The Inference Report

Spatial reasoning over three-dimensional scenes is a core capability for embodied intelligence, yet continuous model improvement remains bottlenecked by the cost of geometric annotation. The self-evolving paradigm offers a promising path, but its reliance on model consensus to construct pseudo-labels causes training to reinforce rather than correct the model's own geometric errors. We identify a property unique to 3D spatial reasoning that circumvents this limitation: ground truth is a deterministic consequence of the underlying geometry, computable exactly from point clouds and camera poses without any model involvement. Building on this insight, we present SpatialEvo, a self-evolving framework for 3D spatial reasoning, centered on the Deterministic Geometric Environment (DGE). The DGE formalizes 16 spatial reasoning task categories under explicit geometric validation rules and converts unannotated 3D scenes into zero-noise interactive oracles, replacing model consensus with objective physical feedback. A single shared-parameter policy co-evolves across questioner and solver roles under DGE constraints: the questioner generates physically valid spatial questions grounded in scene observations, while the solver derives precise answers against DGE-verified ground truth. A task-adaptive scheduler endogenously concentrates training on the model's weakest categories, producing a dynamic curriculum without manual design. Experiments across nine benchmarks demonstrate that SpatialEvo achieves the highest average score at both 3B and 7B scales, with consistent gains on spatial reasoning benchmarks and no degradation on general visual understanding.

While reinforcement learning with verifiable rewards (RLVR) significantly enhances LLM reasoning by optimizing the conditional distribution P(y|x), its potential is fundamentally bounded by the base model's existing output distribution. Optimizing the marginal distribution P(y) in the Pre-train Space addresses this bottleneck by encoding reasoning ability and preserving broad exploration capacity. Yet, conventional pre-training relies on static corpora for passive learning, leading to a distribution shift that hinders targeted reasoning enhancement. In this paper, we introduce PreRL (Pre-train Space RL), which applies reward-driven online updates directly to P(y). We theoretically and empirically validate the strong gradient alignment between log P(y) and log P(y|x), establishing PreRL as a viable surrogate for standard RL. Furthermore, we uncover a critical mechanism: Negative Sample Reinforcement (NSR) within PreRL serves as an exceptionally effective driver for reasoning. NSR-PreRL rapidly prunes incorrect reasoning spaces while stimulating endogenous reflective behaviors, increasing transition and reflection thoughts by 14.89x and 6.54x, respectively. Leveraging these insights, we propose Dual Space RL (DSRL), a Policy Reincarnation strategy that initializes models with NSR-PreRL to expand the reasoning horizon before transitioning to standard RL for fine-grained optimization. Extensive experiments demonstrate that DSRL consistently outperforms strong baselines, proving that pre-train space pruning effectively steers the policy toward a refined correct reasoning subspace.

As language models are increasingly deployed for complex autonomous tasks, their ability to reason accurately over longer horizons becomes critical. An essential component of this ability is planning and managing a long, complex chain-of-thought (CoT). We introduce LongCoT, a scalable benchmark of 2,500 expert-designed problems spanning chemistry, mathematics, computer science, chess, and logic to isolate and directly measure the long-horizon CoT reasoning capabilities of frontier models. Problems consist of a short input with a verifiable answer; solving them requires navigating a graph of interdependent steps that span tens to hundreds of thousands of reasoning tokens. Each local step is individually tractable for frontier models, so failures reflect long-horizon reasoning limitations. At release, the best models achieve <10% accuracy (GPT 5.2: 9.8%; Gemini 3 Pro: 6.1%) on LongCoT, revealing a substantial gap in current capabilities. Overall, LongCoT provides a rigorous measure of long-horizon reasoning, tracking the ability of frontier models to reason reliably over extended periods.

Evaluating LLMs is challenging, as benchmark scores often fail to capture models' real-world usefulness. Instead, users often rely on ``vibe-testing'': informal experience-based evaluation, such as comparing models on coding tasks related to their own workflow. While prevalent, vibe-testing is often too ad hoc and unstructured to analyze or reproduce at scale. In this work, we study how vibe-testing works in practice and then formalize it to support systematic analysis. We first analyze two empirical resources: (1) a survey of user evaluation practices, and (2) a collection of in-the-wild model comparison reports from blogs and social media. Based on these resources, we formalize vibe-testing as a two-part process: users personalize both what they test and how they judge responses. We then introduce a proof-of-concept evaluation pipeline that follows this formulation by generating personalized prompts and comparing model outputs using user-aware subjective criteria. In experiments on coding benchmarks, we find that combining personalized prompts and user-aware evaluation can change which model is preferred, reflecting the role of vibe-testing in practice. These findings suggest that formalized vibe-testing can serve as a useful approach for bridging benchmark scores and real-world experience.

Rhetorical questions are asked not to seek information but to persuade or signal stance. How large language models internally represent them remains unclear. We analyze rhetorical questions in LLM representations using linear probes on two social-media datasets with different discourse contexts, and find that rhetorical signals emerge early and are most stably captured by last-token representations. Rhetorical questions are linearly separable from information-seeking questions within datasets, and remain detectable under cross-dataset transfer, reaching AUROC around 0.7-0.8. However, we demonstrate that transferability does not simply imply a shared representation. Probes trained on different datasets produce different rankings when applied to the same target corpus, with overlap among the top-ranked instances often below 0.2. Qualitative analysis shows that these divergences correspond to distinct rhetorical phenomena: some probes capture discourse-level rhetorical stance embedded in extended argumentation, while others emphasize localized, syntax-driven interrogative acts. Together, these findings suggest that rhetorical questions in LLM representations are encoded by multiple linear directions emphasizing different cues, rather than a single shared direction.

While end-to-end Vision-Language-Action (VLA) models offer a promising paradigm for robotic manipulation, fine-tuning them on narrow control data often compromises the profound reasoning capabilities inherited from their base Vision-Language Models (VLMs). To resolve this fundamental trade-off, we propose HiVLA, a visual-grounded-centric hierarchical framework that explicitly decouples high-level semantic planning from low-level motor control. In high-level part, a VLM planner first performs task decomposition and visual grounding to generate structured plans, comprising a subtask instruction and a precise target bounding box. Then, to translate this plan into physical actions, we introduce a flow-matching Diffusion Transformer (DiT) action expert in low-level part equipped with a novel cascaded cross-attention mechanism. This design sequentially fuses global context, high-resolution object-centric crops and skill semantics, enabling the DiT to focus purely on robust execution. Our decoupled architecture preserves the VLM's zero-shot reasoning while allowing independent improvement of both components. Extensive experiments in simulation and the real world demonstrate that HiVLA significantly outperforms state-of-the-art end-to-end baselines, particularly excelling in long-horizon skill composition and the fine-grained manipulation of small objects in cluttered scenes.

LLM reasoning traces suffer from complex flaws -- *Step Internal Flaws* (logical errors, hallucinations, etc.) and *Step-wise Flaws* (overthinking, underthinking), which vary by sample. A natural approach would be to provide ground-truth labels to guide LLMs' reasoning. Contrary to intuition, we show that this yields no improvement in reasoning ability. We then propose CRAFT, a unified framework that mitigates both types of Step flaws, which builds a Reasoning Knowledge Graph (RKG) based on the consensus parts of multiple candidate traces, and synthesizes a high-quality trace through topological generation. Our approach improves label-prediction accuracy by 10+% on average, and consistently outperforms all baselines across both logical and mathematical reasoning benchmarks. Further, detailed benchmark evaluation proves that our method also improves the quality of LLMs' reasoning traces in multiple dimensions.

Given two symmetric positive-definite matrices $A, B \in \mathbb{R}^{n \times n}$, we study the spectral properties of the interpolation $A^{1-x} B^x$ for $0 \leq x \leq 1$. The presence of `common structures' in $A$ and $B$, eigenvectors pointing in a similar direction, can be investigated using this interpolation perspective. Generically, exact log-linearity of the operator norm $\|A^{1-x} B^x\|$ is equivalent to the existence of a shared eigenvector in the original matrices; stability bounds show that approximate log-linearity forces principal singular vectors to align with leading eigenvectors of both matrices. These results give rise to and provide theoretical justification for a multi-manifold learning framework that identifies common and distinct latent structures in multiview data.

While Large Language Models (LLMs) have empowered AI research agents to perform isolated scientific tasks, automating complex, real-world workflows, such as LLM training, remains a significant challenge. In this paper, we introduce TREX, a multi-agent system that automates the entire LLM training life-cycle. By orchestrating collaboration between two core modules-the Researcher and the Executor-the system seamlessly performs requirement analysis, open-domain literature and data research, formulation of training strategies, preparation of data recipes, and model training and evaluation. The multi-round experimental process is modeled as a search tree, enabling the system to efficiently plan exploration paths, reuse historical results, and distill high-level insights from iterative trials. To evaluate the capability of automated LLM training, we construct FT-Bench, a benchmark comprising 10 tasks derived from real-world scenarios, ranging from optimizing fundamental model capabilities to enhancing performance on domain-specific tasks. Experimental results demonstrate that the TREX agent consistently optimizes model performance on target tasks.

Sequential recommendation has become increasingly prominent in both academia and industry, particularly in e-commerce. The primary goal is to extract user preferences from historical interaction sequences and predict items a user is likely to engage with next. Recent advances have leveraged contrastive learning and graph neural networks to learn more expressive representations from interaction histories -- graphs capture relational structure between nodes, while ID-based representations encode item-specific information. However, few studies have explored multi-view contrastive learning between ID and graph perspectives to jointly improve user and item representations, especially in settings where only interaction data is available without auxiliary information. To address this gap, we propose Multi-View Contrastive learning for sequential recommendation (MVCrec), a framework that integrates complementary signals from both sequential (ID-based) and graph-based views. MVCrec incorporates three contrastive objectives: within the sequential view, within the graph view, and across views. To effectively fuse the learned representations, we introduce a multi-view attention fusion module that combines global and local attention mechanisms to estimate the likelihood of a target user purchasing a target item. Comprehensive experiments on five real-world benchmark datasets demonstrate that MVCrec consistently outperforms 11 state-of-the-art baselines, achieving improvements of up to 14.44\% in NDCG@10 and 9.22\% in HitRatio@10 over the strongest baseline. Our code and datasets are available at https://github.com/sword-Lz/MMCrec.

GUI grounding, which localizes interface elements from screenshots given natural language queries, remains challenging for small icons and dense layouts. Test-time zoom-in methods improve localization by cropping and re-running inference at higher resolution, but apply cropping uniformly across all instances with fixed crop sizes, ignoring whether the model is actually uncertain on each case. We propose \textbf{UI-Zoomer}, a training-free adaptive zoom-in framework that treats both the trigger and scale of zoom-in as a prediction uncertainty quantification problem. A confidence-aware gate fuses spatial consensus among stochastic candidates with token-level generation confidence to selectively trigger zoom-in only when localization is uncertain. When triggered, an uncertainty-driven crop sizing module decomposes prediction variance into inter-sample positional spread and intra-sample box extent, deriving a per-instance crop radius via the law of total variance. Extensive experiments on ScreenSpot-Pro, UI-Vision, and ScreenSpot-v2 demonstrate consistent improvements over strong baselines across multiple model architectures, achieving gains of up to +13.4\%, +10.3\%, and +4.2\% respectively, with no additional training required.

Large Language Models (LLMs) are now capable of generating highly fluent, human-like text. They enable many applications, but also raise concerns such as large scale spam, phishing, or academic misuse. While much work has focused on detecting LLM-generated text, only limited work has gone into understanding the stylistic differences between human-written and machine-generated text. In this work, we perform a large scale analysis of stylistic variation across human-written text and outputs from 11 LLMs spanning 8 different genres and 4 decoding strategies using Douglas Biber's set of lexicogrammatical and functional features. Our findings reveal insights that can guide intentional LLM usage. First, key linguistic differentiators of LLM-generated text seem robust to generation conditions (e.g., prompt settings to nudge them to generate human-like text, or availability of human-written text to continue the style); second, genre exerts a stronger influence on stylistic features than the source itself; third, chat variants of the models generally appear to be clustered together in stylistic space, and finally, model has a larger effect on the style than decoding strategy, with some exceptions. These results highlight the relative importance of model and genre over prompting and decoding strategies in shaping the stylistic behavior of machine-generated text.

Recent work suggests that (stochastic) gradient descent self-organizes near an instability boundary, shaping both optimization and the solutions found. Momentum and mini-batch gradients are widely used in practical deep learning optimization, but it remains unclear whether they operate in a comparable regime of instability. We demonstrate that SGD with momentum exhibits an Edge of Stochastic Stability (EoSS)-like regime with batch-size-dependent behavior that cannot be explained by a single momentum-adjusted stability threshold. Batch Sharpness (the expected directional mini-batch curvature) stabilizes in two distinct regimes: at small batch sizes it converges to a lower plateau $2(1-β)/η$, reflecting amplification of stochastic fluctuations by momentum and favoring flatter regions than vanilla SGD; at large batch sizes it converges to a higher plateau $2(1+β)/η$, where momentum recovers its classical stabilizing effect and favors sharper regions consistent with full-batch dynamics. We further show that this aligns with linear stability thresholds and discuss the implications for hyperparameter tuning and coupling.

Post-training adaptation of language models is commonly achieved through parameter updates or input-based methods such as fine-tuning, parameter-efficient adaptation, and prompting. In parallel, a growing body of work modifies internal activations at inference time to influence model behavior, an approach known as steering. Despite increasing use, steering is rarely analyzed within the same conceptual framework as established adaptation methods. In this work, we argue that steering should be regarded as a form of model adaptation. We introduce a set of functional criteria for adaptation methods and use them to compare steering approaches with classical alternatives. This analysis positions steering as a distinct adaptation paradigm based on targeted interventions in activation space, enabling local and reversible behavioral change without parameter updates. The resulting framing clarifies how steering relates to existing methods, motivating a unified taxonomy for model adaptation.

We present UMI-3D, a multimodal extension of the Universal Manipulation Interface (UMI) for robust and scalable data collection in embodied manipulation. While UMI enables portable, wrist-mounted data acquisition, its reliance on monocular visual SLAM makes it vulnerable to occlusions, dynamic scenes, and tracking failures, limiting its applicability in real-world environments. UMI-3D addresses these limitations by introducing a lightweight and low-cost LiDAR sensor tightly integrated into the wrist-mounted interface, enabling LiDAR-centric SLAM with accurate metric-scale pose estimation under challenging conditions. We further develop a hardware-synchronized multimodal sensing pipeline and a unified spatiotemporal calibration framework that aligns visual observations with LiDAR point clouds, producing consistent 3D representations of demonstrations. Despite maintaining the original 2D visuomotor policy formulation, UMI-3D significantly improves the quality and reliability of collected data, which directly translates into enhanced policy performance. Extensive real-world experiments demonstrate that UMI-3D not only achieves high success rates on standard manipulation tasks, but also enables learning of tasks that are challenging or infeasible for the original vision-only UMI setup, including large deformable object manipulation and articulated object operation. The system supports an end-to-end pipeline for data acquisition, alignment, training, and deployment, while preserving the portability and accessibility of the original UMI. All hardware and software components are open-sourced to facilitate large-scale data collection and accelerate research in embodied intelligence: \href{https://umi-3d.github.io}{https://umi-3d.github.io}.

