relies on unstructured literature and free-form textual claims. We present a pipeline for Earth observation that grounds hypothesis generation directly in the NASA Earth Observation Knowledge Graph. A heterogeneous graph neural network trained on historical co-usage relations ranks candidate dataset pairings, and a three-agent LLM pipeline filters, generates, and evaluates structured research hypotheses. Applied to 1,475 NASA datasets, the system produces 160 hypotheses spanning multiple Earth-science domains, including ecohydrology, glaciology, aerosol--cloud interactions, vegetation phenology, and stratospheric chemistry. Model-predicted novel dataset pairings are rated nearly as plausible as held-out real co-usages from the literature, indicating that the pipeline surfaces scientifically coherent yet unexplored combinations. A 2*2*2 factorial experiment across GPT-5.2 and Claude Sonnet 4.6 shows that hypothesis rankings remain stable, while absolute scores depend strongly on judge identity, highlighting limitations of single-judge LLM evaluation.
The capacity of Large Language Models (LLMs) to reason about pedagogical intent within instructional communication remains underexplored, particularly in educational domains such as translation pedagogy. To address this, we propose the \textbf{Adaptive Pedagogical Vigilance (APV)} framework, a novel computational formalism that reframes communicative vigilance as an adaptive mechanism for optimizing learning through intent inference. APV formalizes the problem via a Bayesian Pedagogical Intent Inference Engine (PIIE), which models how instructors select content to maximize pedagogical utility and how vigilant learners should inversely reason about latent instructional configurations -- encompassing genre, stance, and incentives. We evaluate APV through a three-tier hierarchy: distinguishing instructional genre, reasoning about structured pedagogical setups, and generalizing to authentic educational discourse. Experiments on leading LLMs (e.g., GPT-4o, Claude 3.5) show that APV substantially improves model vigilance. It achieves the strongest discrimination between pedagogical and exposure-based content, correlates highly with human judgments ($r=0.958$), and maintains robust performance on naturalistic data where baseline methods degrade. This work establishes a unified framework for assessing and enhancing LLMs' understanding of pedagogical motives, advancing the development of more reliable AI-assisted learning systems.
Chain-of-thought (CoT) reasoning enables large language models (LLMs) to solve complex problems by generating intermediate reasoning steps. While much attention has been paid to the length and content of these reasoning chains, far less is known about their internal geometry. We study the \emph{geometry} of CoT trajectories in the hidden state space of transformer models, formalizing each reasoning chain as a discrete curve in $\mathbb{R}^d$ and characterizing it through spectral, positional, and kinematic geometric functionals. We introduce the effective dimension $d_ρ$ as a measure of trajectory complexity and show theoretically that trajectories with flatter eigenvalue spectra correspond to harder tasks, as they explore more of the hidden dimensions. Lastly, we explore how kinematic features of the trajectory, mean position, positional dispersion, initial and current hidden states, mean velocity, mean speed, and speed dispersion, can be used to predict solution correctness before generation is complete, and may inform future early-stopping strategies. Experimentally, on mathematical reasoning problems from the MATH500 dataset, $d_ρ$ achieves $0.93$ AUC in distinguishing easy from hard problems, while kinematic features potentially can predict correctness from only the first $20\%$ of generated tokens. These correctness signatures transfer across questions of varying difficulty, establishing that the shape of a model's internal reasoning trajectory is a principled window into both task hardness and solution quality.
Deceptive behavior in LLMs is costly to monitor and prevent, motivating approaches such as Scalable Oversight via Lie Detectors (SOLiD) (Cundy & Gleave, 2025), which uses lie detectors to identify responses for review by high-cost labelers. In this paper, we scale SOLiD to larger models and evaluate it in more diverse and realistic preference-learning settings. We find favorable scaling: undetected deception drops from 34% for 1B-parameter models to 14% for 405B-parameter models at a detector true positive rate of 99%, and expensive human labelers can be removed entirely from the fine-tuning phase without a statistically significant increase in deception. However, SOLiD is sensitive to distribution shift between detector training and preference-training data, which can drive detector false positive rates to impractical levels.
Classical software effort distribution models, including the PNR family and Parr alter native curve, were designed to describe the time distribution of development effort under an implied staffing pattern. Their direct use in agile environments is limited when team capacity is fixed, partially fixed, or externally constrained, the original curve may prescribe a staff demand that the organization cannot allocate. This paper proposes a compact refactoring of Parr model as a capacity-aware forecasting layer for agile projects. The contribution is deliberately narrower than a full causal theory of project dynamics. A normalized Parr shaped latent effort demand is combined with an observed or planned capacity trajectory. The resulting model forecasts aggregate progress, completion time, capacity deficit, and capacity slack without assuming that the same internal activity path is followed under resource restriction. The model uses a small parameter set such as total effort K, a Parr shape parameter, an origin constant c that can match nonzero initial staffing, and the capacity trajectory. A discrete sprint formulation is provided, together with a calibration method from ordinary Scrum records and a rolling origin validation protocol against simple management baselines.
Large Language Models (LLMs) often struggle with persuasion in high-stakes scenarios. People's individual personalities and concerns require tailored strategies rather than a one-size-fits-all approach. To address this challenge, we focus on a fire-rescue scenario in which an operator must persuade a resident to evacuate as a high-stakes persuasion domain and propose Dialogue Policy Selection (DiPS), a Q-learning framework to dynamically select persuasion strategies adapted to the evolving conversational context. Specifically, we train a critic, trained to maximize the chance of evacuation success, to select a persuasion policy at each turn based on the resident's recent utterances.We then evaluate DiPS against multiple baselines in both simulated and real human interactions. We find that DiPS achieves higher evacuation success than a zero-shot LLM and generic RAG-augmented approach.
Transformers have become general-purpose architectures, but their all-to-all self-attention is poorly matched to graph data, whose interactions are sparse, structured and multi-scale. Existing Graph Transformers address this mismatch through structural encodings, hybrid message-passing modules or learned attention constraints, often introducing additional complexity and limited interpretability. Here we introduce X-LogSMask, an explainable multi-head logarithmic structural mask that injects symmetrically normalized graph topology directly into attention logits. The logarithmic transform converts structural connectivity into a topology-aware gating signal, suppressing unsupported node interactions while preserving feature-dependent attention. By assigning different powers of the normalized adjacency matrix to different attention heads, X-LogSMask gives each head a defined structural radius and supports multi-hop information propagation within a single layer. We further show that a standard Transformer encoder can be interpreted as one-step message passing on a complete graph, motivating X-LogSMask as a topology-constrained alternative to unrestricted self-attention. Across 20 node-, edge- and graph-level benchmarks, Transformers equipped with X-LogSMask achieve state-of-the-art performance on 13 datasets and remain competitive in a lightweight one-layer configuration. These results show that simple, interpretable structural masks can make self-attention an effective graph-learning operator without changing the Transformer architecture. The code is available at https://github.com/LiLeyan-0120/X-LogSMask.
Large language models are increasingly used as open-ended search operators in evolutionary optimization. We introduce Evolutionary Feature Engineering (EFE), a framework for using LLM-based evolution to discover preprocessing transformations for structured data. EFE represents transformations as Python programs with a standardized fit/transform interface, allowing them to be inserted directly into existing machine learning pipelines. During evolution, candidate programs are refined using dataset context, summary statistics, and downstream performance feedback on validation set. We instantiate EFE in two settings. For time-series forecasting, EFE-Time learns invertible, dataset-specific normalizations that improve off-the-shelf time-series foundation models. It reduces forecasting errors (MASE, WQL, MAE) 3% or more when averaged across datasets and improvements are as much as 19% on the COVID-Deaths dataset. Notably, these improvements occur with recent TSFMs such as Chronos-2. For tabular prediction, EFE-Tab evolves compact feature programs that add useful interpretable features and remove redundant ones, improving or matching existing LLM-based feature-engineering methods. We found EFE-Tab to be particularly effective on classical decision trees, where small sets of evolved features yield competitive accuracy while preserving interpretability. Overall, EFE demonstrates that LLM-based evolution can improve both accuracy and interpretability when automatically tackling structured data.
This paper studies whether the Metabolic Multi-Agent Optimizer (MMAO) can act as a credible outer-loop optimizer for classification model selection. We propose MMAO-Cls, a mixed-space realization in which each agent jointly encodes a binary feature mask and classifier hyperparameters, while private energy, communal budget, role drift, and lifecycle turnover are mapped to the accuracy-complexity tradeoff of wrapper learning. The implementation is strengthened by deriving feature-budget adaptation from feature-information priors and by regularizing validation reward with both subset compactness and train-validation overfitting gap. We evaluate MMAO-Cls on seven standard tabular benchmarks with three seeds each and compare it against RandomSearch, GA-lite, PSO-lite, and an endogenous no-sharing ablation. On the aggregate validation objective, MMAO-Cls ranks second ($0.9433$) behind GA-lite ($0.9446$). On held-out test performance, it reaches mean score $0.8882$, improving over RandomSearch ($0.8808$) and GA-lite ($0.8857$), remaining close to PSO-lite ($0.8874$) and the no-sharing ablation ($0.8900$), while using the most compact mean held-out feature subset among all compared methods (feature ratio $0.4881$). Pairwise tests show that these margins are not yet statistically significant. The resulting claim is therefore conservative: MMAO-Cls supports classification applicability and compact mixed-space search more clearly than it isolates communal sharing as a decisive standalone advantage.
Language models (LMs) raise an intriguing alternative to vector-based retrieval: conditioning on an in-context corpus and directly generating a relevant answer. However, prior work has largely focused on proprietary systems or the smaller-scale reranking task, leaving corpus-scale in-context retrieval largely unexplored. In this work, we present the first systematic study of in-context retrieval on two scales practical retrievers demand: million-token corpora and length-generalization far beyond training-time sizes. We first introduce BlockSearch, a 0.6B LM retriever whose architectural and training modifications improve over prior LM baselines and length-generalize up to 10 times beyond its training regime. Nevertheless, retrieval still collapses under more extreme extrapolation. We trace this failure to an attention dilution effect: as the corpus grows, irrelevant documents dominate the softmax denominator, reducing the normalized mass on the gold document even when its pre-softmax score stays high. Motivated by this analysis, we introduce length-aware adjustments to the attention softmax and document-level sparse attention. With these modifications, at the million-token scale, our model matches dense retrieval on widely studied benchmarks (e.g, MS MARCO and NQ), while outperforming the concurrent model MSA despite being 7 times smaller. Furthermore, it significantly outperforms dense retrieval on tasks requiring entirely different notions of similarity, such as LIMIT, achieving a 3 times higher score. Together, our results position in-context retrieval a promising alternative to classical retrieval while emphasizing attention control under extreme context growth as a new challenge.
Certified world models estimate how long their predictions remain valid. We turn this validity horizon into an operational sensing clock: a rule for when an agent should stop coasting and re-sense. Starting from an audited equivariant world model, we derive a deadline for no-sensing intervals and show that deployable deadlines in learned world models must be drift-aware: on-manifold Lyapunov rates alone overestimate coasting validity, while calibrated native rollout-drift envelopes carry the deployed guarantee. On a frozen 3D VN-JEPA model, the resulting clock controls held-out interval-simultaneous certificate violation across seeds and data shards. In a cue-conditioned theorem-bed (a synthetic bench where all schedulers share the exact model, isolating the scheduling rule), the clock remains valid on the deployment distribution and substantially reduces eventful-tail violations relative to exact-mixture expected-belief scheduling at matched sensing budget. We also report limits: in the short-horizon frozen VN-JEPA regime, empirical conformal horizons match the deployed clock on validity and budget, and a partial-reset exploration finds no clean budget-matched advantage for the spectral term. Thus the contribution is a certified sensing-clock primitive and drift-aware deployment method, not a claim that spectral clocks empirically dominate all non-spectral schedulers.
Learning how an environment behaves from interaction is central to building agents that adapt to unfamiliar tasks. World models learned with deep networks are flexible but data-hungry and transfer poorly beyond their training distribution. Program-synthesized world models, written as source code by LLMs and refined through counterexample-guided inductive synthesis (CEGIS), are instead data-efficient and reusable, yet they have been demonstrated mainly on structured-state worlds with a given object vocabulary, and a single program search does not scale to pixel-rendered environments whose object structure must be hypothesized flexibly. We introduce OPINE-World, an LLM agent that learns an object-centric programmatic world model online from interaction. OPINE-World couples two cooperating agents in a loop of hypothesis and test, one acting in the environment and one synthesizing the model in code with replay verification and model-based planning, and it steers exploration with a Bayesian measure of object-type adequacy we call ontology error. We evaluate OPINE-World on ARC-AGI-3, a benchmark for skill-acquisition efficiency in which the object vocabulary, the goal, and the action semantics are withheld. OPINE-World solves 20 of 25 games without per-game training and reaches an action-efficiency score of 78.4 against the human baseline.
Understanding user intent is fundamental to delivering relevant search results in e-commerce. However, substantial fraction of real-world queries are under-specified (e.g., "watch" or "shirt"), lacking explicit attributes such as gender or age group. This ambiguity poses a significant challenge for query intent detection models in e-commerce search systems, which must accurately infer latent user intent (e.g., age, gender) to support effective downstream retrieval. We introduce IntentTune, a framework for resolving ambiguous or under-specified query intents by leveraging either (1) user-specific behavioral signals including search history, browsing activity, and profile attributes or (2) population-level demand patterns aggregated across all users. Through experiments on real-world e-commerce data, we first demonstrate that population-level demand patterns alone are insufficient to reliably infer intent in under-specified queries. We then demonstrate that user-specific behavioral signals -- particularly prior search queries -- outperform both population-level statistics and static profile information for inferring gender, age group, product category, and size intent from underspecified queries.
Small multirotor aircraft are increasingly tasked with operations in the atmospheric boundary layer, where turbulent winds comparable to the vehicle's airspeed degrade trajectory tracking and can defeat conventional feedback control. This work illustrates a two-stage learning pipeline that first estimates the local wind from onboard kinematics and dynamics and then exploits that estimate inside a reinforcement learning (RL) flight controller. The wind estimator, an attention-augmented gated recurrent network trained on thousands of simulated flights through von Karman turbulence with power-law shear and veer, recovers the horizontal wind vector with a per-flight root-mean-square error of 0.40 m/s and a direction error of 3.2 degrees on unseen wind regimes, an accuracy near the floor imposed by unresolved turbulence, and generalizes to vertical ascent profiles with a skill score of 0.861 over a constant-wind reference. A proximal policy optimization controller receiving the frozen estimator's output reduces horizontal trajectory tracking error by 48% relative to a wind-blind proportional-derivative baseline across mean winds of 4 m/s to 12 m/s, winning on 100% of evaluation episodes. A three-way ablation decomposes this improvement into a kinematic component, available without wind information, and a wind-perception component; the perception share rises with wind speed, from small in light winds toward roughly half the total benefit in strong winds, consistent with the quadratic scaling of aerodynamic drag. The controller degrades gracefully on out-of-distribution winds of 13 m/s to 15 m/s, where the baseline fails catastrophically.
Room embeddings derived from reverberant speech are often unreliable: speech content and recording degradation can alter the representation even when speaker, room, and source-receiver geometry remain unchanged, degrading downstream task performance. We propose a framework that learns room embeddings robust to speech-content variation and a representation-level uncertainty score from reverberant speech without downstream-task supervision. The embedding is anchored to a structured room impulse response (RIR) latent space and trained using a multi-view data structure with Kullback-Leibler (KL)-based alignment; a multi-positive contrastive term further refines robustness. A lightweight uncertainty head is calibrated using the dispersion of corruption-induced embeddings and optimized with a rank-based objective. Across waveform- and spectrogram-level corruptions, the score is consistent with representation dispersion and enables effective selective prediction while requiring only a single utterance at inference.
This monograph provides an introduction to mean field reinforcement learning through the lens of Markov decision processes arising from large-population stochastic control with mean field interactions and common noise. Starting from the connection between multi-agent reinforcement learning and mean field control, it develops the probabilistic, mathematical, and control-theoretic framework needed to formulate representative-agent learning problems, analyze their relationship with finite-population systems, and study both general and linear-quadratic models. The presentation includes dynamic programming principles, propagation-of-chaos limits, and theoretical analyses of tabular Q-learning and policy-gradient methods. It also discusses numerical implementations, including tabular schemes and deep reinforcement learning methods such as deep deterministic policy gradient. The goal is to give readers a coherent bridge between mean field control theory and reinforcement learning methodology, emphasizing the mathematical structure of the problems and the design of tractable learning approaches for large stochastic populations.
Recurrent memory agents extend LLMs to arbitrarily long contexts by iteratively consolidating input into a fixed-size memory window. Despite their scalability, these agents exhibit a well-documented reliability problem: end-to-end performance degrades systematically as context length grows. We diagnose this failure by decomposing performance into two factors--memory capture and memory retention--and quantitatively confirm that retention is the dominant bottleneck. Retention collapses because existing designs maintain memory as a monolithic text block, forcing every update to risk overwriting previously retained content. Motivated by this diagnosis, we propose Multi-Head Recurrent Memory (MHM), a general, training-free framework that partitions memory into independent heads governed by a stage-wise select-then-update strategy. At each step, exactly one head is selected for update while the remaining heads are structurally shielded from overwriting, shifting the burden of retention from model behavior to architectural design. As a lightweight instantiation, we introduce Least-Recently-Updated MHM (MHM-LRU), which guarantees uniform head utilization with zero additional token overhead. Extensive experiments on long-context benchmarks show that MHM-LRU substantially improves both retention and end-to-end accuracy across the 100K--1M token range, where baselines degrade sharply. On RULER-HQA at 896K tokens, MHM-LRU improves the memory retention rate from less than 30% to 73.96%. These gains generalize across model families, scales, and task types, positioning architectural optimization as a practical and cost-efficient path toward reliable long-context recurrent memory.
Mode shape recognition is a fundamental task in automotive NVH development, yet it remains dependent on manual visual inspection by experienced engineers. Existing approaches based on engineering heuristics, Modal Assurance Criterion (MAC), or geometry-dependent AI representations often exhibit limited robustness across different vehicle architectures, finite element (FE) meshes, and experimental measurement layouts, restricting their industrial applicability. This paper presents a Canonical Engineering Graph Representation and region-aware graph learning framework for robust and explainable 3D mode shape recognition. Rather than learning directly from vehicle-specific FE meshes, heterogeneous FE models and experimental measurements are transformed into a common graph whose nodes represent semantically meaningful structural regions connected through engineering-informed relationships. Geometry-independent regional descriptors are combined with graph attention learning and region-aware pooling to capture structural interactions while preserving engineering semantics and enabling physically interpretable predictions. The resulting representation decouples engineering knowledge from numerical discretization, allowing transfer across different vehicle programs without requiring identical mesh topology or sensor configurations. The proposed framework is validated using FE and experimental datasets from four vehicle programs under severe label scarcity. Results demonstrate high classification accuracy, cross-vehicle transferability, and physically meaningful explanations by directly relating predictions to engineering-defined structural regions used in NVH analysis. Beyond mode shape recognition, the proposed Canonical Engineering Graph Representation provides a reusable engineering abstraction for trustworthy and transferable AI across heterogeneous simulation and experimental workflows.
Transformer inference on long sequences is expensive because softmax attention repeatedly reads from a large KV cache. The prevalent approach to this bottleneck is KV cache compression, which replaces the full cache with a compact summary. Despite its practical importance, the design of such summaries is largely driven by empirical experimentation. On the theoretical side, existing results show that KV cache compression can be impossible in the worst case, but offer little systematic guidance for designing algorithms in regimes where accurate compression is possible. We bridge this gap by characterizing the minimax risk of KV cache compression in terms of the intrinsic compressibility of a cache, revealing when and how accurate compression is possible. These results yield novel design principles for KV cache compression under causal masking that map efficiently to prefill and autoregressive decoding while achieving minimax-optimal risk. We instantiate these principles in a practical algorithm and report promising performance on LongBench in targeted experiments. Overall, our results provide a principled avenue for practical KV cache compression with theoretical guarantees.
How far can a language model improve under a strict artifact budget? Parameter Golf posed this question as an open community challenge in which participants trained the best language model, with the complete artifact (training code + compressed weights) required to fit within 16 MB and be trained in under ten minutes on 8xH100 SXM GPUs. Quality was measured in bits-per-byte (BPB), the average number of bits required to encode each byte of unseen text. We analyze 2,037 pull requests and 1,430 clean scored submissions from the contest, build a taxonomy of 84 optimization techniques, and measure each technique's contribution to BPB. The verified leaderboard score dropped from 1.2244 to 1.058 BPB across three phases -- a 13.6% reduction, despite individual techniques rarely improving BPB by more than 1%. We show that most gains in techniques shrink across competitive submissions, isolating the few methods that improve performance across stacks.
Chain-of-thought (CoT) reasoning has emerged as an effective approach for activating latent reasoning capabilities in large language models. However, most existing CoT methods use reasoning chains mainly as inference-time prompts, while the generated reasoning traces are rarely reused as semi-supervised learning signals. In this report, we define \textbf{Semi-supervised Chain-of-Thought Learning} and propose \textbf{Semi-CoT}, a simple framework that uses unlabeled questions to construct pseudo reasoning supervision. Semi-CoT samples multiple pseudo-CoTs for each unlabeled question, estimates answer-level semantic entropy, and selects low-entropy reasoning chains as reliable pseudo-CoT demonstrations. This extends the self-training view of CoT from inference-time refinement to semi-supervised pseudo-supervision. Pilot experiments on AQuA, SVAMP, GSM8K, and MultiArith show that the entropy gate selects high-precision pseudo-CoTs, with pseudo-answer precision ranging from $91.36\%$ to $100\%$. Semi-CoT also gives small gains on SVAMP and GSM8K, while AQuA shows negative transfer and MultiArith reaches a ceiling. These results suggest that unlabeled questions can provide reliable pseudo reasoning signals, but their effective use still requires stronger demonstration selection or student training.
AI agents that autonomously execute tool calls on a user's behalf raise pressing questions about permission management: what role could users play, and what role should they play? Despite many proposed approaches, the user's role in agentic permission management remains under explored. We introduce Janus, a playground system for implementing and evaluating user-involved agentic permission management designs. Janus consists of two components: Janus-Core, a modular agentic system supporting a diverse spectrum of permission management designs, and Janus-Harness, an automated evaluation framework. Grounded in a conceptual model that identifies key design axes for user involvement, we implement six permission assistants spanning the design space and evaluate them across three scenarios and three synthetic responders. We demonstrate that user input is critical and can significantly strengthen privacy and security, that AI augmentation of user decisions can help reduce cognitive load, and that realistic user behavior including permission fatigue must be accounted for in system design. No single design performs optimally across all contexts, motivating a more principled and context-sensitive approach to deploying permission assistants in agentic systems. Janus is publicly available to support future investigation into this dimension of agentic system design.
Empirical research rarely admits a unique analysis. Different analytical choices can lead to different conclusions from the same data, yet these hidden forking paths are difficult to observe. We show that AI agents capture much of the analytical variation among human researchers while making these paths explicit. Across four high-stakes domains, assigning different personas is sufficient for AI agents to report divergent, often opposing, conclusions from the same data and question, with findings systematically aligned with those beliefs. In a study in which 42 human research teams analyzed the same immigration dataset, AI agents reproduced 72% of the human ideological gap in reported effect estimates. Despite reaching opposing conclusions, it is difficult to identify clear issues in each analysis based on the final AI reports: 86% passed independent AI review and 78% passed majority human expert review. These findings suggest that the central challenge is often not flawed analyses, but selective exploration and reporting from a large space of methodologically defensible analyses. AI agents may amplify this longstanding problem by making such exploration inexpensive and scalable. To address this, we introduce the m-value (multiverse value), the probability that an analysis path would produce a claim at least as extreme as the reported one. We further introduce Agentic Bootstrap, which estimates the m-value by using AI agents to sample plausible analysis paths. Applied to the human immigration study, 13.5% of reported human analyses fell in the most extreme 5% of the analysis space (m<0.05). Scientific evidence should therefore be evaluated not only by a single reported analysis but also by its position within the distribution of analyses that could reasonably have been reported. Agentic Bootstrap makes this distribution observable and turns it into a criterion for scientific credibility.