On-policy knowledge distillation (OPD) trains a student on its own rollouts under token-level supervision from a teacher. Not all token positions matter equally, but existing views of token importance are incomplete. We ask a direct question: which tokens carry the most useful learning signal in OPD? Our answer is that informative tokens come from two regions: positions with high student entropy, and positions with low student entropy plus high teacher--student divergence, where the student is overconfident and wrong. Empirically, student entropy is a strong first-order proxy: retaining $50\%$ of tokens with entropy-based sampling matches or exceeds all-token training while reducing peak memory by up to $47\%$. But entropy alone misses a second important region. When we isolate low-entropy, high-divergence tokens, training on fewer than $10\%$ of all tokens nearly matches full-token baselines, showing that overconfident tokens carry dense corrective signal despite being nearly invisible to entropy-only rules. We organize these findings with TIP (Token Importance in on-Policy distillation), a two-axis taxonomy over student entropy and teacher--student divergence, and give a theoretical explanation for why entropy is useful yet structurally incomplete. This view motivates type-aware token selection rules that combine uncertainty and disagreement. We validate this picture across three teacher--student pairs spanning Qwen3, Llama, and Qwen2.5 on MATH-500 and AIME 2024/2025, and on the DeepPlanning benchmark for long-horizon agentic planning, where Q3-only training on $<$$20\%$ of tokens surpasses full-token OPD. Our experiments are implemented by extending the OPD repository https://github.com/HJSang/OPSD_OnPolicyDistillation, which supports memory-efficient distillation of larger models under limited GPU budgets.

We introduce Multistage Conditional Compositional Optimization (MCCO) as a new paradigm for decision-making under uncertainty that combines aspects of multistage stochastic programming and conditional stochastic optimization. MCCO minimizes a nest of conditional expectations and nonlinear cost functions. It has numerous applications and arises, for example, in optimal stopping, linear-quadratic regulator problems, distributionally robust contextual bandits, as well as in problems involving dynamic risk measures. The naïve nested sampling approach for MCCO suffers from the curse of dimensionality familiar from scenario tree-based multistage stochastic programming, that is, its scenario complexity grows exponentially with the number of nests. We develop new multilevel Monte Carlo techniques for MCCO whose scenario complexity grows only polynomially with the desired accuracy.

Resolving and rewriting references is fundamental in programming languages. Motivated by a real-world decompilation task, we abstract reference rewriting into the problems of direct and indirect indexing by permutation. We create synthetic benchmarks for these tasks and show that well-known sequence-to-sequence machine learning architectures are struggling on these benchmarks. We introduce new sequence-to-sequence architectures for both problems. Our measurements show that our architectures outperform the baselines in both robustness and scalability: our models can handle examples that are ten times longer compared to the best baseline. We measure the impact of our architecture in the real-world task of decompiling switch statements, which has an indexing subtask. According to our measurements, the extended model decreases the error rate by 42%. Multiple ablation studies show that all components of our architectures are essential.

This paper provides a systematic comparison between Fitted Dynamic Programming (DP), where demand is estimated from data, and Reinforcement Learning (RL) methods in finite-horizon dynamic pricing problems. We analyze their performance across environments of increasing structural complexity, ranging from a single typology benchmark to multi-typology settings with heterogeneous demand and inter-temporal revenue constraints. Unlike simplified comparisons that restrict DP to low-dimensional settings, we apply dynamic programming in richer, multi-dimensional environments with multiple product types and constraints. We evaluate revenue performance, stability, constraint satisfaction behavior, and computational scaling, highlighting the trade-offs between explicit expectation-based optimization and trajectory-based learning.

Efficiency and safety of Large Language Models (LLMs), among other factors, rely on the quality of tokenization. A good tokenizer not only improves inference speed and language understanding but also provides extra defense against jailbreak attacks and lowers the risk of hallucinations. In this work, we investigate the efficiency of code tokenization, in particular from the perspective of data source diversity. We demonstrate that code tokenizers are prone to producing unused, and thus under-trained, tokens due to the imbalance in repository and language diversity in the training data, as well as the dominance of source-specific, repetitive tokens that are often unusable in future inference. By modifying the BPE objective and introducing merge skipping, we implement different techniques under the name Source-Attributed BPE (SA-BPE) to regularize BPE training and minimize overfitting, thereby substantially reducing the number of under-trained tokens while maintaining the same inference procedure as with regular BPE. This provides an effective tool suitable for production use.

Parameter space is not function space for neural network architectures. This fact, investigated as early as the 1990s under terms such as ``reverse engineering," or ``parameter identifiability", has led to the natural question of parameter space symmetries\textemdash the study of distinct parameters in neural architectures which realize the same function. Indeed, the quotient space obtained by identifying parameters giving rise to the same function, called the \textit{neuromanifold}, has been shown in some cases to have rich geometric properties, impacting optimization dynamics. Thus far, techniques towards complete classifications have required the analyticity of the activation function, notably excising the important case of ReLU. Here, in contrast, we exploit the non-differentiability of the ReLU activation to provide a complete classification of the symmetries in the shallow case.

Fairness in algorithmic decision-making is often defined in the predictive space, where predictive performance - used as a proxy for decision-maker (DM) utility - is traded off against prediction-based fairness notions, such as demographic parity or equality of opportunity. This perspective, however, ignores how predictions translate into decisions and ultimately into utilities and welfare for both DM and decision subjects (DS), as well as their allocation across social-salient groups. In this paper, we propose a multi-stakeholder framework for fair algorithmic decision-making grounded in welfare economics and distributive justice, explicitly modeling the utilities of both the DM and DS, and defining fairness via a social planner's utility that captures inequalities in DS utilities across groups under different justice-based fairness notions (e.g., Egalitarian, Rawlsian). We formulate fair decision-making as a post-hoc multi-objective optimization problem, characterizing the achievable performance-fairness trade-offs in the two-dimensional utility space of DM utility and the social planner's utility, under different decision policy classes (deterministic vs. stochastic, shared vs. group-specific). Using the proposed framework, we then identify conditions (in terms of the stakeholders' utilities) under which stochastic policies are more optimal than deterministic ones, and empirically demonstrate that simple stochastic policies can yield superior performance-fairness trade-offs by leveraging outcome uncertainty. Overall, we advocate a shift from prediction-centric fairness to a transparent, justice-based, multi-stakeholder approach that supports the collaborative design of decision-making policies.

Recent advances in Generative Artificial Intelligence, particularly Large Language Models (LLMs), have stimulated growing interest in automating or assisting Business Process Modeling tasks using natural language. Several approaches have been proposed to transform textual process descriptions into BPMN and related workflow models. However, the extent to which these approaches effectively support complex process modeling in organizational settings remains unclear. This article presents a literature review of AI-driven methods for transforming natural language into BPMN process models, with a particular focus on the role of LLMs. Following a structured review strategy, relevant studies were identified and analyzed to classify existing approaches, examine how LLMs are integrated into text-to-model pipelines, and investigate the evaluation practices used to assess generated models. The analysis reveals a clear shift from rule-based and traditional NLP pipelines toward LLM-based architectures that rely on prompt engineering, intermediate representations, and iterative refinement mechanisms. While these approaches significantly expand the capabilities of automated process model generation, the literature also exposes persistent challenges related to semantic correctness, evaluation fragmentation, reproducibility, and limited validation in real-world organizational contexts. Based on these findings, this review identifies key research gaps and discusses promising directions for future research, including the integration of contextual knowledge through Retrieval-Augmented Generation (RAG), its integration with LLMs, the development of interactive modeling architectures, and the need for more comprehensive and standardized evaluation frameworks.

Reinforcement learning has shown promise for automating power-grid operation tasks such as topology control and congestion management. However, its deployment in real-world power systems remains limited by strict safety requirements, brittleness under rare disturbances, and poor generalization to unseen grid topologies. In safety-critical infrastructure, catastrophic failures cannot be tolerated, and learning-based controllers must operate within hard physical constraints. This paper proposes a safety-constrained hierarchical control framework for power-grid operation that explicitly decouples long-horizon decision-making from real-time feasibility enforcement. A high-level reinforcement learning policy proposes abstract control actions, while a deterministic runtime safety shield filters unsafe actions using fast forward simulation. Safety is enforced as a runtime invariant, independent of policy quality or training distribution. The proposed framework is evaluated on the Grid2Op benchmark suite under nominal conditions, forced line-outage stress tests, and zero-shot deployment on the ICAPS 2021 large-scale transmission grid without retraining. Results show that flat reinforcement learning policies are brittle under stress, while safety-only methods are overly conservative. In contrast, the proposed hierarchical and safety-aware approach achieves longer episode survival, lower peak line loading, and robust zero-shot generalization to unseen grids. These results indicate that safety and generalization in power-grid control are best achieved through architectural design rather than increasingly complex reward engineering, providing a practical path toward deployable learning-based controllers for real-world energy systems.

Product bundling boosts e-commerce revenue by recommending complementary item combinations. However, existing methods face two critical challenges: (1) collaborative filtering approaches struggle with cold-start items owing to dependency on historical interactions, and (2) LLMs lack inherent capability to model interactive graph directly. To bridge this gap, we propose a dual-enhancement method that integrates interactive graph learning and LLM-based semantic understanding for product bundling. Our method introduces a graph-to-text paradigm, which leverages a Dynamic Concept Binding Mechanism (DCBM) to translate graph structures into natural language prompts. The DCBM plays a critical role in aligning domain-specific entities with LLM tokenization, enabling effective comprehension of combinatorial constraints. Experiments on three benchmarks (POG, POG_dense, Steam) demonstrate 6.3%-26.5% improvements over state-of-the-art baselines.

Reconstructing 3D representations from 2D inputs is a fundamental task in computer vision and graphics, serving as a cornerstone for understanding and interacting with the physical world. While traditional methods achieve high fidelity, they are limited by slow per-scene optimization or category-specific training, which hinders their practical deployment and scalability. Hence, generalizable feed-forward 3D reconstruction has witnessed rapid development in recent years. By learning a model that maps images directly to 3D representations in a single forward pass, these methods enable efficient reconstruction and robust cross-scene generalization. Our survey is motivated by a critical observation: despite the diverse geometric output representations, ranging from implicit fields to explicit primitives, existing feed-forward approaches share similar high-level architectural patterns, such as image feature extraction backbones, multi-view information fusion mechanisms, and geometry-aware design principles. Consequently, we abstract away from these representation differences and instead focus on model design, proposing a novel taxonomy centered on model design strategies that are agnostic to the output format. Our proposed taxonomy organizes the research directions into five key problems that drive recent research development: feature enhancement, geometry awareness, model efficiency, augmentation strategies and temporal-aware models. To support this taxonomy with empirical grounding and standardized evaluation, we further comprehensively review related benchmarks and datasets, and extensively discuss and categorize real-world applications based on feed-forward 3D models. Finally, we outline future directions to address open challenges such as scalability, evaluation standards, and world modeling.

Modern distributed systems generate large volumes of logs that can be analyzed to support essential AIOps tasks such as fault diagnosis, which plays a crucial role in maintaining system reliability. Most existing approaches rely on log-based models that treat logs as linear sequences of events. However, such representations discard the structural context between events that are often present in execution logs, such as parent-child dependencies, fan-out (branching), or temporal features. To better capture these relationships, recent works on Graph Neural Networks (GNNs) suggest that representing logs as graphs offers a promising alternative. Building on these observations, this paper conducts a comparative study of log-based encoder architectures (e.g., BERT) and graph-based models (e.g., GNNs) for automated fault diagnosis. We evaluate our models on TraceBench, a trace-oriented log dataset, and on BGL, a more traditional system log dataset, covering both anomaly detection and fault type classification. Our results show that graph-only models fail to outperform encoder baselines. However, integrating learned representations from log encoders into graph-based models achieves the strongest overall performance. These findings highlight conditions under which graph-augmented architectures can outperform traditional log-based approaches.

Under interpolation-type assumptions such as the strong growth condition, stochastic optimization methods can attain convergence rates comparable to full-batch methods, but their performance, particularly for SGD, remains highly sensitive to step-size selection. To address this issue, we propose a unified stochastic trust-region framework that eliminates manual step-size tuning and extends naturally to equality-constrained problems. For unconstrained optimization, we develop a first-order stochastic trust-region algorithm and show that, under the strong growth condition, it achieves an iteration and stochastic first-order oracle complexity of $O(\varepsilon^{-2} \log(1/\varepsilon))$ for finding an $\varepsilon$-stationary point. For equality-constrained problems, we introduce a quadratic-penalty-based stochastic trust-region method with penalty parameter $μ$, and establish an iteration and oracle complexity of $O(\varepsilon^{-4} \log(1/\varepsilon))$ to reach an $\varepsilon$-stationary point of the penalized problem, corresponding to an $O(\varepsilon)$-approximate KKT point of the original constrained problem. Numerical experiments on deep neural network training and orthogonally constrained subspace fitting demonstrate that the proposed methods achieve performance comparable to well-tuned stochastic baselines, while exhibiting stable optimization behavior and effectively handling hard constraints without manual learning-rate scheduling.

Multimodal Continual Instruction Tuning (MCIT) is essential for sequential task adaptation of Multimodal Large Language Models (MLLMs) but is severely restricted by catastrophic forgetting. While existing literature focuses on the reasoning language backbone, in this work, we expose a critical yet neglected dual-forgetting phenomenon across both perception drift in Cross-modal Projection Space and reasoning collapse in Low-rank Parameter Space. To resolve this, we present \textbf{MAny} (\textbf{M}erge \textbf{Any}thing), a framework that merges task-specific knowledge through \textbf{C}ross-modal \textbf{P}rojection \textbf{M}erging (\textbf{CPM}) and \textbf{L}ow-rank \textbf{P}arameter \textbf{M}erging (\textbf{LPM}). Specifically, CPM recovers perceptual alignment by adaptively merging cross-modal visual representations via visual-prototype guidance, ensuring accurate feature recovery during inference. Simultaneously, LPM eliminates mutual interference among task-specific low-rank modules by recursively merging low-rank weight matrices. By leveraging recursive least squares, LPM provides a closed-form solution that mathematically guarantees an optimal fusion trajectory for reasoning stability. Notably, MAny operates as a training-free paradigm that achieves knowledge merging via efficient CPU-based algebraic operations, eliminating additional gradient-based optimization beyond initial tuning. Our extensive evaluations confirm the superior performance and robustness of MAny across multiple MLLMs and benchmarks. Specifically, on the UCIT benchmark, MAny achieves significant leads of up to 8.57\% and 2.85\% in final average accuracy over state-of-the-art methods across two different MLLMs, respectively.