Rust's ownership type system prevents memory errors in safe code, but certain desirable properties remain orthogonal to compilation: the soundness of unsafe operations (e.g., raw pointer dereferences), functional correctness, and absence of runtime panics. We present Kani, an open-source model checker for Rust that pushes bounded model checking beyond bug-finding to provide correctness guarantees for these properties. Kani compiles proof harnesses from Rust's Mid-level Intermediate Representation (MIR) into CBMC's bit-precise verification engine, automatically checking a comprehensive set of safety properties with no user annotation. To extend verification from bounded to unbounded, Kani provides a specification language comprising function contracts, loop contracts, quantifiers, and function stubbing. We demonstrate feasibility through case studies on industrial Rust projects, where contracts upgraded verification from panic-freedom to functional correctness, uncovering six previously unknown bugs. Kani operates at scale in production CI, with over 16,000 harnesses verified per code change in the Rust standard library verification campaign.
Recent advances in automatic speech recognition (ASR) have explored different sequence models, including Conformer-based models and newer state space models such as Mamba. Although prior work has evaluated these architectures in multiple languages, their effectiveness in African languages remains underexplored. In this work, we evaluate Mamba for ASR on seven South African languages. In monolingual experiments, each model is trained on 50 hours of speech per language, and we compare Mamba to a Conformer baseline of similar parameter scale. Mamba achieves similar recognition accuracy to Conformer while using fewer computational resources and training faster. We further evaluate generalization in this setting and find that both models struggle to generalize to speech that is much longer than what they were trained on. We then study multilingual ASR using Mamba models, where the baseline is pooling all languages together. On top of this, we tested three extensions: training with language-family information by adding both language and language-family embeddings as biases to the downsampled acoustic representations, and multitask learning with a CTC ASR objective and a language identification (LID) head. We find that multilingual training consistently improves performance over monolingual training. However, adding explicit language information does not improve in-domain performance but does improve cross-corpus robustness. We conducted ablation studies in low-resource multilingual settings using 5-hour and 10-hour per-language training data, where we observed gains from using language embeddings and further demonstrated that removing or altering them hurt model performance. Lastly, we analysed these embeddings and find that they do not capture linguistic similarity in a typological sense, but instead act as task-specific control vectors.
We investigate the problem of learning useful policy representations (embeddings) in two-player zero-sum imperfect-information games. We make three contributions: First, we introduce methods of creating datasets of policies for a given game. Second, we propose methods to learn policy representations. Third, we introduce downstream tasks to evaluate the effectiveness of such representations. We evaluate each dataset method, embedding method, and downstream task on Kuhn and Leduc Poker. Although our methods are very basic, we demonstrate that useful behavioral representations are present in the learned embeddings. To our knowledge, this work is among the first to systematically compare self-supervised learning techniques for learning policy representations in games. Our code is available at https://github.com/VitamintK/ssl-project for others to extend.
Matter seeks to resolve longstanding interoperability problems in the Internet of Things (IoT), yet little is known about how developers experience the standard in day to day work. This paper examines over 13,000 issues from the official Project CHIP GitHub repository to understand the kinds of problems contributors report when implementing and integrating Matter. Using topic modeling and qualitative analysis, we identify four recurring areas of concern, Testing, Interoperability, Development, and Platform and Network, and describe how they manifest in the evolution of the codebase and tooling. The findings reveal systematic technical and integration challenges and point to concrete opportunities to refine Matter's test infrastructure, cross vendor guidance, and documentation as the standard continues to mature.
Recent work has established a fundamental trilemma between Byzantine robustness, local differential privacy (LDP), and optimization error in distributed learning. We show that this trilemma does not universally extend to generalization error, but instead depends critically on the privacy regime. Specifically, in the high-noise regime (strong privacy), we prove that increasing privacy reduces the generalization error, i.e., there is no tension between robustness and privacy. In the low-noise regime (weaker privacy), however, the tension between robustness and privacy reappears and increasing privacy indeed degrades generalization. Our theory explains this surprising non-monotonic behavior of the generalization error via matching lower and upper bounds on the algorithmic stability of Byzantine-robust distributed learning under LDP constraints. We corroborate and further analyze these theoretical findings with empirical evaluations.
Reinforcement learning post-training dramatically improves LLM reasoning, but suffers from training instability and diversity collapse. Advantage functions offer an appealing fix: they reshape the training objective, reweight which rollouts drive learning, and are trivial to implement. Yet a proliferation of methods makes it unclear which advantage to use and when. We cut through the confusion with a unifying framework that decomposes any advantage into its positive and negative gradient mass along two orthogonal axes. On the sign axis, imbalanced updates collapse either entropy or weight geometry. On the difficulty axis, hard-problem focus sharpens signal but costs sample size. Both trade-offs shift during training: exploration favors balance and hard focus; exploitation favors suppression and medium focus. This motivates FADE (Focal Advantage with Dynamic Entropy), a self-adapting advantage that reads training dynamics to schedule the gradient weight automatically. FADE reaches peak pass@1 20k steps earlier than the best static baseline at the 7B scale and 2k steps earlier at the 32B , while achieving the best accuracy-diversity trade-off across all pass@k on LiveCodeBench and AIME.
We propose a scaling law that takes into account model size and training data while explicitly splitting the latter into training steps and batch size (called three-term law). Fitting the proposed law on a large set of training runs, we find that it correctly recovers the scaling of the optimal batch size. Moreover, because it makes use of training runs with suboptimal batch size, our proposed law can be robustly fit with a significantly smaller amount of training runs. We further show that the three-term law can be used to derive scaling laws for suboptimal batch sizes, and that it matches previous empirical findings related to the critical batch size.
We present a fully unsupervised Fast-Slow DSVDD detector for continuous State-of-Polarization monitoring on a deployed subsea cable. Trained without event labels, it ranks all five confirmed trawler contacts within the top 13 of 122,174 recordings and surfaces additional corroborated cable-contact events.
Reinforcement learning with verifiable rewards (RLVR), along with recent selfdistillation variants such as SDPO, evaluates each rollout against a verifier and updates the policy from that episode-level signal. However, the richer procedural information in the rollout is rarely retained or reused. Across episodes and epochs, the model repeatedly encounters related problems under a changing policy, producing cross-episode signals that episode-local updates cannot capture: which strategies consistently pass verification, which failure modes persist, which patterns recur. We propose Procedural Memory Distillation (PMD), which converts these crossepisode signals into reusable procedural memory and distills it into the policy's weights during training. This memory functions as a training scaffold, absorbed into the policy itself, yielding a memory-free model at inference. PMD organizes the memory at three levels of abstraction: raw trajectories, self-reflected strategies and lessons, and higher-level behavioral patterns that recur across problems, all extracted online from the model's own trajectories. A memory-conditioned self-teacher draws on the accumulated experience to supervise the student on its own rollouts, enabling student to progressively internalize procedural knowledge within its parameters. The central design principle is co-evolution: the policy generates rollouts that update the memory, and memory shapes the supervision that updates the policy. Empirically, across Qwen3-8B and OLMo3-Instruct-7B, PMD improves over SDPO by 3.8-5.5% on SCIKNOWEVAL and 7.9-13.6% on LIVECODEBENCH. Co-evolution powers these gains: freezing either the memory or the policy trails PMD by more than 10% across SCIKNOWEVAL domains.
We measured quantization-induced decision-boundary changes using local logit-margin radii, first-order boundary displacement, normal variation, slice-boundary Jaccard distance, grid prediction changes, multiclass junction counts, and low-margin boundary-band flips. On the digits benchmark, 8-bit weight quantization preserved all test labels while producing boundary-mask Jaccard \(0.428\) on the PCA slice; at 4 bits, accuracy remained \(0.9733\), while boundary Jaccard rose to \(0.970\) and median local boundary shift reached \(0.0290\). Interpolation between adjacent quantization levels localized the visible reconfigurations at multiclass junctions, with 12, 34, and 17 triple-junction cells in the selected transitions. Calibration-to-test stopping reduced the digits held-out flip rate from \(0.0094\) to \(0.0022\) and boundary Jaccard from \(0.825\) to \(0.524\); the same stopping rule also reduced flips on MNIST and Fashion-MNIST. On official CIFAR-10 subsets, PTQ-W selected by accuracy gave 6-bit flip \(0.0367\) and boundary Jaccard \(0.184\), whereas boundary-aware stopping selected 8-bit flip \(0.0083\) and boundary Jaccard \(0.048\). On full CIFAR-10 with three seeds, 6-bit PTQ-W lost \(0.0029\) accuracy relative to float, changed \(5.3\%\) of held-out decisions, and changed \(24.5\%\) of low-margin boundary-band decisions. A fixed-bit boundary-gap rounding term changed the trade-off at 4 bits by reducing boundary Jaccard from \(0.457\) to \(0.435\) and boundary-band pair-order flip from \(0.3600\) to \(0.3558\), with an accuracy trade-off; the 3-bit stress test exposed the tuning limit of this surrogate. Calibration boundary Jaccard predicted held-out boundary Jaccard across PTQ-W and optimized rounding variants with \(r=0.947\)--\(0.994\).
Class imbalance poses a critical challenge in federated learning (FL), where underrepresented classes suffer from poor predictive performance yet cannot be addressed by standard centralized techniques due to privacy and heterogeneity constraints. We propose FedCGNM (Federated Class-Grouped Normalized Momentum), a client-side optimizer in FL that partitions classes into a small number of groups based on minimum within-group variance, maintains a momentum per group, normalizes each group momentum to unit length, and uses the summation of the normalized group momentums as an update direction. This design both equalizes gradient magnitude across majority and minority groups and mitigates the noise inherent in rare-class gradients. We further provide a theoretical convergence analysis explicitly accounting for time-varying resampling-rates. Additionally, to efficiently optimize these rates in small-client regimes, we introduce FedHOO, an X-armed-bandit (XAB) based algorithm that exploits federated parallelism that evaluates many combinations of two candidate rates per client at linear cost. Empirical evaluation on four public long-tailed benchmarks and a proprietary chip-defect dataset demonstrates that FedCGNM consistently outperforms baselines, with FedHOO yielding further gains in small-scale federations.
Clinical protocol-execution tasks -- checking a lab value, applying a threshold, placing a correctly structured FHIR order -- are natural candidates for RL from world feedback: once clinical SMEs encode decision logic into a verifier, that verifier grades unlimited rollouts without per-episode annotation. But applying RL requires a sound feedback channel and sufficient base capability. We audit MedAgentBench v1/v2, find a 41.7\% silent-finish ceiling that makes inaction the RL dominant strategy, and construct \textbf{MedAgentBench-v3 (MAB-v3)} (508 tasks, 8.9\% ceiling). Training Qwen3-8B exposes two structural barriers: a \emph{capability ceiling} (10/20 task types have 0\% base performance, zero gradient) and a \emph{format-knowledge barrier} (3/20 types require exact clinical codes undiscoverable by exploration). Pure RL reaches 18.2\% pass@1 vs.\ 34.1\% for rule-based SFT; the 15.9~pp gap is attributable entirely to these barriers. A decision/format-knowledge/lookup taxonomy predicts RL learnability and prescribes the fix: SFT to inject codes, RL to learn conditionals.
Large language models are trained to predict the next token, not to act inside a specific API. In niche enterprise SaaS workflows -- where success means hitting the right endpoint with the right nested arguments in the right order -- this objective mismatch shows up as silent failures: dropped required fields, hallucinated tools, or early stops after a single read. We ask whether Reinforcement Learning with Verifiable Rewards (RLVR), applied directly in the target environment, closes the gap. As a proof of concept we build a suite of five synthetic environments emulating the Jira REST v3 and Confluence v2 APIs at schema fidelity; rewards are computed entirely from the tool-call trace, with no live API, no learned judge, and no human label in the loop. Scoring prompted Qwen3-1.7B and Qwen3.5-4B on the same checkers that drive GRPO training, we find that on the four scenarios whose rewards are non-degenerate the RL-trained policy lifts average reward from a 4B-baseline range of 0.35--0.92 to 0.95--1.00, with the largest single gain on Confluence page creation ($0.35 \rightarrow 1.00$). We position this as a preliminary step toward outcome-optimised small models for niche enterprise APIs, and foreground two limitations a workshop reader should weigh: hand-crafting verifiable rewards does not scale beyond the handful of endpoints reported here, and one of our five scenarios (ticket-transition) has a saturating reward shape that the prompted 4B already maxes out.
Text scaling, the task of positioning political actors on an ideological scale, is a fundamental task in political analysis. To ease the need for manual analysis, various NLP methods have been proposed for this task, including classification- and regression-based approaches, showing successes as well as limitations. The goal of our paper is to consolidate the state of the art in this area. We ask two questions: (a) Can the performance of scaling methods be improved by predicting scales not individually but jointly? (b) Is there a middle ground between classification and regression?
Large language models (LLMs) are increasingly applied to resume optimization for applicant tracking systems, introducing hallucination failures distinct from general text generation: anachronistic technology injection, cross-domain terminology contamination, structural mutation, and content fabrication. We present Grounded Optimization, a five-layer framework combining temporal context validation, deterministic contamination detection, structural invariant enforcement, prompt-level grounding, and an evaluator agent. In ablation experiments across three LLMs, four temperature settings, and six layer configurations on 25 synthetic resumes spanning 14 industries, undefended baselines produce 2.48-5.36 detected hallucinations per resume. Among detectors independent of the active defenses, temporal hallucinations are reduced by 50-95% across all conditions; overall detected hallucination rate falls to 0.04-0.24. Prompt-level grounding alone achieves zero detected hallucinations at low temperature with a capable instruction-following model; higher temperatures and weaker models reveal the need for the deterministic layers as a complement. We release the contamination taxonomy, evaluation code, and raw data.
Agent Skills provide on-demand domain knowledge to LLM agents without requiring model retraining. Each Agent Skill is defined by a mandatory SKILL.md file containing metadata and an unstructured Markdown body whose contents are left entirely to the skill author. Despite the rapid adoption of Agent Skills, little is known about how these files are authored or whether existing authoring guidelines are followed in practice. In this paper, we present the first systematic study of SKILL.md files as a software artifact. We qualitatively analyze 238 real-world skills and derive a taxonomy of 13 higher-level and 44 lower-level semantic components. We then conduct a multivocal literature review of 29 sources to identify best practices for authoring SKILL.md files and introduce skill smells as violations of these practices. Finally, we develop an automated detector and apply it to real-world skills, finding that over 99% of SKILL.md files contain at least one skill smell, and once introduced, skill smells rarely disappear as skills evolve. These findings reveal a substantial gap between recommended and actual authoring practices, motivating the development of automated techniques to remediate skill smells while increasing developer awareness of this emerging quality issue.
Language models learn continuous programs over discrete symbols, with the embedding table and LM-head acting as the read/write interface between them. We show that this interface has gradient geometry distinct from dense hidden weights which can be exploited to improve the Pareto frontier across supervised finetuning, RL, and pretraining, while only utilizing kilobytes of optimizer state. We introduce Ember, a lightweight optimizer for embedding and LM-head matrices that utilizes O(V + D) VRAM, instead of Adam's O(2VD), and forgoes the need to shard both token table optimizer states. We provide empirical evidence that Ember scales effectively across batch size and parameter count. We show that the optimization trajectory of tokens can be well described by a simple 1D ray, counter to the popular belief that neural net parameters navigate a heavily nonconvex landscape. We provide a principled view on the surprisingly narrow space of optimizers that suffice for Transformer training. Finally, we open-source our distributed Ember implementation that merges cleanly with existing ZeRO/FSDP setups to support further research at https://github.com/katop1234/ember
We propose a novel hybrid neural architecture, the Geometry-aware R-Structured Kolmogorov-Arnold Network (GRS-KAN), which integrates V.L.Rvachev's R-functions into the Kolmogorov-Arnold Network (KAN) framework. The proposed approach combines two complementary modeling mechanisms: smooth nonlinear structure is learned by KAN branches, while known geometric or logical constraints are encoded analytically using differentiable R-functions. This enables explicit representation of discontinuities, feasible regions, and implicit geometric boundaries within a trainable neural architecture. The framework implements differentiable logical operations through R-conjunctions and R-disjunctions, allowing complex geometric supports to be represented analytically and incorporated directly into regression models. Several GRS-KAN variants are introduced, including additive, multiplicative, and agnostic branch-weighted architectures. The method is demonstrated on regression problems involving discontinuities with circular and rectangular supports. Numerical experiments show that explicit geometric encoding substantially improves predictive accuracy and boundary localization compared with standard KANs. In the considered benchmarks, geometry-aware GRS-KAN models reduce test RMSE by up to 67% while simultaneously improving interpretability through explicit analytical representation of the learned geometric structure. The agnostic variant further demonstrates the ability to automatically determine whether geometric priors are beneficial for a given learning task.
Malware poses a critical and ever-evolving threat, and robust and effective systems for detecting and classifying malware are of essential importance. $n$-grams features are among the common static features used in effective machine learning systems for malware, but these features are inherently brittle. We propose an algorithm for constructing more robust features, hamm-grams, which are a special class of regular expressions having a fixed length and single-character wildcards. We devise an efficient algorithm for finding common hamm-grams using a new locality-sensitive hash designed to produce collisions among pairs of small Hamming distance and a clustering within hash buckets to place wildcards. We then demonstrate the advantages of these features in malware classification and detection tasks.
Mixture-of-Experts (MoE) models offer inference speedups via selective activation but impose substantial memory requirements because the whole network must remain loaded. Structured expert pruning is a practical approach for reducing deployment costs in resource-constrained settings. However, prior studies primarily evaluate benchmark utility, leaving the effect of pruning on factual reliability underexplored, particularly in high-stakes domains such as biomedicine. In this paper, we investigate how domain-specific expert pruning affects both utility and reliability. We assess four MoE models, six pruning methods, and multiple pruning ratios across generation and classification tasks under in-domain (biomedical) and cross-domain settings. Results reveal that moderate pruning preserves in-domain utility without immediate reliability decline, although hallucination risks increase at extreme pruning ratios. When shifting to the general domain, both utility and reliability degrade rapidly. These findings indicate that safe compression depends heavily on the task and domain. Evaluating pruned MoE models solely on utility is inadequate for high-stakes deployment without reliability assessment.
Faithful reasoning is essential in medicine, where clinical decisions require transparent justification grounded in reliable evidence. Current medical LLMs either lack active access to evidence or use retrieved evidence without supervising how it should be appraised and applied during reasoning. To address this, we formalize evidence-based medicine principles as process-level criteria and introduce FaithMed, a framework that combines clinician-designed, automatically refined rubrics with reinforcement learning using step-level process reward assignment and advantage grouping. Across seven medical benchmarks, FaithMed improves over agentic-search baselines (+9% on average) and outcome-only RL (+5.8%), while raising average evidence-based medicine rubric scores over agentic-search Qwen3 baselines (+15.5%). This work demonstrates that explicit step-level supervision can improve both task success and the faithfulness of the reasoning process. Code is available at https://github.com/cxcscmu/FaithMed.
Diffusion language models, which generate text by denoising a token canvas bidirectionally instead of emitting tokens left to right, have become competitive with autoregressive (AR) generation. Medical foundation models, however, remain almost entirely autoregressive. We adapt a mixture-of-experts diffusion language model, DiffusionGemma-26B, and benchmark it against its same-size AR sibling Gemma-4-26B under an identical LoRA recipe on medical visual question answering datasets, scored by a verbosity-robust LLM judge. Diffusion matches or exceeds AR on all of them, and the finetuned model (3.8B active) is competitive with frontier vision-language models; its decoding is also 3.5-4.4x faster. Beyond this parity, the diffusion model offers a drafting capability AR lacks: any-order infill. Because the canvas is denoised bidirectionally, a radiologist can fix report fragments and have the model fill the text between them, an operation inherent to diffusion but not to autoregression, which is subpar at it. This suits real reports, which are often terse or inconsistent across clinicians and institutions.
Significant advancement of immersive technologies such as Virtual and Augmented Reality (VR/AR) and their integration into diverse aspects of modern life need authentication interfaces that are secure, intuitive, and compatible with embodied interaction. Traditional methods such as passwords, PINs, and device-based logins, break immersion and rely on external hardware. Recent 3D-specific behavioral approaches, such as hand-gesture, eye-tracking, and electroencephalography (EEG)-based methods, offer promising alternatives but often require specialized sensors or constrain natural movement, limiting usability in dynamic environments. We present Sign in the Air to Unlock, an in-air signature interface that enables users to authenticate by signing naturally in 3D space which is a familiar, personal, and reproducible gesture. To realize this interface, we design a point-voxel Cross-Attention Network (PV-Net) that jointly models local motion dynamics and global spatial structure from 3D trajectories. The model is evaluated on two datasets: the public DeepAirSig dataset (1,800 signatures from 40 users) and ImmAirsig, a new dataset collected using Meta Quest 2 in immersive VR (880 samples from 22 users). PV-Net achieves an Equal Error Rate of 2.5% on DeepAirSig and 76% classification accuracy on ImmAirSig. These findings highlight the potential of 3D behavioral interfaces for seamless, user-centric authentication that merges security with natural interaction in immersive environments.
Divergent thinking is a crucial aspect of creativity, yet large language models (LLMs) tend to consistently generate similar responses to open-ended questions, in what has been termed the artificial hivemind effect. Here, we introduce CreativityNeuro, a data-free method for enhancing divergent thinking in LLMs via contrastive weight steering. We evaluate our method across multiple creativity assessments and report several main findings. On the Divergent Association Task (DAT), a vocabulary-space creativity test, CreativityNeuro improves performance by up to 14 human percentile points. Next, in a large-scale human evaluation (N=720) on the Alternative Uses Test (AUT) and the Task Task, CreativityNeuro achieves significant improvements in originality, surprise, and creativity, transferring to longer-form and more open-ended tasks. Importantly, we find that across all three tasks, CreativityNeuro demonstrably reduces measures of mode collapse. Moreover, activation steering achieves comparable performance to CreativityNeuro on the DAT, but it does not transfer to the AUT and Task Task, demonstrating the effectiveness of weight-space steering in generalizing to unseen tasks. In conclusion, CreativityNeuro improves divergent thinking and reduces mode collapse without requiring behavioral data, re-training, or gradient-based fine-tuning, providing a straightforward way to enhance LLM performance in creative domains.
We introduce ISOSCI, a benchmark of isomorphic cross-domain science problem pairs that separates reasoning ability from domain knowledge retrieval in LLM evaluation. Each pair shares identical logical structure but requires different domain-specific knowledge, enabling controlled attribution of reasoning-mode gains. Across five model pairs spanning four model families, we find that 91.3% of reasoning-mode gains are knowledge-dependent rather than structure-invariant (63/69 gains; Wilson 95% CI [82.3%, 96.0%]), directly challenging the assumption that chain-of-thought reasoning improves short-horizon procedural scientific problem-solving. Reasoning toggles on highly capable models provide less than 5 percentage points accuracy gain across all domains, and a reasoning-specialized model (o3-mini) that outperforms its standard counterpart on GPQA Diamond (+19.2 percentage points) underperforms on ISOSCI (-24.7 percentage points), showing that benchmark choice determines conclusions about reasoning utility. We release ISOSCI at https://huggingface.co/datasets/isosci/isosci
Autonomous customer-service agents are shifting from conversational interfaces toward operational execution roles: they retrieve firm records, apply service policies, and execute backend writes such as refunds, cancellations, exchanges, order modifications, and reservation changes. This shift creates a service-control problem: firms must keep routine service fast and low-friction while preventing operational errors on requests where customer instructions, policy constraints, firm records, and backend writes interact. We propose a difficulty-routed service-control architecture that asks when service agents should reconsider before acting. A lightweight router keeps routine sessions on a low-cost baseline path and routes operationally coupled sessions to an escalated workflow. The escalated path uses conflict-aware communication and write-triggered reconsideration to concentrate deliberation and safeguards before consequential backend writes, rather than applying additional control uniformly across all service sessions. We evaluate the architecture on human-verified retail and airline tasks from $τ^{2}$-bench. In retail, the method improves reliability consistently on service requests with operational conflict. Routing evidence shows that stronger control is directed toward conflicted requests rather than broadly applied to routine ones. Dialogue and tool-use profiles suggest that gains do not come from indiscriminate interaction expansion or broader tool chains; instead, added turns and tool calls support evidence gathering, write separation, and pre-write reconsideration. Case-level evidence shows that the escalated workflow preserves fallback plans, binds retrieved records to the correct action, sequences writes, and decomposes multi-entity requests. Airline results extend the same service-control logic to reservation operations.