Commit signing is widely promoted as a foundation of software supply-chain security, yet prior work has studied it through the lens of individual repositories or curated project samples, missing the broader picture of how developers behave across an entire platform. Grounded in replicability theory, we vary the sampling unit from repositories to individual developers, following 71,694 active GitHub users, defined as accounts that have authored at least one commit, across all their repositories and their entire commit history, spanning 16 million commits and 874,198 repositories. This platform-wide, user-centric view reveals a fundamental gap that repository sampling cannot detect. The ecosystem's apparent high signing adoption rate is an illusion. Once platform-generated signatures are excluded, fewer than 6% of developers have ever signed a commit themselves, and the vast majority of apparent signers have never signed outside a web browser. Among the minority who do sign locally, signing rarely persists over time or across repositories, and roughly one in eight developer-managed signatures fails verification because signing keys are never uploaded to GitHub. Examining the key registry, we find that expired keys are almost never revoked and more than a quarter of users carry at least one dead key. Together, these findings reveal that commit signing as practiced today cannot serve as a dependable provenance signal at ecosystem scale, and we offer concrete recommendations for closing that gap.

We promote in this paper the processing of radar data in the frequency domain to achieve higher robustness against noise and structural errors, especially in comparison to feature-based methods. This holds also for high dynamics in the scene, i.e., ego-motion of the vehicle with the sensor plus the presence of an unknown number of other moving objects. In addition to the high robustness, the processing in the frequency domain has the so far neglected advantage that the underlying correlation based methods used for, e.g., registration, provide information about all moving structures in the scene. A typical automotive application case is overtaking maneuvers, which in the context of autonomous racing are used here as a motivating example. Initial experiments and results with Fourier SOFT in 2D (FS2D) are presented that use the Boreas dataset to demonstrate radar-only-odometry, i.e., radar-odometry without sensor-fusion, to support our arguments.

Supervised Fine-Tuning (SFT) of large language models often suffers from task interference and catastrophic forgetting. Recent approaches alleviate this issue by isolating task-critical parameters during training. However, these methods represent a static solution to a dynamic problem, assuming that parameter importance remains fixed once identified. In this work, we empirically demonstrate that parameter importance exhibits temporal drift over the course of training. To address this, we propose Evolving Parameter Isolation (EPI), a fine-tuning framework that adapts isolation decisions based on online estimates of parameter importance. Instead of freezing a fixed subset of parameters, EPI periodically updates isolation masks using gradient-based signals, enabling the model to protect emerging task-critical parameters while releasing outdated ones to recover plasticity. Experiments on diverse multi-task benchmarks demonstrate that EPI consistently reduces interference and forgetting compared to static isolation and standard fine-tuning, while improving overall generalization. Our analysis highlights the necessity of synchronizing isolation mechanisms with the evolving dynamics of learning diverse abilities.

Memory-based self-evolution has emerged as a promising paradigm for coding agents. However, existing approaches typically restrict memory utilization to homogeneous task domains, failing to leverage the shared infrastructural foundations, such as runtime environments and programming languages, that exist across diverse real-world coding problems. To address this limitation, we investigate \textbf{Memory Transfer Learning} (MTL) by harnessing a unified memory pool from heterogeneous domains. We evaluate performance across 6 coding benchmarks using four memory representations, ranging from concrete traces to abstract insights. Our experiments demonstrate that cross-domain memory improves average performance by 3.7\%, primarily by transferring meta-knowledge, such as validation routines, rather than task-specific code. Importantly, we find that abstraction dictates transferability; high-level insights generalize well, whereas low-level traces often induce negative transfer due to excessive specificity. Furthermore, we show that transfer effectiveness scales with the size of the memory pool, and memory can be transferred even between different models. Our work establishes empirical design principles for expanding memory utilization beyond single-domain silos. Project page: https://memorytransfer.github.io/

Diffusion language models have recently emerged as a leading alternative to standard language models, due to their ability for bidirectional attention and parallel text generation. In this work, we explore variants for their use in speech recognition. Specifically, we introduce a comprehensive guide to incorporating masked diffusion language models (MDLM) and uniform-state diffusion models (USDMs) for rescoring ASR hypotheses. Additionally, we design a new joint-decoding method that combines CTC and USDM by integrating the framewise probability distributions derived from CTC with the labelwise probability distributions computed by USDM at each decoding step, thereby generating new candidates that combine strong language knowledge from USDM and acoustic information from CTC. Our findings reveal that USDM, as well as MDLM, can significantly improve the accuracy of recognized text. We publish all our code and recipes.

The lack of transparency about code datasets used to train large language models (LLMs) makes it difficult to detect, evaluate, and mitigate data leakage. We present a perturbation-based method to quantify memorization advantage in code LLMs, defined as the performance gap between likely seen and unseen inputs. We evaluate 8 open-source code LLMs on 19 benchmarks across four task families: code generation, code understanding, vulnerability detection, and bug fixing. Sensitivity patterns vary widely across models and tasks. For example, StarCoder reaches high sensitivity on some benchmarks (up to 0.8), while QwenCoder remains lower (mostly below 0.4), suggesting differences in generalization behavior. Task categories also differ: code summarization tends to show low sensitivity, whereas test generation is substantially higher. We then analyze two widely discussed benchmarks, CVEFixes and Defects4J, often suspected of leakage. Contrary to common concerns, both show low memorization advantage across models: CVEFixes remains below 0.1, and Defects4J is lower than other program repair benchmarks. These results suggest that, for these datasets, models may rely more on learned generalization than direct memorization. Overall, our findings provide evidence that memorization risk is highly task- and model-dependent, and highlight the need for stronger evaluation protocols, especially in security-focused settings.

Physical reasoning over visual inputs demands tight integration of visual perception, domain knowledge, and multi-step symbolic inference. Yet even state-of-the-art Vision Language Models (VLMs) fall far short of human performance on physics benchmarks. While post-training algorithms such as Supervised Fine-Tuning (SFT) and Group Relative Policy Optimization (GRPO) have demonstrated strong reasoning gains in language models, how reward design shapes VLM physical reasoning behavior remains poorly understood. We present a systematic reward ablation study for GRPO-based VLM training on physical reasoning. We compare four reward signals of increasing semantic richness: format compliance, answer accuracy, a composite rubric reward (answer correctness, physics principle identification, and unit consistency), and a novel internal reward derived from model attention weights over input image regions. We evaluate on PhyX, a 3,000-problem benchmark spanning six physics domains and six reasoning types across multiple-choice and open-ended formats, using IBM Granite Vision 3.3 (2B). Across both formats, GRPO with accuracy-based rewards outperforms SFT on most domains, though gains vary substantially by reward type and domain. Reward design does not uniformly improve performance. Instead, it induces domain-specific reasoning behaviors. Accuracy-based rewards provide the strongest overall gains. Rubric rewards improve structured reasoning quality without consistent accuracy improvements. Attention-based rewards enhance spatial reasoning while degrading performance in symbolic domains. Our internal attention-weight reward requires no spatial annotations and improves spatial relation accuracy from 0.27 to 0.50, suggesting that supervising where the model attends during generation is a promising direction for visually grounded physical reasoning.

Accurate methane sorption prediction across heterogeneous coal ranks requires models that combine thermodynamic consistency, efficient knowledge transfer across data-scarce geological systems, and calibrated uncertainty estimates, capabilities that are rarely addressed together in existing frameworks. We present a physics-informed transfer learning framework that adapts a hydrogen sorption PINN to methane sorption prediction via Elastic Weight Consolidation, coal-specific feature engineering, and a three-phase curriculum that progressively balances transfer preservation with thermodynamic fine-tuning. Trained on 993 equilibrium measurements from 114 independent coal experiments spanning lignite to anthracite, the framework achieves R2 = 0.932 on held-out coal samples, a 227% improvement over pressure-only classical isotherms, while hydrogen pre-training delivers 18.9% lower RMSE and 19.4% faster convergence than random initialization. Five Bayesian uncertainty quantification approaches reveal a systematic divergence in performance across physics-constrained architectures. Monte Carlo Dropout achieves well-calibrated uncertainty at minimal overhead, while deep ensembles, regardless of architectural diversity or initialization strategy, exhibit performance degradation because shared physics constraints narrow the admissible solution manifold. SHAP and ALE analyses confirm that learned representations remain physically interpretable and aligned with established coal sorption mechanisms: moisture-volatile interactions are most influential, pressure-temperature coupling captures thermodynamic co-dependence, and features exhibit non-monotonic effects. These results identify Monte Carlo Dropout as the best-performing UQ method in this physics-constrained transfer learning framework, and demonstrate cross-gas transfer learning as a data-efficient strategy for geological material modeling.

Large language models (LLMs) are prone to generating factually incorrect outputs. Recent work has applied conformal prediction to provide uncertainty estimates and statistical guarantees for the factuality of LLM generations. However, existing approaches are typically not prompt-adaptive, limiting their ability to capture input-dependent variability. As a result, they may filter out too few items (leading to over-coverage) or too many (under-coverage) for a given task or prompt. We propose an adaptive conformal prediction approach that extends conformal score transformation methods to LLMs, with applications to long-form generation and multiple-choice question answering. This enables prompt-dependent calibration, retaining marginal coverage guarantees while improving conditional coverage. In addition, the approach naturally supports selective prediction, allowing unreliable claims or answer choices to be filtered out in downstream applications. We evaluate our approach on multiple white-box models across diverse domains and show that it significantly outperforms existing baselines in terms of conditional coverage.

Objective: Investigate whether hypnogram 'realism' can be used to guide an unsupervised method for handling arbitrary types of signal degradation in mobile sleep monitoring. Approach: Combining a pretrained, state-of-the-art 'u-sleep' model with a 'discriminator' network, we align features from a target domain with a feature space learned during pretraining. To test the approach, we distort the source domain with realistic signal degradations, to see how well the method can adapt to different types of degradation. We compare the performance of the resulting model with best-case models designed in a supervised manner for each type of transfer. Main Results: Depending on the type of distortion, we find that the unsupervised approach can increase Cohen's kappa with as little as 0.03 and up to 0.29, and that for all transfers, the method does not decrease performance. However, the approach never quite reaches the estimated theoretical optimal performance, and when tested on a real-life domain mismatch between two sleep studies, the benefit was insignificant. Significance: 'Discriminator-guided fine tuning' is an interesting approach to handling signal degradation for 'in the wild' sleep monitoring, with some promise. In particular, what it says about sleep data in general is interesting. However, more development will be necessary before using it 'in production'.

Predicting the effects of perturbations in-silico on cell state can identify drivers of cell behavior at scale and accelerate drug discovery. However, modeling challenges remain due to the inherent heterogeneity of single cell gene expression and the complex, latent gene dependencies. Here, we present PRiMeFlow, an end-to-end flow matching based approach to directly model the effects of genetic and small molecule perturbations in the gene expression space. The distribution-fitting approach taken by PRiMeFlow enables it to accurately approximate the empirical distribution of single-cell gene expression, which we demonstrate through extensive benchmarking inside PerturBench. Through ablation studies, we also validate important model design choices such as operating in gene expression space and parameterizing the velocity field with a U-Net architecture. The PRiMeFlow architecture was used as the basis for the model that won the Generalist Prize in the first ARC Virtual Cell Challenge.

Antibody lead optimization is inherently a multi-objective challenge in drug discovery. Achieving a balance between different drug-like properties is crucial for the development of viable candidates, and this search becomes exponentially challenging as desired properties grow. The ever-growing zoo of sophisticated in silico tools for predicting antibody properties calls for an efficient joint optimization procedure to overcome resource-intensive sequential filtering pipelines. We present BOAT, a versatile Bayesian optimization framework for multi-property antibody engineering. Our `plug-and-play' framework couples uncertainty-aware surrogate modeling with a genetic algorithm to jointly optimize various predicted antibody traits while enabling efficient exploration of sequence space. Through systematic benchmarking against genetic algorithms and newer generative learning approaches, we demonstrate competitive performance with state-of-the-art methods for multi-objective protein optimization. We identify clear regimes where surrogate-driven optimization outperforms expensive generative approaches and establish practical limits imposed by sequence dimensionality and oracle costs.

Open-world Question Answering (OW-QA) over knowledge graphs (KGs) aims to answer questions over incomplete or evolving KGs. Traditional KGQA assumes a closed world where answers must exist in the KG, limiting real-world applicability. In contrast, open-world QA requires inferring missing knowledge based on graph structure and context. Large language models (LLMs) excel at language understanding but lack structured reasoning. Graph neural networks (GNNs) model graph topology but struggle with semantic interpretation. Existing systems integrate LLMs with GNNs or graph retrievers. Some support open-world QA but rely on structural embeddings without semantic grounding. Most assume observed paths or complete graphs, making them unreliable under missing links or multi-hop reasoning. We present GLOW, a hybrid system that combines a pre-trained GNN and an LLM for open-world KGQA. The GNN predicts top-k candidate answers from the graph structure. These, along with relevant KG facts, are serialized into a structured prompt (e.g., triples and candidates) to guide the LLM's reasoning. This enables joint reasoning over symbolic and semantic signals, without relying on retrieval or fine-tuning. To evaluate generalization, we introduce GLOW-BENCH, a 1,000-question benchmark over incomplete KGs across diverse domains. GLOW outperforms existing LLM-GNN systems on standard benchmarks and GLOW-BENCH, achieving up to 53.3% and an average 38% improvement. GitHub code and data are available.

Synthetic data is a standard component in training large language models, yet systematic comparisons across design dimensions, including rephrasing strategy, generator model, and source data, remain absent. We conduct extensive controlled experiments, generating over one trillion tokens, to identify critical factors in rephrasing web text into synthetic pretraining data. Our results reveal that structured output formats, such as tables, math problems, FAQs, and tutorials, consistently outperform both curated web baselines and prior synthetic methods. Notably, increasing the size of the generator model beyond 1B parameters provides no additional benefit. Our analysis also demonstrates that the selection of the original data used for mixing substantially influences performance. By applying our findings, we develop \textbf{\textsc{FinePhrase}}, a 486-billion-token open dataset of rephrased web text. We show that \textsc{FinePhrase} outperforms all existing synthetic data baselines while reducing generation costs by up to 30 times. We provide the dataset, all prompts, and the generation framework to the research community.