Understanding large, complex codebases, especially those with obfuscated structures and incomplete documentation, remains a significant challenge. Existing code summarization solutions often rely on a single language model or coding assistant like Claude Code, and treat source code as flat text, underutilizing the rich interdependencies and hierarchical information within a repository. To address these shortcomings, we propose Agent4cs - a multi-agent framework that summarizes large codebases in a bottom-up fashion, where a summarization agent focuses on producing robust summaries; a keyword-extraction agent proactively identifies critical information from subfolders; and a quality-assurance agent iteratively refines the outputs for readability, coherence, and completeness. Evaluated on 7 frontier models, Agent4cs improves semantic consistency across all folder levels by average 8% compared to two structured prompting baselines with code segments. Furthermore, extensive evaluation on real-world datasets demonstrates up to 38% gains in normalized keyword coverage rate over the same baselines.
Engineering management research has produced mature frameworks for software risk: ownership by feature, escalation by severity, and assurance by test coverage. These frameworks implicitly assume deterministic behavior, discrete and auditable change events, and clear component-to-owner mappings. Teams that build and operate agentic AI systems violate all three assumptions at once: outputs are probabilistic, systems take autonomous multi-step actions, and the risk surface mutates silently between deployments. Existing AI risk literature addresses this from above (policy frameworks such as the NIST AI RMF and ISO/IEC 42001) or below (threat taxonomies such as OWASP's agentic AI guidance), but not at the layer where an engineering manager (EM) operates: roles, decision rights, and escalation structures. This paper contributes (i) a seven-dimension profile distinguishing pure software-engineering, hybrid, and AI-native teams; (ii) a six-cluster failure-mode taxonomy including a previously unarticulated cluster, dependency-boundary determinism mismatch; and (iii) a synthetic framework-adequacy methodology scoring how well each profile's risk architecture detects, contains, and escalates a defined scenario set. Because the object of study is framework adequacy rather than human behavior, the evaluation yields derived rather than observed coverage claims. Coverage degrades as teams move from pure software engineering to AI-native operation, monotonically in the median and abruptly in the count of uncovered, high-consequence failures appearing only at the AI-native step. The degradation concentrates in specific failure-mode categories, and the most severe, least-covered failures arise not inside AI-native teams but at the organizational boundary where their probabilistic outputs are consumed by determinism-assuming dependencies.
As grounded QA systems are increasingly deployed in AI assistants, accurately attributing generated answers to evidence is critical for user trust and model safety. While unimodal attributions have been explored in depth, the multimodal setting remains relatively under-researched. As a result, we introduce MultAttnAttrib, a training-free attribution-generation method that leverages a model's prefill pass, selected attention heads, and calibrated thresholds to locate source evidence within a document. To establish baseline results for the method, we introduce MultAttrEval, a complementary benchmark dataset annotated with fine-grained, ground-truth attributions for answer components grounded in multimodal source documents. To our knowledge, this is the first evaluation dataset designed specifically for multimodal attribution in long-form documents. Experimental results show that MultAttnAttrib consistently outperforms a variety of attribution-generation methods, including several strong prompting-based approaches and matches the latest frontier models such as GPT 5.4. Our method not only substantially improves attribution accuracy for both unimodal and multimodal attribution types, but also produces attributions at up to one-seventh of the direct inference latency compared to prompting on the same base model.
Organizations rolling out agentic command line tools like Anthropic's Claude Code and GitHub's Copilot CLI need to know who will try them, who will keep using them, and whether the tools produce enough output to justify their cost. At organizational scale, token spend can run into millions of dollars annually, so misreading adoption, retention, or impact can make a rollout expensive without changing engineering velocity. Studying tens of thousands of engineers at Microsoft over its early-2026 rollout, we find that first use spread primarily through social networks, retention was associated more with engineers' coding activity than with demographics, and adopters merged roughly 24% more pull requests than they would have otherwise. We use merged pull requests as our proxy for output -- acknowledging that a merged PR is not the same as the value it delivers -- and the lift persists across our four-month window. These results suggest that CLI coding agents are neither uniformly adopted nor mere novelty effects and that organizations should treat visible peer use as central to rollout strategy.
Conditional inference trees (CIT) and conditional inference forests (CIF) reduce split-selection bias by testing features before choosing split thresholds, but repeated permutation tests and threshold searches can make these methods computationally expensive. We study CIT and CIF as top-$k$ feature-ranking methods for downstream prediction using real-data benchmarks, runtime ablations, and synthetic feature-recovery experiments. At a fixed node, if the features and permutation budget do not depend on the node responses, Bonferroni-corrected $+1$ Monte Carlo permutation $p$-values control nodewise rejection under the complete permutation null. CIF ranks 4th among 17 classification methods on 22 datasets and 3rd among 18 regression methods on 8 datasets. With Bonferroni correction held fixed, the CIF runtime ablations indicate that adaptive stopping and the number of thresholds searched have the largest measured effect on runtime: turning off adaptive stopping and using exact threshold search increase fitting time by 4.0--8.4$\times$ and 1.9--10.8$\times$, respectively, while downstream score changes are at most 0.011. Sparse high-$p$ simulations indicate that forest feature sampling can leave informative features out of many split decisions. Overall, the results support CIF as a top-$k$ feature-ranking method in the evaluated downstream prediction benchmarks.
Coding-agent reinforcement learning treats execution infrastructure as a background implementation detail, despite relying on large numbers of interactive software rollouts. This is a missed opportunity: measuring infrastructure overhead can reveal practical efficiency gains for RL post-training, where small per-rollout savings compound at scale. We present a comparative study of four execution substrates: single containers, hosted sandboxes, Kubernetes-orchestrated containers, and cloud virtual machines. We find up to $110\times$ variation in cold-start latency and a $1.8\times$ spread in projected worker-hours for one million 150-step trajectories. Our results suggest that future coding-agent RL systems should optimize execution substrates as part of the training system itself, not merely as deployment plumbing.
Sim2real transfer for robot policy learning suffers due to mismatch between simulation and reality. Existing methods typically address each gap in isolation through separate adaptation modules, which are composed or layered when both gaps coexist. Yet the basis for attempting sim2real in the first place is that there is shared structure between a task in simulation and reality, where equivalent actions from equivalent configurations produce equivalent long term outcomes regardless of domain specific differences in rendering or physics. In this paper, we study whether we can identify and exploit this shared structure from raw observations to train a policy that enables zero shot transfer. We introduce BIFROST, which learns a shared history encoder on paired cross-domain data via cross-domain bisimulation objective: observation-action sequences leading to equivalent long-term behavior are mapped to nearby latent states, regardless of domain. Policies trained on these latent states in simulation transfer zero-shot to reality. We provide empirical evidence on sim2sim visual navigation and sim2real contact rich manipulation task and visual servoing task that BIFROST achieves effective transfer where domain adaptation and co-training baselines fail under both visual and dynamics domain gaps.
GPU training jobs fail often, roughly two in five on large production clusters, yet the operator typically learns of a failure only by reconnecting hours later. Experiment trackers require editing the training script and maintaining a cloud connection; the scheduler's mail hook delivers a single status line with no cause and no logs. GPUAlert is a command-line wrapper that monitors any training command at the process boundary, and with no change to that command, emails a structured notification on completion carrying a classified failure cause, durable logs, and output artifacts. The tool is organized around three reliability primitives: a pre-launch log guarantee that establishes the durable destination before the child process can crash, notifier isolation that makes the wrapper's exit code a pure function of the child's status regardless of whether the email succeeds, and a non-silent artifact budget that bounds attachment size without ever dropping output silently. We release a labelled corpus of 474 GPU training logs across 15 failure classes and a reproducible evaluation harness. On the twelve hardware-reproduced classes, the ordered-rule classifier reaches 0.997 macro-F1, against 0.830 for unordered keyword matching and 0.133 for exit-code inspection. Wrapper overhead is a constant approximately 3ms per job; the pre-launch guarantee preserves a log where a shell redirect yields nothing; and across all 15 failure modes the wrapper returns the child's exit code unchanged even when the SMTP relay is unreachable.
$\mathrm{E}(3)$-equivariant networks are promising for 3D atomistic system modeling, yet their scalability is limited by the $O(L^6)$ complexity of the Clebsch-Gordan Tensor Product (CGTP). The recently proposed Gaunt Tensor Product (GTP) reduces the complexity but is unable to capture the antisymmetric paths, resulting in incomplete expressivity. In this work, we present SpinGTP, an approach to overcome the GTP incompleteness by generalizing from scalar functions to Spin-Weighted Spherical Harmonics (SWSH). By relying on the algebraic properties of SWSH, SpinGTP recovers the missing antisymmetric interactions while maintaining the asymptotic efficiency of GTP. It also allows for a more expressive equivariant basis that naturally accounts for the parity-odd components of tensor products. We evaluate SpinGTP across diverse benchmarks, including Tetris, 3BPA, SPICE-MACE-OFF, and OC20. Our results show that SpinGTP achieves accuracies comparable to full CGTP. Notably, by explicitly capturing antisymmetric paths, SpinGTP exhibits superior performance in tasks involving chiral materials and non-centrosymmetric geometries. This work provides a complete, scalable, and mathematically rigorous path toward high-order equivariance in large-scale 3D atomistic system simulations.
INTRODUCTION: Accurate MRI-based identification of Alzheimer's disease (AD), mild cognitive impairment (MCI), and related dementias remains challenging because disease-related structural changes are often subtle and heterogeneous. We developed NeuroBridge, a clinically guided multi-task MRI framework for neurodegenerative disease diagnosis. METHODS: NeuroBridge integrates large-scale self-supervised MRI pretraining with hippocampal segmentation, hippocampal atrophy classification, and reconstruction objectives, followed by gated fusion fine-tuning. Performance was evaluated across ADNI and OASIS cohorts, including cross-cohort transfer, probability-based analysis, and opportunistic screening. RESULTS: NeuroBridge achieved the highest performance across evaluated classification tasks, reaching 88.17% accuracy for AD versus cognitively normal controls in ADNI and 82.78% in OASIS. The largest gains occurred in MCI-related and mixed-diagnosis settings. The framework demonstrated strong cross-cohort generalization, systematic associations between predicted-class probability and accuracy, and the feasibility of probability-based opportunistic screening. DISCUSSION: Clinically guided multi-task representation learning improves neurodegenerative MRI diagnosis beyond conventional single-task approaches. NeuroBridge provides a robust and scalable framework for dementia assessment and MRI-based opportunistic screening.
Deep multimodal brain-encoding models now predict fMRI responses to naturalistic video with high accuracy. Whether their predicted neural signals also forecast behavioral engagement is unknown. We run TRIBE, the winning model of the 2025 Algonauts brain-encoding challenge (Llama-3.2 + V-JEPA2 + Wav2Vec-BERT), on 48 YouTube videos and reduce its predicted cortical response to a per-second engagement curve, the global field power. Correlated against each video's "most replayed" heatmap, a passively-collected proxy for which moments viewers return to, the curve shows no evidence of predicting re-watch behavior. The pooled position-controlled partial correlation is +0.058 (95% CI [-0.04, 0.15]; one-sample t(47)=1.21, p=0.23), indistinguishable from zero and not significantly above simple loudness and motion baselines (loudness +0.04, paired p=0.74). The raw correlation is also near zero; the moderate values reported for music videos reflect a genre-specific intro/onset-replay artifact rather than content prediction, and do not generalize. The null holds across six cortical-network readouts and under an autocorrelation-preserving permutation test. We release the code, the video-ID manifest, and an acquisition method that works despite YouTube's SABR-only streaming.
At the heart of human visual perception lies the ability to maintain a continuous and coherent understanding of the external world. By integrating observations with accumulated experience, the human visual system can continuously adapt to variations in both the target and its surrounding environment, while preserving robust visual continuity as scene dynamics evolve. Human vision can therefore integrate prior knowledge, spatial geometry, and semantic context to understand complex scenes and their changes. As a core problem in computer vision, visual object tracking aims to bring machine perception closer to human visual perception. These capabilities are central to the task of Generic Object Tracking (GOT). In this task, a visual tracker is initialized only with the bounding box of an arbitrarily specified target in the first frame, and must continuously localize the target in subsequent dynamic visual streams. However, future events, observations, and real-world variations are inherently unpredictable; therefore, the model's generalization and online adaptation capabilities remain bottlenecks. Tracking reliability can deteriorate when the target undergoes severe deformation, is affected by complex distractors, encounters significant environmental changes, or belongs to a category unseen during training. This dissertation aims to narrow the gap between machine visual tracking systems and human visual perception by proposing a series of methods that systematically enhance the target discrimination, robust adaptation, and geometric reasoning capabilities of tracking models.
We present Wiola, a fully original Small Language Model (SLM) architecture built from first principles, sharing no structural lineage with any existing model family including GPT, LLaMA, Mistral, or Falcon. Wiola introduces five independently novel components: (i) Spiral Rotary Positional Encoding (SRPE), which embeds token positions on a three-dimensional helical manifold combining absolute, relative, and hierarchical positional signals; (ii) Gated Cross-Layer Attention (GCLA), providing each decoder layer with soft cross-attention access to compressed summaries of two preceding layers for inter-layer coherence; (iii) Adaptive Token Merging (ATM), which dynamically merges se mantically redundant adjacent tokens in middle network layers to reduce attention complexity without information loss; (iv) Dual Stream Feed-Forward (DSFF), replacing the conventional MLP with two parallel streams fused by a learned per-dimension gate; and (v) WiolaRMSNorm, a modified normalisation introducing a per-dimension learned offset vector that prevents representation collapse. We provide complete mathematical derivations, architectural block diagrams, complexity analyses, and systematic comparisons against GPT-2, LLaMA-2, and Mistral. Wiola is released in four sizes (120M, 360M, 700M, and 1.5B parameters) and is fully compatible with the HuggingFace Transformers ecosystem, with all 22 architectural unit tests passing.
Aligning large language models with diverse and heterogeneous human values requires multi-objective alignment methods to effectively trade off conflicting preference dimensions. Current methods achieve this trade-off by training policies conditioned on preference vectors and leveraging online direct preference optimization. However, exploration uncertainty can cause the reward distributions of responses generated under different preference vectors to overlap, and the generated responses may fail to effectively align with the corresponding preference vectors. In this paper, we propose Multi-Objective Exploration and Preference Optimization via Mutual Information (MI-EPO), an information-theoretic framework. It unifies multi-objective exploration and alignment by maximizing the joint conditional mutual information among generated responses, preference feedback, and preference vectors. By incorporating a probabilistic routing mechanism, MI-EPO naturally decomposes objective alignment and preference-aware exploration, encouraging the model to generate responses that are distinguishable and aligned with different preference conditions. Experiments on safe alignment and helpful assistant tasks show that MI-EPO significantly improves the alignment between generated responses and preference vectors, makes the outputs more controllable, and achieves stable trade-offs across multiple objectives.
How do we encode numeric values in transformer-based sequence processing, particularly in electronic health record (EHR) data? We systematically compare discrete, continuous, and hybrid value encoding strategies using synthetic arithmetic tasks embedded within real-world EHR data, as well as real-world clinical prediction tasks. Our study reveals trade-offs between numeric precision, optimisation stability, and architectural flexibility. We find that approaches that explicitly model value-concept interactions perform best on precision-sensitive arithmetic tasks when architectural constraints permit. Hybrid token-based approaches that retain numeric values but apply binning prior to projection provide a more robust and broadly applicable alternative, with the optimal number of bins following a simple empirically derived power-law in dataset size. Across tasks, models consistently exhibit reliable "good enough" numeric computation rather than exact arithmetic, while clinical gains from incorporating laboratory values are task-dependent. This suggests that robustness and deployability often outweigh maximal numeric precision in practice, motivating hybrid token-based approaches as a practical default.
Multi-step symbolic reasoning is essential for robust financial analysis, yet most benchmarks neglect intermediate reasoning steps. FINCHAIN introduced verifiable Chain-of-Thought (CoT) evaluation but is limited to English. FINESSE-Bench includes a Russian block but relies on multiple-choice questions without step-level supervision. We present RusFinChain, the first Russian-language symbolic benchmark for verifiable CoT reasoning in finance. It spans 17 domains, 172 topics, and comprises 5,280 parameterized examples from executable Python templates, ensuring contamination-free evaluation. Each example includes a gold-standard reasoning chain with intermediate numeric values for automatic verification. We also introduce enhanced metrics: Fuzzy Numeric Alignment and Soft-Attention Alignment. We evaluate 8 open-weight LLMs on a stratified sample, generating 8,100 responses. Results reveal a substantial reasoning gap: models achieve Hard F1 of ~0.65 for step alignment, but only ~29% of final answers are correct. Our fuzzy and soft metrics show stronger correlation with final-answer correctness (Spearman rho approx 0.48) than the original ChainEval (rho approx 0.38-0.46), demonstrating superior diagnostic power. We release dataset, code, and evaluation framework to foster verifiable financial AI for the Russian-speaking community.
Recommender systems are vital in helping users navigate vast amounts of information, offering personalized suggestions and effective explanations for these recommendations. While previous efforts have attempted to provide such explanations, evaluating their effectiveness across various scenarios remains a challenge. Enhancing these explanations is essential for improving user engagement, trust, and decision-making. To facilitate effective explanations within the recommender system, we propose a Bi-level Neural Architecture Search (Bi-NAS) framework to optimize explanations. This approach simultaneously refines cross-attention mechanisms and feature interaction functions by exploring both intra-layer and inter-layer design spaces. Furthermore, we integrate Large Language Models (LLMs) to enhance explanation generation, leveraging zero-shot prompting to produce more effective and personalized justifications. By aligning user feature preferences with item quality scores, our approach ensures that explanations reflect both user intent and item attributes, improving transparency and reasoning depth. Extensive evaluations on four real-world datasets demonstrate that Bi-NAS not only boosts recommendation accuracy but also significantly improves the effectiveness of explanations for recommender systems, providing users with clear and reliable insights into the suggestions they receive.
Gravitational Waves (GWs) represent the newest window of astronomy, furthering our understanding of compact objects like black holes and neutron stars in the Universe. The signal from two merging neutron stars is especially interesting since it brings the prospect of concordant electromagnetic and neutrino emissions. Such multi-messenger observations have a transformational impact on fundamental physics, nuclear matter, astrophysics, and gravity. It was first witnessed in 2017 with the detection of the binary neutron star (BNS) merger GW170817. However, searching for BNS signals in real-time in the LIGO-Virgo-KAGRA (LVK) GW detectors presents a computational challenge, as the data streaming out must be matched against $\sim$ million reference waveforms, which requires up to a thousand CPU cores. We present a different approach using neural networks to learn the presence of a signal in the data. Our algorithm, called Aframe, was deployed in the LVK's fourth observing run and was the first artificial intelligence (AI)-enabled search to detect multiple binary black holes (BBHs) live. In this work, we demonstrate that the approach extends to the lower-mass BNS regime, and is the first AI-enabled search that achieves sensitivity comparable to matched-filter pipelines at lower computational and latency costs. The challenge of the longer-duration BNS signals is addressed by heterodyning the data, following which the network architecture used for BBHs is sufficient to distinguish signal versus background. We also show that this analysis requires a single non-flagship GPU for online deployment. Furthermore, the design and adoption of inference-as-a-service tools allow rapid offline analysis using a distributed pool of GPU resources. Hence, aside from the use case of rapid online data analysis, we also establish the use of Aframe for efficient archival data analysis.
Federated learning (FL) research often depends on many small but consequential algorithmic choices: optimizer variants, server aggregation rules, local training schedules, normalization, regularization, and model architecture. These choices are expensive to explore manually and difficult to compare fairly when candidate changes can also alter the FL training or evaluation path. In this work, we present Auto-FL-Research (AFR), a constrained coding-agent workflow for FL algorithmic recipe search. Agents may propose and implement candidate training algorithms, including server aggregation rules, client update schedules, local objectives, and registered model variants, while task profiles fix the mutation surface, compute budget, communication contract, and final model evaluation. Each campaign records candidate scores, runtime, edited files, artifacts, and failure status. We evaluate AFR on five healthcare cross-silo FLamby tasks and on grouped-client profiles for the five fixed LEAF datasets plus the LEAF synthetic task. Five-seed repeat evaluations support gains on four FLamby tasks and five of six LEAF profiles, while also exposing seed-sensitive and search-selected failure cases. Same-budget controls show that several gains correspond to FL-recipe changes, whereas other improvements are recovered by fixed-surface scalar controls or fail under repeat or held-out evaluation. These mixed outcomes are part of the contribution: they show how agent-generated candidates can be separated into repeated FL mechanisms, fixed-surface tuning effects, and selected single-run artifacts.
Given one or more images of a railway crossing, can we leverage visual cues that allow us to robustly estimate how safe it is? Can we improve our ability to do so by introducing structured data (such as official accident reports) about the accident history of that crossing into our models? In this work, we explore how to best answer those questions towards building an AI system that can ingest multi-modal data for railway crossings and provide safety assessment and scores that align with expert opinion and with safety scoring used by the Federal Railroad Administration (FRA). To that end, we propose a proof-of-concept pipeline that delivers on that goal, while at the same time exploring and tackling a number of critical research challenges that pertain to different parts of the pipeline, from data preparation to different learning paradigms that can allow us to realize such a system. Indicatively, our proposed system identifies HIGH-RISK and LOW-RISK crossings with a macro F1 score of 0.757 and estimates FRA-based safety scores with an RMSE of 0.078 and correlation of 0.492 using a routed fine-tuned compact VLM pipeline, while producing qualitative results that align with domain-expert assessment.
Quantum mechanical (QM) cluster models provide an effective framework for mechanistic studies of enzymatic reactions but remain computationally demanding. Neural network potentials (NNPs) offer a promising route to reduce this cost, but enzymes present challenges beyond small molecules, including large system sizes, implicit-solvent environments, substantial polarization, and charge transfer. Here, we present an integrated software framework for efficient NNP training for mechanistic studies of enzymes, demonstrated on QM cluster models of S-adenosyl-L-methionine-dependent methyltransferases (MTases). Our Enerzyme code introduces modular electrostatics-aware NNP architectures and combines automated QM-cluster construction with reactive dataset generation. The Enerzymette subpackage automates reaction pathway exploration at both NNP and DFT levels. We show that iterative flexible scans and nudged elastic band calculations impose stricter requirements on NNPs than conventional dataset metrics. Nevertheless, NNPs trained on fewer than 1,000 system-specific datapoints reproduce reaction energetics and transition-state structures for MTase clusters containing up to 545 atoms with near-chemical accuracy. Direct supervision of atomic charges and consistent dielectric screening substantially improve simulation stability and accuracy, while multitask-learned atomic charges capture charge transfer and polarization trends and provide chemically meaningful descriptors of reactivity. Finally, transferability across chemically diverse catechol O-methyltransferase substrates indicates that NNPs learn generalizable reactivity patterns as training data expand across multiple enzymes. Together, these results establish a foundation for accelerating enzyme mechanistic studies and guide future NNP development for biomolecular reactivity.
Large language models (LLMs) are typically evaluated on code generation and program repair using binary functional correctness: a generated program or patch either passes or fails a test suite. This protocol is simple but coarse, as it ignores partial progress, feedback use, regressions, and the refinement trajectory through which models often improve code. We introduce PAIR-Bench, a progressive and adaptive benchmark for evaluating code improvement: transforming an incorrect or incomplete program into a more correct one through feedback-guided refinement. PAIR-Bench uses progressive hinting, a structured feedback protocol with two controls. Failure-region control determines what the feedback targets by grouping hidden failing tests into failure scenarios, while hint-depth control determines how much repair-relevant information is revealed, from coarse symptoms to implementation-level guidance. This design enables PAIR-Bench to measure whether a model repairs targeted failures, generalizes beyond the hint, preserves already-correct behavior, and how much assistance it requires. By evaluating repair trajectories progressive metrics rather than only final pass/fail outcomes, PAIR-Bench provides a finer-grained assessment of LLM code-improvement capability.