We study value adaptation in offline-to-online reinforcement learning under general function approximation. Starting from an imperfect offline pretrained $Q$-function, the learner aims to adapt it to the target environment using only a limited amount of online interaction. We first characterize the difficulty of this setting by establishing a minimax lower bound, showing that even when the pretrained $Q$-function is close to optimal $Q^\star$, online adaptation can be no more efficient than pure online RL on certain hard instances. On the positive side, under a novel structural condition on the offline-pretrained value functions, we propose O2O-LSVI, an adaptation algorithm with problem-dependent sample complexity that provably improves over pure online RL. Finally, we complement our theory with neural-network experiments that demonstrate the practical effectiveness of the proposed method.

As AI moves from data centers to robots and wearables, scaling ever-larger models becomes insufficient. Physical AI operates under tight latency, energy, privacy, and reliability constraints, and its performance depends not only on model capacity but also on how signals are acquired through controllable sensors in dynamic environments. We present Artificial Tripartite Intelligence (ATI), a bio-inspired, sensor-first architectural contract for physical AI. ATI is tripartite at the systems level: a Brainstem (L1) provides reflexive safety and signal-integrity control, a Cerebellum (L2) performs continuous sensor calibration, and a Cerebral Inference Subsystem spanning L3/L4 supports routine skill selection and execution, coordination, and deep reasoning. This modular organization allows sensor control, adaptive sensing, edge-cloud execution, and foundation model reasoning to co-evolve within one closed-loop architecture, while keeping time-critical sensing and control on device and invoking higher-level inference only when needed. We instantiate ATI in a mobile camera prototype under dynamic lighting and motion. In our routed evaluation (L3-L4 split inference), compared to the default auto-exposure setting, ATI (L1/L2 adaptive sensing) improves end-to-end accuracy from 53.8% to 88% while reducing remote L4 invocations by 43.3%. These results show the value of co-designing sensing and inference for embodied AI.

Text-to-image (T2I) systems enable rapid generation of high-fidelity imagery but are misaligned with how visual ideas develop. T2I systems generate outputs that make implicit visual decisions on behalf of the user, often introduce fine-grained details that can anchor users prematurely and limit their ability to keep options open early on, and cause unintended changes during editing that are difficult to correct and reduce users' sense of control. To address these concerns, we present Creo, a multi-stage T2I system that scaffolds image generation by progressing from rough sketches to high-resolution outputs, exposing intermediary abstractions where users can make incremental changes. Sketch-like abstractions invite user editing and allow users to keep design options open when ideas are still forming due to their provisional nature. Each stage in Creo can be modified with manual changes and AI-assisted operations, enabling fine-grained, step-wise control through a locking mechanism that preserves prior decisions so subsequent edits affect only specified regions or attributes. Users remain in the loop, making and verifying decisions across stages, while the system applies diffs instead of regenerating full images, reducing drift as fidelity increases. A comparative study with a one-shot baseline shows that participants felt stronger ownership over Creo outputs, as they were able to trace their decisions in building up the image. Furthermore, embedding-based analysis indicates that Creo outputs are less homogeneous than one-shot results. These findings suggest that multi-stage generation, combined with intermediate control and decision locking, is a key design principle for improving controllability, user agency, creativity, and output diversity in generative systems.

According to constructivist theory, students learn software security more effectively when examples are grounded in their own code. Generic examples often fail to connect with students' prior work, limiting engagement and understanding. Advances in LLMs are now making it possible to automatically generate personalized examples by embedding security vulnerabilities directly into student-authored code. This paper introduces a method that uses LLMs to inject instances of specific Common Weakness Enumerations (CWEs) into students' own assignment code, creating individualized instructional materials. We present an agentic AI framework, using autonomous LLM-based agents equipped with task-specific tools to orchestrate injection, evaluation, ranking, and learning outcome generation. We report the experience of deploying this system in two undergraduate computer science courses (N=71), where students reviewed code samples containing LLM-injected vulnerabilities and completed a post-project survey. We compared responses with a baseline using a widely adopted set of generic security instructional materials. Students qualitatively reported finding CWE injections into their own code more relevant, clearer, and more engaging than the textbook-style examples. However, our quantitative findings revealed limited statistically significant differences, suggesting that while students valued the personalization, further studies and refinement of the approach are needed to establish stronger empirical support.

Existing agent-safety evaluation has focused mainly on externally induced risks. Yet agents may still enter unsafe trajectories under benign conditions. We study this complementary but underexplored setting through the lens of \emph{intrinsic} risk, where intrinsic failures remain latent, propagate across long-horizon execution, and eventually lead to high-consequence outcomes. To evaluate this setting, we introduce \emph{non-attack intrinsic risk auditing} and present \textbf{HINTBench}, a benchmark of 629 agent trajectories (523 risky, 106 safe; 33 steps on average) supporting three tasks: risk detection, risk-step localization, and intrinsic failure-type identification. Its annotations are organized under a unified five-constraint taxonomy. Experiments reveal a substantial capability gap: strong LLMs perform well on trajectory-level risk detection, but their performance drops to below 35 Strict-F1 on risk-step localization, while fine-grained failure diagnosis proves even harder. Existing guard models transfer poorly to this setting. These findings establish intrinsic risk auditing as an open challenge for agent safety.

This study evaluates colorectal risk factors and compares classical models against Quantum Neural Networks (QNNs) for anastomotic leak prediction. Analyzing clinical data with 14\% leak prevalence, we tested ZZFeatureMap encodings with RealAmplitudes and EfficientSU2 ansatze under simulated noise. $F_β$-optimized quantum configurations yielded significantly higher sensitivity (83.3\%) than classical baselines (66.7\%). This demonstrates that quantum feature spaces better prioritize minority class identification, which is critical for low-prevalence clinical risk prediction. Our work explores various optimizers under noisy conditions, highlighting key trade-offs and future directions for hardware deployment.

We show how causal interventions in Transformer models provide insights into English syntax by focusing on a long-standing challenge for syntactic theory: syntactic islands. Extraction from coordinated verb phrases is often degraded, yet acceptability varies gradiently with lexical content (e.g., "I know what he hates art and loves" vs. "I know what he looked down and saw"). We show that modern Transformer language models replicate human judgments across this gradient. Using causal interventions that isolate functionally relevant subspaces in Transformer blocks, attention modules, and MLPs, we demonstrate that extraction from coordination islands engages the same filler-gap mechanisms as canonical wh-dependencies, but that these mechanisms are selectively blocked to varying degrees. By projecting a large corpus of unrelated text onto these causally identified subspaces, we derive a novel linguistic hypothesis: the conjunction "and" is represented differently in extractable versus non-extractable constructions, corresponding to expressions encoding relational dependencies versus purely conjunctive uses. These results illustrate how mechanistic interpretability can inform syntax, generating new hypotheses about linguistic representation and processing.

Automated code generation remains a persistent challenge in software engineering, as conventional multi-agent frameworks are often constrained by static planning, isolated execution, high computational overhead, and limited adaptability to complex tasks. This paper introduces CollabCoder, a novel Plan-Code Co-Evolution framework that improves code generation through dynamic multi-agent collaboration. The core idea is to design a collaborative decision-making process between the plan module and the code module to decide which module should be executed for the debugging process. Extensive experiments on widely used benchmarks demonstrate that CollabCoder consistently improves code quality and robustness across tasks. Importantly, CollabCoder achieves performance comparable to or exceeding current state-of-the-art methods while reducing computational overhead, with efficiency gains becoming more pronounced as benchmark difficulty increases. On the more challenging LiveCodeBench and xCodeEval benchmarks, our approach improves performance by 11-20% over strong baselines while reducing the number of API calls by an average of 4-10 per execution.

Scientific peer review faces mounting strain as submission volumes surge, making it increasingly difficult to sustain review quality, consistency, and timeliness. Recent advances in AI have led the community to consider its use in peer review, yet a key unresolved question is whether AI can generate technically sound reviews at real-world conference scale. Here we report the first large-scale field deployment of AI-assisted peer review: every main-track submission at AAAI-26 received one clearly identified AI review from a state-of-the-art system. The system combined frontier models, tool use, and safeguards in a multi-stage process to generate reviews for all 22,977 full-review papers in less than a day. A large-scale survey of AAAI-26 authors and program committee members showed that participants not only found AI reviews useful, but actually preferred them to human reviews on key dimensions such as technical accuracy and research suggestions. We also introduce a novel benchmark and find that our system substantially outperforms a simple LLM-generated review baseline at detecting a variety of scientific weaknesses. Together, these results show that state-of-the-art AI methods can already make meaningful contributions to scientific peer review at conference scale, opening a path toward the next generation of synergistic human-AI teaming for evaluating research.

Software engineering research has focused on automating maintenance and evolution processes to reduce costs and improve reliability. The emergence of foundation models (FMs) with strong code understanding and reasoning abilities offers new opportunities for autonomous software behavior. Inspired by Artificial Life (ALife), we propose a fundamental shift in the Software Development Life-Cycle (SDLC) by introducing self-construction mechanisms that enable software to evolve and maintain autonomously. This position paper explores the potential of Autopoietic Architectures, specifically Psi-Arch, as a foundational framework for self-constructing software. We first analyze the limitations of traditional maintenance approaches and identify gaps in current SDLC automation. Subsequently, we outline the core challenges in achieving self-construction, including the integration of foundation-model-based reasoning units and the establishment of novel architectural paradigms. Although this paper does not present a definitive solution, it seeks to catalyze discourse and inspire research toward a new paradigm in software engineering, one in which self-constructing software represents the next frontier in SDLC automation.

Industrial time-series data from real production environments exhibits substantially higher complexity than commonly used benchmark datasets, primarily due to heterogeneous, multi-stage operational processes. As a result, anomaly detection methods validated under simplified conditions often fail to generalize to industrial settings. This work presents an empirical study on a unique dataset collected from fully operational industrial machinery, explicitly capturing pronounced process-induced variability. We evaluate which model classes are capable of capturing this complexity, starting with a classical Isolation Forest baseline and extending to multiple autoencoder architectures. Experimental results show that Isolation Forest is insufficient for modeling the non-periodic, multi-scale dynamics present in the data, whereas autoencoders consistently perform better. Among them, temporal convolutional autoencoders achieve the most robust performance, while recurrent and variational variants require more careful tuning.

Time-series anomaly detection (TSAD) is critical in domains such as industrial monitoring, healthcare, and cybersecurity, but it remains challenging due to rare and heterogeneous anomalies and the scarcity of labelled data. This scarcity makes unsupervised approaches predominant, yet existing methods often rely on reconstruction or forecasting, which struggle with complex data, or on embedding-based approaches that require domain-specific anomaly synthesis and fixed distance metrics. We propose ASTER, a framework that generates pseudo-anomalies directly in the latent space, avoiding handcrafted anomaly injections and the need for domain expertise. A latent-space decoder produces tailored pseudo-anomalies to train a Transformer-based anomaly classifier, while a pre-trained LLM enriches the temporal and contextual representations of this space. Experiments on three benchmark datasets show that ASTER achieves state-of-the-art performance and sets a new standard for LLM-based TSAD.

Computational fluid dynamics (CFD) has become an essential tool for predicting fire behavior, yet maintaining both efficiency and accuracy remains challenging. A major source of computational cost in fire simulations is the modeling of radiation transfer, which is usually the dominant heat transfer mechanism in fires. Solving the high-dimensional radiative transfer equation (RTE) with traditional numerical methods can be a performance bottleneck. Here, we present a machine learning framework based on Fourier-enhanced multiple-input neural operators (Fourier-MIONet) as an efficient alternative to direct numerical integration of the RTE. We first investigate the performance of neural operator architectures for a small-scale 2D pool fire and find that Fourier-MIONet provides the most accurate radiative solution predictions. The approach is then extended to 3D CFD fire simulations, where the computational mesh is locally refined across multiple levels. In these high-resolution settings, monolithic surrogate models for direct field-to-field mapping become difficult to train and computationally inefficient. To address this issue, a nested Fourier-MIONet is proposed to predict radiation solutions across multiple mesh-refinement levels. We validate the approach on 3D McCaffrey pool fires simulated with FireFOAM, including fixed fire sizes and a unified model trained over a continuous range of heat release rates (HRRs). The proposed method achieves global relative errors of 2-4% for 3D varying-HRR scenarios while providing faster inference than the estimated cost of one finite-volume radiation solve in FireFOAM for the 16-solid-angle case. With fast and accurate inference, the surrogate makes higher-fidelity radiation treatments practical and enables the incorporation of more spectrally resolved radiation models into CFD fire simulations for engineering applications.

Reinforcement Learning with Verifiable Rewards (RLVR) has catalyzed significant advances in the reasoning capabilities of Large Language Models (LLMs). However, effectively managing the exploration and exploitation trade-off remains a critical challenge. In this paper, we fully analyze the exploration and exploitation dilemma of extremely hard and easy samples during the training and propose a new fine-grained trade-off mechanism. Concretely, we introduce a perplexity space disentangling strategy that divides the sample space into distinct exploration (high perplexity) and exploitation (low perplexity) subspaces, thereby mining fine-grained samples requiring exploration-exploitation trade-off. Subsequently, we propose a bidirectional reward allocation mechanism with a minimum impact on verification rewards to implement perplexity-guided exploration and exploitation, enabling more stable policy optimization. Finally, we have evaluated our method on two mainstream tasks: mathematical reasoning and function calling, and experimental results demonstrate the superiority of the proposed method, confirming its effectiveness in enhancing LLM performance by fine-grained exploration-exploitation trade-off.

Instruction-tuned LLMs can annotate thousands of instances from a short prompt at negligible cost. This raises two questions for active learning (AL): can LLM labels replace human labels within the AL loop, and does AL remain necessary when entire corpora can be labelled at once? We investigate both questions on a new dataset of 277,902 German political TikTok comments (25,974 LLM-labelled, 5,000 human-annotated), comparing seven annotation strategies across four encoders to detect anti-immigrant hostility. A classifier trained on 25,974 GPT-5.2 labels (\$43) achieves comparable F1-Macro to one trained on 3,800 human annotations (\$316). Active learning offers little advantage over random sampling in our pre-enriched pool and delivers lower F1 than full LLM annotation at the same cost. However, comparable aggregate F1 masks a systematic difference in error structure: LLM-trained classifiers over-predict the positive class relative to the human gold standard. This divergence concentrates in topically ambiguous discussions where the distinction between anti-immigrant hostility and policy critique is most subtle, suggesting that annotation strategy should be guided not by aggregate F1 alone but by the error profile acceptable for the target application.

We present an open Molecular Crystal (MC) database of Machine-Learned Interatomic Potentials (MLIP) called MolCryst-MLIPs. The first release comprises fine-tuned MACE models for nine molecular crystal systems -- Benzamide, Benzoic acid, Coumarin, Durene, Isonicotinamide, Niacinamide, Nicotinamide, Pyrazinamide, and Resorcinol -- developed using the Automated Machine Learning Pipeline (AMLP), which streamlines the entire MLIP development workflow, from reference data generation to model training and validation, into a reproducible and user-friendly pipeline. Models are fine-tuned from the MACE-MH-1 foundation model (omol head), yielding a mean energy MAE of 0.141 kJ/mol/atom and a mean force MAE of 0.648 kJ/mol/Angstrom across all systems. Dynamical stability and structural integrity, as assessed through energy conservation, P2 orientational order parameters, and radial distribution functions, are evaluated using molecular dynamics simulations. The released models and datasets constitute a growing open database of validated MLIPs, ready for production MD simulations of molecular crystal polymorphism under different thermodynamic conditions.

Automated driving at unsignalized intersections is challenging due to complex multi-vehicle interactions and the need to balance safety and efficiency. Model Predictive Control (MPC) offers structured constraint handling through optimization but relies on hand-crafted rules that often produce overly conservative behavior. Deep Reinforcement Learning (RL) learns adaptive behaviors from experience but often struggles with safety assurance and generalization to unseen environments. In this study, we present an integrated MPC-RL framework to improve navigation performance in multi-agent scenarios. Experiments show that MPC-RL outperforms standalone MPC and end-to-end RL across three traffic-density levels. Collectively, MPC-RL reduces the collision rate by 21% and improves the success rate by 6.5% compared to pure MPC. We further evaluate zero-shot transfer to a highway merging scenario without retraining. Both MPC-based methods transfer substantially better than end-to-end PPO, which highlights the role of the MPC backbone in cross-scenario robustness. The framework also shows faster loss stabilization than end-to-end RL during training, which indicates a reduced learning burden. These results suggest that the integrated approach can improve the balance between safety performance and efficiency in multi-agent intersection scenarios, while the MPC component provides a strong foundation for generalization across driving environments. The implementation code is available open-source.

Why do capitalist economies recurrently generate crises whose severity is disproportionate to the size of the triggering shock? This paper proposes a structural answer grounded in the evolutionary geometry of production networks. As economies evolve through specialization, integration, and competitive selection, their inter-sectoral linkages drift toward configurations of increasing geometric fragility, eventually crossing a threshold beyond which small disturbances generate disproportionately large cascades. We introduce Sandpile Economics, a formal framework that interprets macroeconomic instability as an emergent property of disequilibrium production networks. The key state variable is the Forman--Ricci curvature of the input--output graph, capturing local substitution possibilities when supply chains are disrupted. We show that when curvature falls below an endogenous threshold, the distribution of cascade sizes follows a power law with tail index $α\in (1,2)$, implying a regime of unbounded amplification. The underlying mechanism is evolutionary: specialization reduces input substitutability, pushing the economy toward criticality, while crisis episodes induce endogenous network reconfiguration and path dependence. These dynamics are inherently non-ergodic and cannot be captured by representative-agent frameworks. Empirically, using global input--output data, we document that production networks operate in persistently negative curvature regimes and that curvature robustly predicts medium-run output dynamics. A one-standard-deviation increase in curvature is associated with higher cumulative growth over three-year horizons, and curvature systematically outperforms standard network metrics in explaining cross-country differences in resilience.

The integration of Large Language Models (LLMs) into Geographic Information Systems (GIS) marks a paradigm shift toward autonomous spatial analysis. However, evaluating these LLM-based agents remains challenging due to the complex, multi-step nature of geospatial workflows. Existing benchmarks primarily rely on static text or code matching, neglecting dynamic runtime feedback and the multimodal nature of spatial outputs. To address this gap, we introduce GeoAgentBench (GABench), a dynamic and interactive evaluation benchmark tailored for tool-augmented GIS agents. GABench provides a realistic execution sandbox integrating 117 atomic GIS tools, encompassing 53 typical spatial analysis tasks across 6 core GIS domains. Recognizing that precise parameter configuration is the primary determinant of execution success in dynamic GIS environments, we designed the Parameter Execution Accuracy (PEA) metric, which utilizes a "Last-Attempt Alignment" strategy to quantify the fidelity of implicit parameter inference. Complementing this, a Vision-Language Model (VLM) based verification is proposed to assess data-spatial accuracy and cartographic style adherence. Furthermore, to address the frequent task failures caused by parameter misalignments and runtime anomalies, we developed a novel agent architecture, Plan-and-React, that mimics expert cognitive workflows by decoupling global orchestration from step-wise reactive execution. Extensive experiments with seven representative LLMs demonstrate that the Plan-and-React paradigm significantly outperforms traditional frameworks, achieving the optimal balance between logical rigor and execution robustness, particularly in multi-step reasoning and error recovery. Our findings highlight current capability boundaries and establish a robust standard for assessing and advancing the next generation of autonomous GeoAI.

Modern machine learning models typically represent inputs as fixed points in a high-dimensional embedding space. While this approach has been proven powerful for a wide range of downstream tasks, it fundamentally differs from the way humans process information. Because humans are constantly adapting to their environment, they represent objects and their relationships in a highly context-sensitive manner. To address this gap, we propose a method for context-sensitive similarity computation from neural network embeddings, applied to modeling a triplet odd-one-out task with an anchor image serving as simultaneous context. Modeling context enables us to achieve up to a 15% improvement in odd-one-out accuracy over a context-insensitive model. We find that this improvement is consistent across both original and "human-aligned" vision foundation models.

The evaluation of supervised machine learning models is a critical stage in the development of reliable predictive systems. Despite the widespread availability of machine learning libraries and automated workflows, model assessment is often reduced to the reporting of a small set of aggregate metrics, which can lead to misleading conclusions about real-world performance. This paper examines the principles, challenges, and practical considerations involved in evaluating supervised learning algorithms across classification and regression tasks. In particular, it discusses how evaluation outcomes are influenced by dataset characteristics, validation design, class imbalance, asymmetric error costs, and the choice of performance metrics. Through a series of controlled experimental scenarios using diverse benchmark datasets, the study highlights common pitfalls such as the accuracy paradox, data leakage, inappropriate metric selection, and overreliance on scalar summary measures. The paper also compares alternative validation strategies and emphasizes the importance of aligning model evaluation with the intended operational objective of the task. By presenting evaluation as a decision-oriented and context-dependent process, this work provides a structured foundation for selecting metrics and validation protocols that support statistically sound, robust, and trustworthy supervised machine learning systems.

Driver drowsiness significantly impairs the ability to accurately judge safe braking distances and is estimated to contribute to 10%-20% of road accidents in Europe. Traditional driver-assistance systems lack adaptability to real-time physiological states such as drowsiness. This paper proposes a deep reinforcement learning-based autonomous braking system that integrates vehicle dynamics with driver physiological data. Drowsiness is detected from ECG signals using a Recurrent Neural Network (RNN), selected through an extensive benchmark analysis of 2-minute windows with varying segmentation and overlap configurations. The inferred drowsiness state is incorporated into the observable state space of a Double-Dueling Deep Q-Network (DQN) agent, where driver impairment is modeled as an action delay. The system is implemented and evaluated in a high-fidelity CARLA simulation environment. Experimental results show that the proposed agent achieves a 99.99% success rate in avoiding collisions under both drowsy and non-drowsy conditions. These findings demonstrate the effectiveness of physiology-aware control strategies for enhancing adaptive and intelligent driving safety systems.

Deep neural networks (DNNs) deliver state-of-the-art accuracy on regression and classification tasks, yet two structural deficits persistently obstruct their deployment in safety-critical, resource-constrained settings: (i) opacity of the learned function, which precludes formal verification, and (ii) reliance on heterogeneous, library-bound activation functions that inflate latency and silicon area on edge hardware. The recently introduced Exp-Minus-Log (EML) Sheffer operator, eml(x, y) = exp(x) - ln(y), was shown by Odrzywolek (2026) to be sufficient - together with the constant 1 - to express every standard elementary function as a binary tree of identical nodes. We propose to embed EML primitives inside conventional DNN architectures, yielding a hybrid DNN-EML model in which the trunk learns distributed representations and the head is a depth-bounded, weight-sparse EML tree whose snapped weights collapse to closed-form symbolic sub-expressions. We derive the forward equations, prove computational-cost bounds, analyse inference and training acceleration relative to multilayer perceptrons (MLPs) and physics-informed neural networks (PINNs), and quantify the trade-offs for FPGA/analog deployment. We argue that the DNN-EML pairing closes a literature gap: prior neuro-symbolic and equation-learner approaches (EQL, KAN, AI-Feynman) work with heterogeneous primitive sets and do not exploit a single hardware-realisable Sheffer element. A balanced assessment shows that EML is unlikely to accelerate training, and on commodity CPU/GPU it is also unlikely to accelerate inference; however, on a custom EML cell (FPGA logic block or analog circuit) the asymptotic latency advantage can reach an order of magnitude with simultaneous gain in interpretability and formal-verification tractability.

We consider the well-studied setting of minimizing a convex Lipschitz function using either gradient descent (GD) or its stochastic variant (SGD), and examine the last iterate convergence. By now, it is known that standard stepsize choices lead to a last iterate convergence rate of $\log T/\sqrt{T}$ after $T$ steps. A breakthrough result of Jain et al. [2019] recovered the optimal $1/\sqrt{T}$ rate by constructing a non-standard stepsize sequence. However, this sequence requires choosing $T$ in advance, as opposed to common stepsize schedules which apply for any time horizon. Moreover, Jain et al. conjectured that without prior knowledge of $T$, no stepsize sequence can ensure the optimal error for SGD's last iterate, a claim which so far remained unproven. We prove this conjecture, and in fact show that even in the noiseless case of GD, it is impossible to avoid an excess poly-log factor in $T$ when considering an anytime last iterate guarantee. Our proof further suggests that such (slightly) suboptimal stopping times are unavoidably common.

This paper develops a unified Markov Decision Process (MDP) framework for optimising two recurring in-match decisions in T20 cricket namely batting order selection and bowling plan assignment, directly in terms of win and defend probability rather than expected runs. A three-phase player profile engine (Powerplay, Middle, Death) with James-Stein shrinkage is estimated from 1,161 IPL ball-by-ball records (2008-2025). Win/defend probabilities are evaluated by vectorised Monte Carlo simulation over N = 50,000 innings trajectories. Batting orders are searched by exhaustive enumeration. Bowling plans are computed by simulated annealing over the remaining quota with the constraint that the same bowler cannot bowl consecutive overs. Applied to two 2026 IPL matches, the optimal batting order improves Mumbai Indians' win probability by 4.1 percentage points (52.4% to 56.5%), and the optimal Gujarat Titans bowling plan improves defend probability by 5.2 percentage points (39.1% to 44.3%). In both cases the observed sub-optimality is consistent with phase-agnostic deployment in decisions that appear reasonable by aggregate metrics but are exposed as costly when phase-specific profiles are applied.

The rapid proliferation of Model Context Protocol (MCP)-based agentic systems has introduced a new category of security threats that existing frameworks are inadequately equipped to address. We present MCPThreatHive, an open-source platform that automates the end-to-end lifecycle of MCP threat intelligence: from continuous, multi-source data collection through AI-driven threat extraction and classification, to structured knowledge graph storage and interactive visualization. The platform operationalizes the MCP-38 threat taxonomy, a curated set of 38 MCP-specific threat patterns mapped to STRIDE, OWASP Top 10 for LLM Applications, and OWASP Top 10 for Agentic Applications. A composite risk scoring model provides quantitative prioritization. Through a comparative analysis of representative existing MCP security tools, we identify three critical coverage gaps that MCPThreatHive addresses: incomplete compositional attack modeling, absence of continuous threat intelligence, and lack of unified multi-framework classification.

While sparse attention mitigates the computational bottleneck of long-context LLM training, its distributed training process exhibits extreme heterogeneity in both \textit{1)} sequence length and \textit{2)} sparsity sensitivity, leading to a severe imbalance problem and sub-optimal model accuracy. Existing algorithms and training frameworks typically focus on single issue, failing to systematically co-optimize these two problems. Therefore, we propose SparseBalance, a novel algorithm-system co-design framework, which exploits the sparsity and sequence heterogeneity to optimize model accuracy and system efficiency jointly. First, we propose workload-aware dynamic sparsity tuning, which employs a bidirectional sparsity adjustment to eliminate stragglers and exploit inherent bubbles for free accuracy. Second, we propose a sparsity-aware batching strategy to achieve coarse-grained balance, which complements dynamic sparsity tuning. Experimental results demonstrate that SparseBalance achieves up to a 1.33$\times$ end-to-end speedup while still improving the long-context capability by 0.46\% on the LongBench benchmark.

Personalized Large Language Models (LLMs) facilitate more natural, human-like interactions in human-centric applications. However, existing personalization methods are constrained by limited controllability and high resource demands. Furthermore, their reliance on static personality modeling restricts adaptability across varying situations. To address these limitations, we first demonstrate the existence of situation-dependency and consistent situation-behavior patterns within LLM personalities through a multi-perspective analysis of persona neurons. Building on these insights, we propose IRIS, a training-free, neuron-based Identify-Retrieve-Steer framework for advanced situational personality steering. Our approach comprises situational persona neuron identification, situation-aware neuron retrieval, and similarity-weighted steering. We empirically validate our framework on PersonalityBench and our newly introduced SPBench, a comprehensive situational personality benchmark. Experimental results show that our method surpasses best-performing baselines, demonstrating IRIS's generalization and robustness to complex, unseen situations and different models architecture.

Reward models are central to aligning large language models, yet they often overfit to spurious cues such as response length and overly agreeable tone. Most prior work weakens these cues directly by penalizing or controlling specific artifacts, but it does not explicitly encourage the model to ground preferences in the prompt's intent. We learn a decoder that maps a candidate answer to the latent intent embedding of the input. The reconstruction error is used as a signal to regularize the reward model training. We provide theoretical evidence that this signal emphasizes prompt-dependent information while suppressing prompt-independent shortcuts. Across math, helpfulness, and safety benchmarks, the decoder selects shorter and less sycophantic candidates with 0.877 accuracy. Incorporating this signal into RM training in Gemma-2-2B-it and Gemma-2-9B-it increases RewardBench accuracy from 0.832 to 0.868. For Best-of-N selection, our framework increases length-controlled win rates while producing shorter outputs, and remains robust to lengthening and mild off-topic drift in controlled rewrite tests.