Turn-taking naturalness is central to full-duplex spoken dialogue systems, yet its automatic evaluation remains limited. Existing evaluations often rely on human judgments or behavior-specific timing metrics, making it difficult to compare heterogeneous timing failures within a unified framework. We propose TurnNat, a likelihood-based framework for automatic turn-taking naturalness evaluation in two-channel spoken dialogue. A causal turn-taking prediction model trained on natural conversations estimates future two-speaker voice-activity states, and the negative log-likelihood (NLL) of the observed future activity measures timing atypicality. TurnNat pools frame-level NLLs over turn-taking boundary units (TBUs) extracted from utterance onsets and offsets, and aggregates mean and tail TBU scores into a dialogue-level naturalness score. We further construct a controlled perturbation benchmark of paired natural and perturbed dialogue clips, validated by human naturalness judgments. Experiments on this benchmark show that TurnNat successfully identifies unnatural turn-taking perturbations across heterogeneous timing failures.
Neural Quantum States (NQS) are a remarkably expressive class of variational ansätze for quantum many-body wavefunctions, yet little is understood about their internal mechanisms: trained on variational objectives alone, how do NQS accurately capture physical observables that they have never been explicitly optimized for? In this work, we present a systematic approach to analyze the internal activations of NQS using sparse autoencoders. We extract features from the residual stream and demonstrate that these features strongly correlate with physical observables such as order parameters, staggered magnetization, and half-chain correlators, across both ground state representation and real-time dynamics. Remarkably, the discovery of these features is entirely unsupervised, with no physical labels provided. We further establish that such features causally affect the corresponding observables predicted by NQS, by showing that targeted, post-training intervention on a \textit{single} feature smoothly and monotonically steers the corresponding observable, while leaving the variational energy nearly unchanged. These results demonstrate that NQS are not merely functional approximators, but encode rich, interpretable internal representations of physical information. Our approach provides both a diagnostic and an intervention tool for NQS, and serves as a foundation for using mechanistic interpretability towards more reliable, transparent NQS.
The geometric and topological structure of quantum cost landscapes (QCLs) governs the optimization and thus the predictive power of variational quantum algorithms (VQAs). We systematically analyze ravines - low-cost paths connecting local minima - using an adapted version of the nudged elastic band (NEB) algorithm, a method originating from theoretical chemistry. By training quantum neural networks (QNNs) to classify the concentratable entanglement of quantum states, we apply the NEB algorithm and numerically identify ravine structures in QCLs of hardware-efficient ansatzes. Beyond visualizing these ravines, we construct an ensemble prediction framework by averaging predictions from QNNs parameterized along the low-cost NEB path. We introduce a resource-light pre-training metric which quantifies local-prediction variability and serves as a strong performance indicator for VQAs, even beyond the scope of this study. When base classifiers are drawn from circuit and weight initializations exhibiting high local-prediction variability, the quantum-based NEB ensembles outperform both classical and naive quantum alternatives. Moreover, a complexity analysis shows that leveraging the ravine-like structure of QCLs with the QNN NEB approach substantially reduces computational costs compared to naive QNN ensembling. A depth and qubit scaling analysis indicates that ravines persist across both scalings, and that, despite the expected growth in resource requirements with the qubit scaling, the NEB approach also accelerates convergence over the naive alternative.
Reinforcement learning (RL) has become a central component of post-training large language models (LLMs), yet little is understood about how RL adaptation is distributed across transformer layers. Existing approaches typically update all model parameters uniformly, implicitly assuming that every layer contributes similarly to the gains obtained during RL post-training. In this work, we challenge this assumption through a systematic layer-wise study of RL training. Surprisingly, we find that training a single transformer layer can recover most of the gains achieved by full-parameter RL training, and in some cases even surpass it. To quantify this phenomenon, we introduce the quantity layer contribution, which measures the fraction of full RL improvement recovered by training a layer in isolation. Across seven models spanning two model families (Qwen3, Qwen2.5), three RL algorithms (GRPO, GiGPO, Dr. GRPO), and multiple task domains including mathematical reasoning, code generation, and agentic decision-making, we observe a remarkably stable pattern: RL gains are highly concentrated in a small subset of, and in many cases even a single, transformer layers. More strikingly, the same structural pattern consistently emerges: high-contribution layers concentrate in the middle of the transformer stack, while layers near the input and output ends contribute substantially less. The resulting layer rankings remain strongly correlated across datasets, tasks, model families, and RL algorithms.
When should an AI system's answer be trusted? Formal proof assistants offer certainty but cannot reach most of the problem distribution; scalar LLM judges offer coverage but produce opaque scores that cannot be audited after the fact and are subject to the same coherence issues as any LLM. We present Theoria, a verification architecture that closes this gap. A candidate solution is rewritten into a sequence of typed state transitions, each licensed by an explicit justification, whether that be a citation, computation, or problem-given fact, and every transition is independently auditable. The foundational invariant is completeness of change: every difference between consecutive proof states must be accounted for, so hidden premises surface as unlicensed mutations rather than passing silently. On HLE-Verified Gold (185 text-only expert problems), Theoria certifies 105 at 91.4% strict precision (Wilson 95% CI [84.5%, 95.4%]). Every certification produces a human readable proof trace in which each step can be independently challenged. Holistic LLM judges achieve comparable precision at matched coverage but fail on different problems (Jaccard 0.14-0.36), making the approaches complementary. On 95 adversarial poisoned proofs across 15 domains, structured judges catch 94.7% versus 83.2% for holistic judging (p= 0.0017). The overall 11.5 pp gap concentrates in hidden premises (90.6% vs. 62.5%, a 28 pp difference) and fabricated citations (100% vs. 90%), the error classes where the formal analysis predicts an advantage; performance is identical on arithmetic and theorem-misapplication errors, where no advantage is predicted. On GPQA Diamond (n= 65), certified precision is 97.1% (Wilson CI [85.1%, 99.5%]).
In practice, most commercial LLM providers do not publicly release details of underlying LLM architectures. However, prior work has shown that given limited API access to an LLM (namely, top-$k$ logits and/or a logit bias function), one can recover certain architectural details of an LLM, such as the hidden dimension of the feed-forward network. Perhaps in response to these results, most commercial LLM providers have restricted their APIs to expose only the single logit for each decoded token, and they no longer give users the ability to bias logits. We show that even under current restrictive APIs, several architectural parameters are still recoverable. We present NightVision, an attack that uses restrictive black-box API access to estimate the hidden dimension, depth, and parameter count of an LLM. Algorithmically, NightVision relies on a novel common set prompting technique in which multiple prompts expose log probabilities for the same set of output tokens; a spectral analysis of these results is used to infer hidden dimension. NightVision additionally uses end-to-end time to first token (TTFT) measurements and the estimated hidden dimension to estimate depth and parameter count. We empirically evaluate NightVision on 32 open-source LLMs, recovering hidden dimension to within 23% average relative error across all models (9% on MoE models), and depth and parameter count to within 53% for models exceeding three billion parameters. We run extensive ablations to demonstrate how these accuracies scale with token budget and model properties. Overall, our results suggest that current LLM APIs are not sufficiently restricted to fully obfuscate the architectural details of their underlying models.
Quantum computing has emerged as a promising computational paradigm for machine learning (ML), with the potential to offer computational advantages over classical approaches. At this stage, the evidence supporting the performance and advantages of quantum machine learning (QML) models relative to classical models is insufficient. To address this gap, this paper presents an empirical study on the performance of QML models and their classical counterparts. We compare seven model pairs spanning supervised learning and reinforcement learning. Our results indicate that the evaluated quantum machine learning models do not yet surpass the classical baselines in overall prediction performance, policy stability, or training time. Nevertheless, QML remains a promising approach for filtering noise and controlling false positives. Our research findings summarize the challenges facing quantum machine learning across hardware environments, training efficiency, and convergence stability, providing a foundation for research into the robustness and parameter optimization of QML. This work is publicly available at https://github.com/Z-537-437/QML.
Deep learning has outgrown any single mathematical explanation. From Approximation to Emergence develops a unified, proof-oriented account of modern deep learning theory, tracing a path from the classical foundations of approximation, optimization, and generalization to the contemporary mechanisms of overparameterization, robustness, generative modeling, transformers, in-context learning, scaling laws, interpretability, alignment, and emergence. Rather than presenting isolated results, the book organizes a broad literature into a coherent research narrative: each theory is examined through the object it controls, the assumptions that make it valid, and the phenomena it leaves unexplained. Written for researchers, graduate students, and mathematically trained practitioners, this monograph offers a rigorous map of deep learning theory as it stands today: powerful, incomplete, and increasingly centered on the question of how learned mechanisms arise from scale, data, architecture, and training.
NA methylation profiling has become a powerful approach for central nervous system (CNS) tumor classification, yet important challenges remain regarding cross-cohort transferability, methodological correctness, and robust multiclass evaluation. In this work, we propose a novel and methodologically rigorous machine-learning approach for methylation-based CNS tumor classification that combines Sparse Random Projection for dimensionality reduction with multinomial logistic regression for classification. We evaluate the proposed approach in the same general experimental setting established by a widely used reference classifier. On the 2,801-sample reference cohort, our method achieves a mean accuracy of 96\% under stratified 3-fold cross-validation. On the independent 1,104-sample clinical evaluation cohort, it reaches 86\% accuracy at the 91-class level and 93\% when predictions are evaluated at the methylation class family level. These results improve upon the corresponding state-of-the-art reference figures of 82\% class-level concordance and 88\% family-level concordance, yielding absolute gains of approximately 4 and 5 percentage points, respectively. This improvement is clinically relevant: in a diagnostic setting, a 5-point increase in correct tumor classification can directly affect cancer subtype assignment and, in turn, influence treatment selection and downstream clinical decision-making. Our results show that the proposed model, grounded in stronger methodological practice in machine learning, consistently outperforms the previous state of the art across evaluation settings and can materially improve the reliability of CNS tumor classification.
Counterfactual explanations explain machine learning predictions by identifying minimal input changes that would alter a model's decision. Although many existing methods successfully generate prediction-changing alternatives, they often produce unrealistic or infeasible recommendations due to a lack of explicit mechanisms for incorporating domain knowledge and intervention constraints. Neuro-symbolic AI offers a promising direction by combining data-driven predictive models with symbolic reasoning capable of representing human-understandable rules and feasible actions. This paper presents PACE, a modular neuro-symbolic framework for generating feasibility-aware counterfactual explanations. The framework separates prediction and reasoning into two components: a neural predictive model for classification and a symbolic reasoning layer that enforces domain-specific constraints during counterfactual generation. By explicitly modeling feasible interventions, the framework produces explanations consistent with domain knowledge while remaining interpretable and actionable. The approach is model-agnostic and adaptable to domains requiring realistic decision support. A case study is conducted on the Adult Income dataset, combining a multilayer perceptron classifier with Answer Set Programming (ASP) rules encoding feasible modifications to education, occupation, and working hours while preserving immutable attributes. Results highlight the trade-off between counterfactual validity and plausibility and show that symbolic constraints yield explanations that better satisfy domain-specific feasibility requirements, illustrating the potential of neuro-symbolic methods for transparent, feasibility-aware counterfactual explanation in explainable AI.
Intrusion Detection Systems (IDSs) are essential for monitoring network traffic and identifying malicious activities in modern cyber-physical, Internet of Things (IoT), enterprise, and distributed network environments. However, developing reliable IDS models remains challenging because attack behaviors evolve over time, realistic datasets are difficult to obtain, traffic records may be incomplete, attack classes are often imbalanced, and privacy constraints limit centralized data collection. Recent advances in generative artificial intelligence (AI) and Federated Learning (FL) provide new opportunities to address these limitations. Generative models can support anomaly detection, synthetic traffic generation, data augmentation, data imputation, adversarial traffic generation, and IDS alert explanation. FL enables distributed IDS training without directly sharing local network traffic, making it suitable for privacy-sensitive and geographically distributed environments. This survey provides a structured review of generative AI and FL techniques for IDS. We first summarize representative IDS research directions, including adversarial machine learning, anomaly-based detection, IoT-oriented IDS, explainable IDS, and benchmark datasets. We then categorize generative AI applications in IDS according to model families and task objectives, covering autoencoder-based models, Generative Adversarial Networks (GANs), diffusion models, and Large Language Models (LLMs). Finally, we review emerging studies that integrate generative AI with FL-based IDS and discuss open challenges, including synthetic data quality, realistic traffic generation, dual-use adversarial risks, non-IID client distributions, communication-efficient model sharing, federated IDS benchmarking, and domain-specific LLMs for network security.