Integro-differential equations arise in a wide range of applications, including transport, kinetic theory, radiative transfer, and multiphysics modeling, where nonlocal integral operators couple the solution across phase space. Such nonlocality often introduces dense coupling blocks in deterministic discretizations, leading to increased computational cost and memory usage, while physics-informed neural networks may suffer from expensive nonconvex training and sensitivity to hyperparameter choices. In this work, we present randomized neural networks (RaNNs) as a mesh-free collocation framework for linear integro-differential equations. Because the RaNN approximation is intrinsically dense through globally supported random features, the nonlocal integral operator does not introduce an additional loss of sparsity, while the approximate solution can still be represented with relatively few trainable degrees of freedom. By randomly fixing the hidden-layer parameters and solving only for the linear output weights, the training procedure reduces to a convex least-squares problem in the output coefficients, enabling stable and efficient optimization. As a representative application, we apply the proposed framework to the steady neutron transport equation, a high-dimensional linear integro-differential model featuring scattering integrals and diverse boundary conditions. Extensive numerical experiments demonstrate that, in the reported test settings, the RaNN approach achieves competitive accuracy while incurring substantially lower training cost than the selected neural and deterministic baselines, highlighting RaNNs as a robust and efficient alternative for the numerical simulation of nonlocal linear operators.

User simulators are essential for the scalable training and evaluation of interactive AI systems. However, existing approaches often rely on shallow user profiling, struggle to maintain persona consistency over long interactions, and are largely limited to English or single-domain settings. We present MUSE, a multi-domain Chinese user simulation framework designed to generate human-like, controllable, and behaviorally consistent responses. First, we propose Iterative Profile Self-Evolution (IPSE), which gradually optimizes user profiles by comparing and reasoning discrepancies between simulated trajectories and real dialogue behaviors. We then apply Role-Reversal Supervised Fine-Tuning to improve local response realism and human-like expression. To enable fine-grained behavioral alignment, we further train a specialized rubric-based reward model and incorporate it into rubric-guided multi-turn reinforcement learning, which optimizes the simulator at the dialogue level and enhances long-horizon behavioral consistency. Experiments show that MUSE consistently outperforms strong baselines in both utterance-level and session-level evaluations, generating responses that are more realistic, coherent, and persona-consistent over extended interactions.

Sentiment analysis in software engineering focuses on understanding emotions expressed in software artifacts. Previous research highlighted the limitations of applying general off-the-shelf sentiment analysis tools within the software engineering domain and indicated the need for specialized tools tailored to various software engineering contexts. The development of such tools heavily relies on supervised machine learning techniques that necessitate annotated datasets. Acquiring such datasets is a substantial challenge, as it requires domain-specific expertise and significant effort. Objective: This study explores the potential of ZSL to address the scarcity of annotated datasets in sentiment analysis within software engineering Method:} We conducted an empirical experiment to evaluate the performance of various ZSL techniques, including embedding-based, NLI-based, TARS-based, and generative-based ZSL techniques. We assessed the performance of these techniques under different labels setups to examine the impact of label configurations. Additionally, we compared the results of the ZSL techniques with state-of-the-art fine-tuned transformer-based models. Finally, we performed an error analysis to identify the primary causes of misclassifications. Results: Our findings demonstrate that ZSL techniques, particularly those combining expert-curated labels with embedding-based or generative-based models, can achieve macro-F1 scores comparable to fine-tuned transformer-based models. The error analysis revealed that subjectivity in annotation and polar facts are the main contributors to ZSL misclassifications. Conclusion: This study demonstrates the potential of ZSL for sentiment analysis in software engineering. ZSL can provide a solution to the challenge of annotated dataset scarcity by reducing reliance on annotated dataset.

World models have been emerging as critical components for assessing the consequences of actions generated by interactive agents in online planning and offline evaluation. In text-based environments, world models are typically evaluated and trained with single-step metrics such as Exact Match, aiming to improve the similarity between predicted and real-world states, but such metrics have been shown to be insufficient for capturing actual agent behavior. To address this issue, we introduce a new behavior-aligned training paradigm aimed at improving the functional consistency between the world model and the real environment. This paradigm focuses on optimizing a tractable step-level metric named Behavior Consistency Reward (BehR), which measures how much the likelihood of a logged next action changes between the real state and the world-model-predicted state under a frozen Reference Agent. Experiments on WebShop and TextWorld show that BehR-based training improves long-term alignment in several settings, with the clearest gains in WebShop and less movement in near-ceiling regimes, while preserving or improving single-step prediction quality in three of four settings. World models trained with BehR also achieve lower false positives in offline surrogate evaluation and show modest but encouraging gains in inference-time lookahead planning.

MLLM-based GUI agents have demonstrated strong capabilities in complex user interface interaction tasks. However, long-horizon scenarios remain challenging, as these agents are burdened with tasks beyond their intrinsic capabilities, suffering from memory degradation, progress confusion, and math hallucination. To address these challenges, we present UI-Copilot, a collaborative framework where the GUI agent focuses on task execution while a lightweight copilot provides on-demand assistance for memory retrieval and numerical computation. We introduce memory decoupling to separate persistent observations from transient execution context, and train the policy agent to selectively invoke the copilot as Retriever or Calculator based on task demands. To enable effective tool invocation learning, we propose Tool-Integrated Policy Optimization (TIPO), which separately optimizes tool selection through single-turn prediction and task execution through on-policy multi-turn rollouts. Experimental results show that UI-Copilot-7B achieves state-of-the-art performance on challenging MemGUI-Bench, outperforming strong 7B-scale GUI agents such as GUI-Owl-7B and UI-TARS-1.5-7B. Moreover, UI-Copilot-7B delivers a 17.1% absolute improvement on AndroidWorld over the base Qwen model, highlighting UI-Copilot's strong generalization to real-world GUI tasks.

Large language models (LLMs) have shown remarkable capabilities in dialogue generation and reasoning, yet their effectiveness in eliciting user-known but concealed information in open-ended conversations remains limited. In many interactive AI applications, such as personal assistants, tutoring systems, and legal or clinical support, users often withhold sensitive or uncertain information due to privacy concerns, ambiguity, or social hesitation. This makes it challenging for LLMs to gather complete and contextually relevant inputs. In this work, we define the problem of information elicitation in open-ended dialogue settings and propose Reinforcement Prompt Selection (RPS), a lightweight reinforcement learning framework that formulates prompt selection as a sequential decision-making problem. To analyze this problem in a controlled setting, we design a synthetic experiment, where a reinforcement learning agent outperforms a random query baseline, illustrating the potential of policy-based approaches for adaptive information elicitation. Building on this insight, RPS learns a policy over a pool of prompts to adaptively elicit concealed or incompletely expressed information from users through dialogue. We also introduce IELegal, a new benchmark dataset constructed from real legal case documents, which simulates dialogue-based information elicitation tasks aimed at uncovering case-relevant facts. In this setting, RPS outperforms static prompt baselines, demonstrating the effectiveness of adaptive prompt selection for eliciting critical information in LLM-driven dialogue systems.

Determining the number of clusters is a central challenge in unsupervised learning, where ground-truth labels are unavailable. The Silhouette coefficient is a widely used internal validation metric for this task, yet its standard micro-averaged form tends to favor larger clusters under size imbalance. Macro-averaging mitigates this bias by weighting clusters equally, but may overemphasize noise from under-represented groups. We introduce Composite Silhouette, an internal criterion for cluster-count selection that aggregates evidence across repeated subsampled clusterings rather than relying on a single partition. For each subsample, micro- and macro-averaged Silhouette scores are combined through an adaptive convex weight determined by their normalized discrepancy and smoothed by a bounded nonlinearity; the final score is then obtained by averaging these subsample-level composites. We establish key properties of the criterion and derive finite-sample concentration guarantees for its subsampling estimate. Experiments on synthetic and real-world datasets show that Composite Silhouette effectively reconciles the strengths of micro- and macro-averaging, yielding more accurate recovery of the ground-truth number of clusters.

Recent advances in artificial intelligence (AI) have shown promise in automating key aspects of Agile project management, yet their impact on team cognition remains underexplored. In this work, we investigate cognitive offloading in Agile sprint planning by conducting a controlled, three-condition experiment comparing AI-only, human-only, and hybrid planning models on a live client deliverable at a mid-sized digital agency. Using quantitative metrics -- including estimation accuracy, rework rates, and scope change recovery time -- alongside qualitative indicators of planning robustness, we evaluate each model's effectiveness beyond raw efficiency. We find that while AI-only planning minimizes time and cost, it significantly degrades risk capture rates and increases rework due to unstated assumptions, whereas human-only planning excels at adaptability but incurs substantial overhead. Drawing on these findings, we propose a theoretical framework for hybrid AI-human sprint planning that assigns algorithmic tools to estimation and backlog formatting while mandating human deliberation for risk assessment and ambiguity resolution. Our results challenge the assumption that efficiency equates to effectiveness, offering actionable governance strategies for organizations seeking to augment rather than erode team cognition.

Merging is a core operation in version control systems such as Git, but traditional line-based algorithms often yield spurious conflicts, particularly in the presence of refactorings or parallel edits. While syntax- and semantics-aware merging approaches can reduce conflicts, they introduce drawbacks such as loss of formatting, dependence on language-specific parsers, and limited flexibility across heterogeneous artifacts. To address this gap, we present Summer, a novel textual token-based merge algorithm independent of document formats. Dividing text into tokens, our approach formulates token-level changes in one branch into string-rewriting rules and move rules, and applies these rules to the text of the other branch to construct a merge. Despite being independent on programming languages, our move rules model extracting and inlining functions. We evaluated Summer on ConflictBench, a large benchmark of real-world merge scenarios, comparing it with five pioneering merge tools across Java and non-Java files. Experimental results show that Summer achieved the highest 36% accuracy in reproducing merges verbatim identical to developers', and ranked second in semantic accuracy.

Quantum circuit optimization is a central task in Quantum Computing, as current Noisy Intermediate Scale Quantum devices suffer from error propagation that often scales with the number of operations. Among quantum operations, the CNOT gate is of fundamental importance, being the only 2-qubit gate in the universal Clifford+T set. The problem of CNOT gates minimization has been addressed by heuristic algorithms such as the well-known Patel-Markov-Hayes (PMH) for linear reversible synthesis (i.e., CNOT minimization with no topological constraints), and more recently by Reinforcement Learning (RL) based strategies in the more complex case of topology-aware synthesis, where each CNOT can act on a subset of all qubits pairs. In this work we introduce AlphaCNOT, a RL framework based on Monte Carlo Tree Search (MCTS) that address effectively the CNOT minimization problem by modeling it as a planning problem. In contrast to other RL- based solution, our method is model-based, i.e. it can leverage lookahead search to evaluate future trajectories, thus finding more efficient sequences of CNOTs. Our method achieves a reduction of up to 32% in CNOT gate count compared to PMH baseline on linear reversible synthesis, while in the constraint version we report a consistent gate count reduction on a variety of topologies with up to 8 qubits, with respect to state-of-the-art RL-based solutions. Our results suggest the combination of RL with search-based strategies can be applied to different circuit optimization tasks, such as Clifford minimization, thus fostering the transition toward the "quantum utility" era.

Large Language Models (LLMs) are widely used across many domains, but their scale makes deployment challenging. Post-Training Quantization (PTQ) reduces memory footprint without retraining by leveraging a small calibration set. Recent Hessian-based PTQ methods compensate quantization error via cross-channel dependencies, but such approaches degrade at low bit-widths due to noisy curvature estimates from limited calibration data. We propose DASH-Q, a robust PTQ framework using diagonal Hessian approximation and iterative weighted least squares. By discarding noise-prone dependencies, DASH-Q filters sampling noise while prioritizing the preservation of salient feature power. We outperform other PTQ baselines in ultra low-bit regime, improving zero-shot accuracy by 7.01% on average and up to 14.01% over the strongest baselines across five baseline LLM models, while showing robust and stable performance with very small calibration data.

The rapid evolution of multimodal large models has revolutionized the simulation of diverse characters in speech dialogue systems, enabling a novel interactive paradigm. Character attributes are manifested not only in textual responses but also through vocal features, as speech conveys rich paralinguistic information that is challenging to quantify. This poses significant difficulties in evaluating the character alignment of role-playing agents. To address these challenges, we present RoleJudge, an evaluation framework that leverages audio large language models to systematically assess the alignment between speech and character across multiple modalities and dimensions. Furthermore, we introduce RoleChat, the first voice role-playing evaluation dataset enriched with chain-of-thought reasoning annotations, comprising a diverse set of authentic and LLM-generated speech samples. Utilizing this dataset, we implement a multi-stage training paradigm and incorporate Standard Alignment in reinforcement learning to mitigate reward misalignment during optimization. Experimental results in terms of accuracy and subjective assessment demonstrate that RoleJudge outperforms various baseline models, validating the effectiveness of our multidimensional evaluation framework.

Vision-language models are increasingly deployed in high-stakes settings, yet their susceptibility to sycophantic manipulation remains poorly understood, particularly in relation to how these models represent visual information internally. Whether models whose visual representations more closely mirror human neural processing are also more resistant to adversarial pressure is an open question with implications for both neuroscience and AI safety. We investigate this question by evaluating 12 open-weight vision-language models spanning 6 architecture families and a 40$\times$ parameter range (256M--10B) along two axes: brain alignment, measured by predicting fMRI responses from the Natural Scenes Dataset across 8 human subjects and 6 visual cortex regions of interest, and sycophancy, measured through 76,800 two-turn gaslighting prompts spanning 5 categories and 10 difficulty levels. Region-of-interest analysis reveals that alignment specifically in early visual cortex (V1--V3) is a reliable negative predictor of sycophancy ($r = -0.441$, BCa 95\% CI $[-0.740, -0.031]$), with all 12 leave-one-out correlations negative and the strongest effect for existence denial attacks ($r = -0.597$, $p = 0.040$). This anatomically specific relationship is absent in higher-order category-selective regions, suggesting that faithful low-level visual encoding provides a measurable anchor against adversarial linguistic override in vision-language models. We release our code on \href{https://github.com/aryashah2k/Gaslight-Gatekeep-Sycophantic-Manipulation}{GitHub} and dataset on \href{https://huggingface.co/datasets/aryashah00/Gaslight-Gatekeep-V1-V3}{Hugging Face}

In daily fantasy sports (DFS), match participation is highly time-sensitive. Users must act within a narrow window before a game begins, making match recommendation a time-critical task to prevent missed engagement and revenue loss. Existing recommender systems, typically designed for static item catalogs, are ill-equipped to handle the hard temporal deadlines inherent in these live events. To address this, we designed and deployed a recommendation engine using the Deep Interest Network (DIN) architecture. We adapt the DIN architecture by injecting temporality at two levels: first, through real-time urgency features for each candidate match (e.g., time-to-round-lock), and second, via temporal positional encodings that represent the time-gap between each historical interaction and the current recommendation request, allowing the model to dynamically weigh the recency of past actions. This approach, combined with a listwise neuralNDCG loss function, produces highly relevant and urgency-aware rankings. To support this at industrial scale, we developed a multi-node, multi-GPU training architecture on Ray and PyTorch. Our system, validated on a massive industrial dataset with over 650k users and over 100B interactions, achieves a +9% lift in nDCG@1 over a heavily optimized LightGBM baseline with handcrafted features. The strong offline performance of this model establishes its viability as a core component for our planned on-device (edge) recommendation system, where on-line A/B testing will be conducted.