Creativity research has debated whether creativity is domain-specific (e.g., visual, writing, science), and if it is psychometrically separable from general intelligence. Both questions now apply to LLMs, but a unified benchmark of AI creativity remains elusive. We introduce AGC-Bench, an artificial general creativity benchmark built from a systematic review of the AI creativity literature (3,101 papers screened, 497 benchmarks identified), paired with an agentic harness that converts idiosyncratic codebases into HELM-standardized benchmarks. The first release covers 78 datasets spanning brainstorming, problem solving, STEM, narrative, figurative language, and humor. To address bias in LLM-as-judge, we apply Judge Response Theory -- a psychometric calibration of judge leniency/severity; we then fine-tune Qwen3-30B on the bias-corrected ratings of three frontier LLMs to produce AGC-Judge, an open-weight model that robustly scores new creativity benchmarks it was not trained on. Results reveal frontier models at the top of the AGC-Bench leaderboard, with open models close behind. LLMs show different creative strengths, ranking higher on some domains (e.g., writing) than others (e.g., scientific ideation). Extensive experiments yield three main findings. First, applying factor analysis across 83 LLMs, we recover a single creativity factor 'c', analogous to the 'g' factor of general intelligence, that explains 81.5% of variance, related to but separable from general knowledge/reasoning. Second, we show that prompting models to "be creative" boosts their performance far more than enabling reasoning, evidence that the benchmark tracks creativity over general ability. Third, on a human-matched subset, we find the top human still leads the top LLM on creativity. We release AGC-Bench with a public leaderboard, AGC-Judge, and human data as open infrastructure for measuring AI creativity at scale.
High-throughput RLHF systems often decouple rollout generation from policy optimization, leading to the use of stale rollouts during learner updates. In this work, we study the effect of such staleness in asynchronous GRPO. We make the behavior policy explicit in the GRPO surrogate objective and distinguish between the surrogate-gradient mapping used by the learner and the true total derivative of a distribution-dependent population objective. Under assumptions of local boundedness, distributional smoothness, and behavior-policy smoothness, we show that stale rollouts introduce a per-step surrogate-gradient bias of order O(S * eta), where S denotes the maximum rollout lag and eta denotes the learning rate. We further derive a conditional collapse-time scaling law: when within-cycle drift remains below a batch-level clipping radius, collapse is governed primarily by cumulative learner drift T * eta; when the stale-rollout constraint is active, stability instead depends explicitly on S * eta. This yields a two-constraint stability condition eta << min{R_batch / (S * G_upd), R_crit / (T * G_upd)}, explaining why the maximum stable learning rate may appear weakly dependent on staleness in the horizon-limited regime.
Presentation Attack Detection (PAD) serves as a crucial safeguard for face recognition systems against presentation attacks such as printed photos, replayed videos, and 3D masks. Despite significant progress, existing PAD models still struggle to generalize across unseen domains due to variations in sensors, lighting, and attack materials. Recent Vision-Language Models (VLMs) have shown strong generalization ability, yet their applications in PAD remain limited because learned prompts, typically optimized under class-label supervision, fail to explicitly align with fine-grained attack-relevant visual semantics. As a result, the learned representations often overfit domain-specific artifacts instead of capturing transferable attack cues. To address this, we propose Concept-Informed Prompts Guided Presentation Attack Detection (CPG-PAD), a framework that introduces model-level concept guidance into the prompt learning process. Specifically, we design a Visual Concept-driven Enhancement (VCE) module that employs eXplainable AI (XAI) techniques to automatically discover PAD-relevant visual concepts and generate concept-associated heatmaps providing localized fine-grained guidance. Guided by these heatmaps, a Prompt-based Concept Injection (PCI) mechanism integrates these concepts into the prompt space through a Visual-Prompt Decoder (VPD) and a concept-mapping loss, enabling prompts to align with the model's internal concept space. This design enables CPG-PAD to capture generalizable and domain-invariant attack cues while effectively suppressing dataset-specific biases. Extensive experiments across nine benchmark datasets demonstrate that CPG-PAD consistently achieves state-of-the-art cross-domain performance under multi-source, limited-source, and single-source settings.
Memory has emerged as a cornerstone of modern LLM-based agents, supporting their evolution from single-turn assistants to long-term collaborators. However, memory is not always beneficial: retrieved memories often induce a critical issue of sycophancy, causing agents to over-align with the user at the cost of factual accuracy or objective reasoning. Despite this emerging risk, existing memory benchmarks primarily evaluate whether memories are correctly stored, retrieved, or updated, while overlooking how retrieved memories influence downstream reasoning and decision-making. To bridge this gap, we propose MemSyco-Bench, a comprehensive benchmark for evaluating memory-induced sycophancy in agent systems. MemSyco-Bench measures when memory should influence a decision and how valid memory should be used. Specifically, it covers five tasks that assess whether agents can reject memory as factual evidence, respect its applicable scope, resolve conflicts between memory and objective evidence, track memory updates, and use valid memory for personalization. All related resources are collected for the community at https://github.com/XMUDeepLIT/MemSyco-Bench.
Vision-language models (VLMs) have achieved strong performance across diverse multimodal tasks, yet they remain vulnerable to unreliable reasoning. Existing self-correction methods mitigate these issues but typically rely on post-training or carefully engineered feedback, incurring high computational cost. In this work, we revisit this challenge through the lens of emotional cues, asking whether they can activate latent self-correction behaviors in VLMs without additional training. \textbf{We find that emotional signals serve as an effective trigger for self-correction, encouraging more cautious and reflective reasoning}. Motivated by this finding, we propose \escabstract (\textbf{\underline{E}}motional \textbf{\underline{S}}elf-\textbf{\underline{C}}orrection), a training-free self-correction framework. ESC introduces an external verifier that detects potentially incorrect initial responses and injects emotional feedback to encourage model to reflect, and produce a better revised response without additional training. Extensive experiments across safety, hallucination, vision-centric perception, and multimodal reasoning benchmarks show that ESC consistently improves reliability while preserving overall model utility. These results suggest that emotion can function not only as an ability to be recognized, but also as a practical control signal for scalable self-correction in VLMs. \textbf{We therefore believe that ESC provides a strong foundation for a new reliable human-like, emotion-integrated research direction.} Our project is publicly available at \textcolor{red}{https://genai4e.github.io/ESC/}.
This paper introduces Svarna, a free, open-source, web-based corpus workbench for modern Greek. Svarna integrates five databases covering various registers, institutional, literary, dialectal, social media, and historical, to provide a total of more than 507 million words and around 29 million sentences. This platform addresses the chronic gaps in Greek language technology. Although various corpus resources exist, they are scattered across different platforms, and in many cases, institutional access is restricted or they are no longer available online. Svarna integrates these resources into a single interface that can be used without logging in, installation, or specialized training. This system provides a concordancer with KWIC marking capabilities, frequency analysis including register-by-register normalization, collocation extraction using mutual information, a dictionary of 93 Greek discourse markers providing distribution profiles, text-level analysis tools including n-grams, variants, and collocation networks, register comparison using log-ratio, regular expression search, and an optional LLM layer for pragmatic annotation and free research mode. This platform is built upon SQLite FTS5 full-text indexes provided via a FastAPI backend, deployed as Docker containers on Azure, and released under the MIT license. Source code, build scripts, and deployment configurations are publicly available on GitHub. Users can add their own corpora and deploy their own instances. This document describes the system design, corpus structure, and use cases demonstrating the various queries supported by the platform. Svarna serves as the first step in exploring available data and is expected to lay the foundation for more comprehensive research in the future.
Most existing audio classification methods suppose that each query (testing) sample belongs to a class of support (training) samples, and misrecognize samples of unseen classes as seen classes (cannot reject samples of unseen classes). In this study, we propose a method for Few-shot Open-set Audio Classification (FOAC), which can recognize query samples of seen classes after updating the model using a few support samples, and meanwhile reject query samples from unseen classes. We design a model consisting of an encoder and a classifier. The encoder is the backbone of a ResNet used for extracting embeddings. The classifier consists of prototype generators of few-shot classes and open-set classes. Prototypes of few-shot classes are obtained by fusing the class-discriminative information of support and query embeddings and by assigning larger weighting coefficient to representative part of the support embeddings. One prototype is generated for open-set classes using the proposed prototype generator. The encoder is trained with abundant samples of base classes in supervised manner, and then the prototypes of base classes are generated under the supervision of a joint loss. The classifier is trained using a few samples of few-shot classes in a meta-training way. Three public datasets (LS-100, NSynth-100, and FSC-89) are used to assess the performance of our method. Experiments show that our method has advantage over prior methods in AUROC and accuracy. This advantage has statistical significance for most prior methods. Our method has lower computational complexity than most prior methods. The code is at https://github.com/Jessytan/FOAC-AIFP.
Acoustic imaging visualization is a core methodology in acoustics, enabling spatial analysis of sound sources and acoustic scenes. However, limited sensor availability in practical systems motivate approaches that enhance spatial resolution without increasing the hardware complexity. In this paper, we focus on upsampling virtually a tetrahedral 4-microphone array to a spherical 32-microphone array by estimating the covariance matrices of the channels employing deep learning techniques. Five neural network architectures are investigated for covariance upsampling for acoustic imaging using the real-world STARSS23 dataset. These models are developed to estimate a 32-microphone, time-frequency covariance matrix from a 4-microphone input covariance representation. The proposed architectures are based on 2D convolutional layers to capture the underlying spatial-spectral structure of covariance matrices, and are further enhanced with frequency dynamic convolution to model their frequency-dependent properties. The proposed architectures are evaluated in terms of root mean square error (RMSE) and using delay-and-sum beamforming acoustic imaging. Quantitative results show that all models outperform a random-guess baseline, which yields an RMSE of 0.548, with the best-performing architecture achieving an RMSE of 0.432. We analyze qualitatively the performance of the proposed models through beamforming heatmap visualizations derived from the 4-channel input covariance, the 32-channel ground truth, and the predicted 32-channel covariance matrices. These results demonstrate that covariance upsampling significantly enhances the effective performance of the 4-channel microphone array, producing sound maps that closely resemble those obtained with the 32-channel array.
We present RuleChef, a framework that uses large language models (LLMs) to generate executable rules for NLP tasks such as text classification, Named Entity Recognition (NER), or relation extraction. Rules are generated based on a task description and a set of labeled examples, then they are iteratively improved based both on additional examples and on human feedback overexisting rules. RuleChef can also be used to bootstrap rules using the observed input-output pairs from any existing model for a given task. LLMs are used only at learning time, synthesizing rules and iteratively patching them based on failures measured on a held-out split. The result of this process is a fast, deterministic, and inspectable rule system. Preliminary evaluation is performed on both classification and NER tasks. We release RuleChef as open-source software under an Apache 2.0
Private continual counting is a fundamental problem in differential privacy: given a binary stream of length $n$, where each $1$ corresponds to the contribution of one individual, the goal is to release all running counts while protecting the privacy of each individual. The standard algorithm is the binary tree mechanism, whose Gaussian-noise variant achieves expected $\ell_\infty$ error proportional to $\log^{3/2} n$ for approximate differential privacy. Whether this dependence on the stream length is necessary has remained a central open problem. In this work, we resolve the dependence on $n$ by proving that every differentially private mechanism for continual counting must incur expected $\ell_\infty$ error $Ω(\log^{3/2} n)$. This shows that the binary tree mechanism is asymptotically optimal in the approximate-DP setting. As a consequence, we also obtain a largest-possible separation between hereditary discrepancy and private $\ell_\infty$ error for linear queries, showing that the known general upper bound in terms of hereditary discrepancy has the optimal dependence on the number of queries.
World models are increasingly used in embodied intelligence and generative simulation, yet their scope remains ambiguous across communities. This tutorial presents a design-space view of world models as action-conditioned predictive models that estimate the future evolution of task-relevant observations or states. We categorize existing methods into observation-space and state-space world models, comparing their trade-offs in visual fidelity, spatial structure, physical interpretability, and control usability. We further introduce world action models, which connect predicted futures with executable robot actions, and summarize four representative paradigms: imagine-then-execute, video-feature-conditioned action prediction, joint video-action modeling, and auxiliary video prediction for policy learning. The goal of this tutorial is to clarify the conceptual scope of world (action) models and provide a structured taxonomy for embodied prediction and control.