Vision transformers (ViT) have been shown to allow for more flexible feature detection and can outperform convolutional neural network (CNN) when pre-trained on sufficient data. Due to their promising feature detection capabilities, we deployed ViTs for morphological classification of anaplastic large cell lymphoma (ALCL) versus classic Hodgkin lymphoma (cHL). We had previously designed a ViT model which was trained on a small dataset of 1,200 image patches in fully supervised training. That model achieved a diagnostic accuracy of 100% and an F1 score of 1.0 on the independent test set. Since fully supervised training is not a practical method due to lack of expertise resources in both the training and testing phases, we conducted a recent study on a modified approach to training data (weakly supervised training) and show that labeling training image patch automatically at the slide level of each whole-slide-image is a more practical solution for clinical use of Vision Transformer. Our ViT model, trained on a larger dataset of 100,000 image patches, yields evaluation metrics with significant accuracy, F1 score, and area under the curve (AUC) at 91.85%, 0.92, and 0.98, respectively. These are respectable values that qualify this ViT model, with weakly supervised training, as a suitable tool for a deep learning module in clinical model development using automated image patch extraction.

Large Language Models (LLMs) enhance their problem-solving capability by utilizing external tools. However, in open-world scenarios with massive and evolving tool repositories, existing methods relying on static embedding retrieval or parameter memorization of tools struggle to align user intent with tool semantics or generalize to unseen tools, respectively, leading to suboptimal accuracy of open-world tool retrieval and execution. To address these, we present ToolOmni, a unified agentic framework that enables LLMs for open-world tool use by proactive retrieval and grounded execution within a reasoning loop. First, we construct a cold-start multi-turn interaction dataset to instill foundational agentic capabilities via Supervised Fine-Tuning (SFT). Then, we introduce open-world tool learning based on a Decoupled Multi-Objective GRPO algorithm, which simultaneously optimizes LLMs for both tool retrieval accuracy and execution efficacy in online environments. Extensive experiments demonstrate that ToolOmni achieves state-of-the-art performance both in retrieval and execution, surpassing strong baselines by a significant margin of +10.8% in end-to-end execution success rate, while exhibiting exceptional robustness and generalization capabilities.

As large language models become standard backends for content generation, practical provenance increasingly requires multi-bit watermarking. In provider-internal deployments, a key requirement is message symmetry: the message itself should not systematically affect either text quality or verification outcomes. Vocabulary-partition watermarks can break message symmetry in low-entropy decoding: some messages are assigned most of the probability mass, while others are forced to use tail tokens. This makes embedding quality and message decoding accuracy message-dependent. We propose QuantileMark, a white-box multi-bit watermark that embeds messages within the continuous cumulative probability interval $[0, 1)$. At each step, QuantileMark partitions this interval into $M$ equal-mass bins and samples strictly from the bin assigned to the target symbol, ensuring a fixed $1/M$ probability budget regardless of context entropy. For detection, the verifier reconstructs the same partition under teacher forcing, computes posteriors over latent bins, and aggregates evidence for verification. We prove message-unbiasedness, a property ensuring that the base distribution is recovered when averaging over messages. This provides a theoretical foundation for generation-side symmetry, while the equal-mass design additionally promotes uniform evidence strength across messages on the detection side. Empirical results on C4 continuation and LFQA show improved multi-bit recovery and detection robustness over strong baselines, with negligible impact on generation quality. Our code is available at GitHub (https://github.com/zzzjunlin/QuantileMark).

Deep Learning (DL) is becoming more and more widespread in clone detection, motivated by achieving near-perfect performance for this task. In particular in case of semantic code clones, which share only limited syntax but implement the same or similar functionality, Deep Learning appears to outperform conventional tools. In this paper, we want to investigate the generalizability of DL-based clone detectors for Java. We therefore replicate and evaluate the performance of five state-of-the-art DL-based clone detectors, including Transformers like CodeBERT and single-task models like FA-AST+GMN, in a zero-shot evaluation scenario, where we train/fine-tune and evaluate on different datasets and functionalities. Our experiments demonstrate that the models' generalizability to unseen code is limited. Further analysis reveals that the conventional clone detector NiCad even outperforms the DL-based clone detectors in such a zero-shot evaluation scenario.

Large language models (LLMs) may memorize sensitive or copyrighted content, raising significant privacy and legal concerns. While machine unlearning has emerged as a potential remedy, prevailing paradigms rely on user-provided forget sets, making unlearning requests difficult to audit and exposing systems to secondary leakage and malicious abuse. We propose MAGE, a Memory-grAph Guided Erasure framework for user-minimized, corpus-free unlearning. Given only a lightweight user anchor that identifies a target entity, MAGE probes the target LLM to recover target-related memorization, organizes it into a weighted local memory graph, and synthesizes scoped supervision for unlearning. MAGE is model-agnostic, can be plugged into standard unlearning methods, and requires no access to the original training corpus. Experiments on two benchmarks, TOFU and RWKU, demonstrate that MAGE's self-generated supervision achieves effective unlearning performance comparable to supervision generated with external reference, while preserving overall utility. These results support a practical and auditable unlearning workflow driven by minimal anchors rather than user-supplied forget corpora.

Watermarking is becoming the default mechanism for AI content authentication, with governance policies and frameworks referencing it as infrastructure for content provenance. Yet across text, image, and audio modalities, watermark signal strength, detectability, and robustness depend on statistical properties of the content itself, properties that vary systematically across languages, cultural visual traditions, and demographic groups. We examine how this content dependence creates modality-specific pathways to bias. Reviewing the major watermarking benchmarks across modalities, we find that, with one exception, none report performance across languages, cultural content types, or population groups. To address this, we propose three concrete evaluation dimensions for pluralistic watermark benchmarking: cross-lingual detection parity, culturally diverse content coverage, and demographic disaggregation of detection metrics. We connect these to the governance frameworks currently mandating watermarking deployment and show that watermarking is held to a lower fairness standard than the generative systems it is meant to govern. Our position is that evaluation must precede deployment, and that the same bias auditing requirements applied to AI models should extend to the verification layer.

To date, various paradigms of soft-Computing have been used to solve many modern problems. Among them, a self organizing combination of fuzzy systems and neural networks can make a powerful decision making system. Here, a Dynamic Growing Fuzzy Neural Controller (DGFNC) is combined with an adaptive strategy and applied to a 3PSP parallel robot position control problem. Specifically, the dynamic growing mechanism is considered in more detail. In contrast to other self-organizing methods, DGFNC adds new rules more conservatively; hence the pruning mechanism is omitted. Instead, the adaptive strategy 'adapts' the control system to parameter variation. Furthermore, a sliding mode-based nonlinear controller ensures system stability. The resulting general control strategy aims to achieve faster response with less computation while maintaining overall stability. Finally, the 3PSP is chosen due to its complex dynamics and the utility of such approaches in modern industrial systems. Several simulations support the merits of the proposed DGFNC strategy as applied to the 3PSP robot.

Sparse mixture-of-experts (MoE) layers have been shown to substantially increase model capacity without a proportional increase in computational cost and are widely used in transformer architectures, where they typically replace feed-forward network blocks. In contrast, integrating sparse MoE layers into convolutional neural networks (CNNs) remains inconsistent, with most prior work focusing on fine-grained MoEs operating at the filter or channel levels. In this work, we investigate a coarser, patch-wise formulation of sparse MoE layers for semantic segmentation, where local regions are routed to a small subset of convolutional experts. Through experiments on the Cityscapes and BDD100K datasets using encoder-decoder and backbone-based CNNs, we conduct a design analysis to assess how architectural choices affect routing dynamics and expert specialization. Our results demonstrate consistent, architecture-dependent improvements (up to +3.9 mIoU) with little computational overhead, while revealing strong design sensitivity. Our work provides empirical insights into the design and internal dynamics of sparse MoE layers in CNN-based dense prediction. Our code is available at https://github.com/KASTEL-MobilityLab/moe-layers/.

Large language model (LLM) agents on multi-step tasks suffer reasoning degradation, looping, drift, stuck states, at rates up to 30% on hard tasks. Current solutions include hard step limits (abrupt) or LLM-as-judge monitoring (10-15% overhead per step). This paper introduces the Cognitive Companion, a parallel monitoring architecture with two implementations: an LLM-based Companion and a novel zero-overhead Probe-based Companion. We report a three-batch feasibility study centered on Gemma 4 E4B, with an additional exploratory small-model analysis on Qwen 2.5 1.5B and Llama 3.2 1B. In our experiments, the LLM-based Companion reduced repetition on loop-prone tasks by 52-62% with approximately 11% overhead. The Probe-based Companion, trained on hidden states from layer 28, showed a mean effect size of +0.471 at zero measured inference overhead; its strongest probe result achieved cross-validated AUROC 0.840 on a small proxy-labeled dataset. A key empirical finding is that companion benefit appears task-type dependent: companions are most helpful on loop-prone and open-ended tasks, while effects are neutral or negative on more structured tasks. Our small-model experiments also suggest a possible scale boundary: companions did not improve the measured quality proxy on 1B-1.5B models, even when interventions fired. Overall, the paper should be read as a feasibility study rather than a definitive validation. The results provide encouraging evidence that sub-token monitoring may be useful, identify task-type sensitivity as a practical design constraint, and motivate selective companion activation as a promising direction for future work.

The next generation of autonomous AI systems will be constrained not only by model capability, but by how intelligence is structured across heterogeneous hardware. Current paradigms -- cloud-centric AI, on-device inference, and edge-cloud pipelines -- treat planning, reasoning, and execution as a monolithic process, leading to unnecessary latency, energy consumption, and fragmented behavioral continuity. We introduce the Tri-Spirit Architecture, a three-layer cognitive framework that decomposes intelligence into planning (Super Layer), reasoning (Agent Layer), and execution (Reflex Layer), each mapped to distinct compute substrates and coordinated via an asynchronous message bus. We formalize the system with a parameterized routing policy, a habit-compilation mechanism that promotes repeated reasoning paths into zero-inference execution policies, a convergent memory model, and explicit safety constraints. We evaluate the architecture in a reproducible simulation of 2000 synthetic tasks against cloud-centric and edge-only baselines. Tri-Spirit reduces mean task latency by 75.6 percent and energy consumption by 71.1 percent, while decreasing LLM invocations by 30 percent and enabling 77.6 percent offline task completion. These results suggest that cognitive decomposition, rather than model scaling alone, is a primary driver of system-level efficiency in AI hardware.

The potential of Multimodal Large Language Models (MLLMs) in domain of medical imaging raise the demands of systematic and rigorous evaluation frameworks that are aligned with the real-world medical imaging practice. Existing practices that report single or coarse-grained metrics are lack the granularity required for specialized clinical support and fail to assess the reliability of reasoning mechanisms. To address this, we propose a paradigm shift toward multidimensional, fine-grained and in-depth evaluation. Based on a two-stage systematic construction pipeline designed for this paradigm, we instantiate it with MedRCube. We benchmark 33 MLLMs, \textit{Lingshu-32B} achieve top-tier performance. Crucially, MedRCube exposes a series of pronounced insights inaccessible under prior evaluation settings. Furthermore, we introduce a credibility evaluation subset to quantify reasoning credibility, uncover a highly significant positive association between shortcut behavior and diagnostic task performance, raising concerns for clinically trustworthy deployment. The resources of this work can be found at https://github.com/F1mc/MedRCube.

We propose a new partial-observability model for online learning problems where the learner, besides its own loss, also observes some noisy feedback about the other actions, depending on the underlying structure of the problem. We represent this structure by a weighted directed graph, where the edge weights are related to the quality of the feedback shared by the connected nodes. Our main contribution is an efficient algorithm that guarantees a regret of $\widetilde{O}(\sqrt{α^* T})$ after $T$ rounds, where $α^*$ is a novel graph property that we call the effective independence number. Our algorithm is completely parameter-free and does not require knowledge (or even estimation) of $α^*$. For the special case of binary edge weights, our setting reduces to the partial-observability models of Mannor and Shamir (2011) and Alon et al. (2013) and our algorithm recovers the near-optimal regret bounds.

Thompson Sampling (TS) has attracted a lot of interest due to its good empirical performance, in particular in the computational advertising. Though successful, the tools for its performance analysis appeared only recently. In this paper, we describe and analyze SpectralTS algorithm for a bandit problem, where the payoffs of the choices are smooth given an underlying graph. In this setting, each choice is a node of a graph and the expected payoffs of the neighboring nodes are assumed to be similar. Although the setting has application both in recommender systems and advertising, the traditional algorithms would scale poorly with the number of choices. For that purpose we consider an effective dimension d, which is small in real-world graphs. We deliver the analysis showing that the regret of SpectralTS scales as d*sqrt(T ln N) with high probability, where T is the time horizon and N is the number of choices. Since a d*sqrt(T ln N) regret is comparable to the known results, SpectralTS offers a computationally more efficient alternative. We also show that our algorithm is competitive on both synthetic and real-world data.

We investigate stochastic combinatorial semi-bandits, where the entire joint distribution of outcomes impacts the complexity of the problem instance (unlike in the standard bandits). Typical distributions considered depend on specific parameter values, whose prior knowledge is required in theory but quite difficult to estimate in practice; an example is the commonly assumed sub-Gaussian family. We alleviate this issue by instead considering a new general family of sub-exponential distributions, which contains bounded and Gaussian ones. We prove a new lower bound on the expected regret on this family, that is parameterized by the unknown covariance matrix of outcomes, a tighter quantity than the sub-Gaussian matrix. We then construct an algorithm that uses covariance estimates, and provide a tight asymptotic analysis of the regret. Finally, we apply and extend our results to the family of sparse outcomes, which has applications in many recommender systems.

Recommender systems have historically developed along two largely independent paradigms: feature interaction models for modeling correlations among multi-field categorical features, and sequential models for capturing user behavior dynamics from historical interaction sequences. Although recent trends attempt to bridge these paradigms within shared backbones, we empirically reveal that naive unifying these two branches may lead to a failure mode of Sequential Collapse Propagation (SCP). That is, the interaction with those dimensionally ill non-sequence fields leads to the dimensional collapse of the sequence features. To overcome this challenge, we propose TokenFormer, a unified recommendation architecture with the following innovations. First, we introduce a Bottom-Full-Top-Sliding (BFTS) attention scheme, which applies full self-attention in the lower layers and shrinking-window sliding attention in the upper layers. Second, we introduce a Non-Linear Interaction Representation (NLIR) that applies one-sided non-linear multiplicative transformations to the hidden states. Extensive experiments on public benchmarks and Tencent's advertising platform demonstrate state-of-the-art performance, while detailed analysis confirm that TokenFormer significantly improves dimensional robustness and representation discriminability under unified modeling.

This work identifies a necessary condition for any variational quantum approach to reach the exact ground state. Briefly, the norms of the projections of the input and the ground state onto each group module must match, implying that module weights of the solution state have to be known in advance in order to reach the exact ground state. An exemplary case is provided by matchgate circuits applied to problems whose solutions are classical bit strings, since all computational basis states share the same module-wise weights. Combined with the known classical simulability of quantum circuits for which observables lie in a small linear subspace, this implies that certain problems admit a classical surrogate for exact solution with each step taking $O(n^5)$ time. The Maximum Cut problem serves as an illustrative example.

Reinforcement learning (RL) enables high-frequency, closed-loop control for robotic manipulation, but scaling to long-horizon tasks with sparse or imperfect rewards remains difficult due to inefficient exploration and poor credit assignment. Vision-Language-Action (VLA) models leverage large-scale multimodal pretraining to provide generalist, task-level reasoning, but current limitations hinder their direct use in fast and precise manipulation. In this paper, we propose Vision-Language-Action Jump-Starting (VLAJS), a method that bridges sparse VLA guidance with on-policy RL to improve exploration and learning efficiency. VLAJS treats VLAs as transient sources of high-level action suggestions that bias early exploration and improve credit assignment, while preserving the high-frequency, state-based control of RL. Our approach augments Proximal Policy Optimization (PPO) with a directional action-consistency regularization that softly aligns the RL agent's actions with VLA guidance during early training, without enforcing strict imitation, requiring demonstrations, or relying on continuous teacher queries. VLA guidance is applied sparsely and annealed over time, allowing the agent to adapt online and ultimately surpass the guiding policy. We evaluate VLAJS on six challenging manipulation tasks: lifting, pick-and-place, peg reorientation, peg insertion, poking, and pushing in simulation, and validate a subset on a real Franka Panda robot. VLAJS consistently outperforms PPO and distillation-style baselines in sample efficiency, reducing required environment interactions by over 50% in several tasks. Real-world experiments demonstrate zero-shot sim-to-real transfer and robust execution under clutter, object variation, and external perturbations.

Multi-page Document Visual Question Answering requires reasoning over semantics, layouts, and visual elements in long, visually dense documents. Existing OCR-free methods face a trade-off between capacity and precision: end-to-end models scale poorly with document length, while visual retrieval-based pipelines are brittle and passive. We propose Doc-$V^*$, an \textbf{OCR-free agentic} framework that casts multi-page DocVQA as sequential evidence aggregation. Doc-$V^*$ begins with a thumbnail overview, then actively navigates via semantic retrieval and targeted page fetching, and aggregates evidence in a structured working memory for grounded reasoning. Trained by imitation learning from expert trajectories and further optimized with Group Relative Policy Optimization, Doc-$V^*$ balances answer accuracy with evidence-seeking efficiency. Across five benchmarks, Doc-$V^*$ outperforms open-source baselines and approaches proprietary models, improving out-of-domain performance by up to \textbf{47.9\%} over RAG baseline. Other results reveal effective evidence aggregation with selective attention, not increased input pages.

We present a hybrid retrieval system for COVID-19 scientific literature, evaluated on the TREC-COVID benchmark (171,332 papers, 50 expert queries). The system implements six retrieval configurations spanning sparse (SPLADE), dense (BGE), rank-level fusion (RRF), and a projection-based vector fusion (B5) approach. RRF fusion achieves the best relevance (nDCG@10 = 0.828), outperforming dense-only by 6.1% and sparse-only by 14.9%. Our projection fusion variant reaches nDCG@10 = 0.678 on expert queries while being 33% faster (847 ms vs. 1271 ms) and producing 2.2x higher ILD@10 than RRF. Evaluation across 400 queries -- including expert, machine-generated, and three paraphrase styles -- shows that B5 delivers the largest relative gain on keyword-heavy reformulations (+8.8%), although RRF remains best in absolute nDCG@10. On expert queries, MMR reranking increases intra-list diversity by 23.8-24.5% at a 20.4-25.4% nDCG@10 cost. Both fusion pipelines evaluated for latency remain below the sub-2 s target across all query sets. The system is deployed as a Streamlit web application backed by Pinecone serverless indices.

Repository-level code intelligence tasks require large language models (LLMs) to process long, multi-file contexts. Such inputs introduce three challenges: crucial context can be obscured by noise, truncated due to limited windows, and increased inference latency. Context compression mitigates these risks by condensing inputs. While studied in NLP, its applicability to code tasks remains largely unexplored. We present the first systematic empirical study of context compression for repository-level code intelligence, organizing eight methods into three paradigms: discrete token sequences, continuous latent vectors, and visual tokens. We evaluate them on code completion and generation, measuring performance and efficiency. Results show context compression is effective: at 4x compression, continuous latent vector methods surpass full-context performance by up to 28.3% in BLEU score, indicating they filter noise rather than just truncating. On efficiency, all paradigms reduce inference cost. Both visual and text-based compression achieve up to 50% reduction in end-to-end latency at high ratios, approaching the cost of inference without repository context. These findings establish context compression as a viable approach and provide guidance for paradigm selection.

Physics-Informed Neural Networks (PINNs) recast PDE solving as an optimisation problem in function space by minimising a residual-based objective, yet many applications require additional derivative-based relations that are just as fundamental as the governing equations. In this paper, we present Derivative-Constrained PINNs (DC-PINNs), a general framework that treats constrained PDE solving as an optimisation guided by a minimum objective function criterion where the physics resides in the minimum principle. DC-PINNs embed general nonlinear constraints on states and derivatives, e.g., bounds, monotonicity, convexity, incompressibility, computed efficiently via automatic differentiation, and they employ self-adaptive loss balancing to tune the influence of each objective, reducing reliance on manual hyperparameters and problem-specific architectures. DC-PINNs consistently reduce constraint violations and improve physical fidelity versus baseline PINN variants, representative hard-constraint formulations on benchmarks, including heat diffusion with bounds, financial volatilities with arbitrage-free, and fluid flow with vortices shed. Explicitly encoding derivative constraints stabilises training and steers optimisation toward physically admissible minima even when the PDE residual alone is small, providing reliable solutions of constrained PDEs grounded in energy minimum principles.

The technical support team of a supercomputing centre accumulates, over the course of decades, a large volume of resolved incidents that constitute critical operational knowledge. At the Galician Supercomputing Center (CESGA) this history has been managed for over twenty years with Request Tracker (RT), whose built-in search engine has significant limitations that hinder knowledge reuse by the support staff. This paper presents Fragata, a semantic ticket search system that combines modern information retrieval techniques with the full RT history. The system can find relevant past incidents regardless of language, the presence of typos, or the specific wording of the query. The architecture is deployed on CESGA's infrastructure, supports incremental updates without service interruption, and offloads the most expensive stages to the FinisTerrae III supercomputer. Preliminary results show a substantial qualitative improvement over RT's native search.

LLM-as-a-judge, using a language model to score or rank candidate responses, is widely used as a scalable alternative to human evaluation in RLHF pipelines, benchmarking, and application layer evaluations (evals). However, judgment reliability depends heavily on prompting and aggregation strategy. We present an empirical investigation of practical, drop-in techniques that improve GPT-5.4 judge accuracy on RewardBench 2 without any finetuning. Two techniques account for nearly all available gains: task-specific criteria injection (+3.0pp at negligible cost) and ensemble scoring (+9.8pp at 5x cost). Combined, they reach 83.6% accuracy, +11.9pp over the 71.7% baseline. Our investigation also covers three further techniques (calibration context, adaptive model escalation, and soft blending) which did not reliably improve on criteria + ensembling at comparable cost. Cheaper model tiers benefit disproportionately from ensembling: GPT-5.4 mini with k=8 achieves 79.2% at 1.2x baseline cost, and GPT-5.4 nano with k=8 reaches 71.4% at 0.4x baseline cost, making high-accuracy LLM judges accessible at low cost.

Large Audio-Language Models (LALMs) enable general audio understanding and demonstrate remarkable performance across various audio tasks. However, these models still face challenges in temporal perception (e.g., inferring event onset and offset), leading to limited utility in fine-grained scenarios. To address this issue, we propose Audio-Side Time Prompt and leverage Reinforcement Learning (RL) to develop the TimePro-RL framework for fine-grained temporal perception. Specifically, we encode timestamps as embeddings and interleave them within the audio feature sequence as temporal coordinates to prompt the model. Furthermore, we introduce RL following Supervised Fine-Tuning (SFT) to directly optimize temporal alignment performance. Experiments demonstrate that TimePro-RL achieves significant performance gains across a range of audio temporal tasks, such as audio grounding, sound event detection, and dense audio captioning, validating its robust effectiveness.

Metaphor detection models achieve strong benchmark performance, yet it remains unclear whether this reflects transferable generalization or lexical memorization. To address this, we analyze generalization in metaphor detection through RoBERTa, the shared backbone of many state-of-the-art systems, focusing on English verbs using the VU Amsterdam Metaphor Corpus. We introduce a controlled lexical hold-out setup where all instances of selected target lemmas are strictly excluded from fine-tuning, and compare predictions on these Held-out lemmas against Exposed lemmas (verbs seen during fine-tuning). While the model performs best on Exposed lemmas, it maintains robust performance on Held-out lemmas. Further analysis reveals that sentence context alone is sufficient to match full-model performance on Held-out lemmas, whereas static verb-level embeddings are not. Together, these results suggest that generalization is primarily driven by "learning the cue" (transferable contextual patterns), while "learning the word" (verb-specific memorization) provides an additive boost when lexical exposure is available.

Professional fact-checkers rely on domain knowledge and deep contextual understanding to verify claims. Large language models (LLMs) and large reasoning models (LRMs) lack such grounding and primarily reason from available evidence alone, creating a mismatch between expert-led and fully automated claim verification. To mitigate this gap, we posit human-AI collaboration as a more promising path forward, where expert feedback, grounded in real-world knowledge and domain expertise, guides the model's reasoning. However, existing LRMs are hard to calibrate to natural language feedback, particularly in a multi-turn interaction setup. We propose Co-FactChecker, a framework for human-AI collaborative claim verification. We introduce a new interaction paradigm that treats the model's thinking trace as a shared scratchpad. Co-FactChecker translates expert feedback into trace-edits that introduce targeted modifications to the trace, sidestepping the shortcomings of dialogue-based interaction. We provide theoretical results showing that trace-editing offers advantages over multi-turn dialogue, and our automatic evaluations demonstrate that Co-FactChecker outperforms existing autonomous and human-AI collaboration approaches. Human evaluations further show that Co-FactChecker is preferred over multi-turn dialogue, producing higher quality reasoning and verdicts along with relatively easier to interpret and more useful thinking traces.

Fairness in language models is typically studied as a property of a single, centrally optimized model. As large language models become increasingly agentic, we propose that fairness emerges through interaction and exchange. We study this via a controlled hospital triage framework in which two agents negotiate over three structured debate rounds. One agent is aligned to a specific ethical framework via retrieval-augmented generation (RAG), while the other is either unaligned or adversarially prompted to favor demographic groups over clinical need. We find that alignment systematically shapes negotiation strategies and allocation patterns, and that neither agent's allocation is ethically adequate in isolation, yet their joint final allocation can satisfy fairness criteria that neither would have reached alone. Aligned agents partially moderate bias through contestation rather than override, acting as corrective patches that restore access for marginalized groups without fully converting a biased counterpart. We further observe that even explicitly aligned agents exhibit intrinsic biases toward certain frameworks, consistent with known left-leaning tendencies in LLMs. We connect these limits to Arrow's Impossibility Theorem: no aggregation mechanism can simultaneously satisfy all desiderata of collective rationality, and multi-agent deliberation navigates rather than resolves this constraint. Our results reposition fairness as an emergent, procedural property of decentralized agent interaction, and the system rather than the individual agent as the appropriate unit of evaluation.

Large language models (LLMs) have enabled agentic AI systems for scientific discovery, but most approaches remain limited to textbased reasoning without automated experimental verification. We propose MIND, an LLM-driven framework for automated hypothesis validation in materials research. MIND organizes the scientific discovery process into hypothesis refinement, experimentation, and debate-based validation within a multi-agent pipeline. For experimental verification, the system integrates Machine Learning Interatomic Potentials, particularly SevenNet-Omni, enabling scalable in-silico experiments. We also provide a web-based user interface for automated hypothesis testing. The modular design allows additional experimental modules to be integrated, making the framework adaptable to broader scientific workflows. The code is available at: https://github.com/IMMS-Ewha/MIND, and a demonstration video at: https://youtu.be/lqiFe1OQzN4.

Reliable and interpretable decision-making is essential in medical imaging, where diagnostic outcomes directly influence patient care. Despite advances in deep learning, most medical AI systems operate as opaque black boxes, providing little insight into why a particular diagnosis was reached. In this paper, we introduce Med-CAM, a framework for generating minimal and sharp maps as evidence-based explanations for Medical decision making via Classifier Activation Matching. Med-CAM trains a segmentation network from scratch to produce a mask that highlights the minimal evidence critical to model's decision for any seen or unseen image. This ensures that the explanation is both faithful to the network's behaviour and interpretable to clinicians. Experiments show, unlike prior spatial explanation methods, such as Grad-CAM and attention maps, which yield only fuzzy regions of relative importance, Med-CAM with its superior spatial awareness to shapes, textures, and boundaries, delivers conclusive, evidence-based explanations that faithfully replicate the model's prediction for any given image. By explicitly constraining explanations to be compact, consistent with model activations, and diagnostic alignment, Med-CAM advances transparent AI to foster clinician understanding and trust in high-stakes medical applications such as pathology and radiology.

Mechanistic interpretability seeks to localize model behavior to the internal components that causally realize it. Prior work has advanced activation-space localization and causal tracing, but modules that appear important in activation space may merely aggregate or amplify upstream signals rather than encode the target capability in their own parameters. To address this gap, we propose Weight Patching, a parameter-space intervention method for source-oriented analysis in paired same-architecture models that differ in how strongly they express a target capability under the inputs of interest. Given a base model and a behavior-specialized counterpart, Weight Patching replaces selected module weights from the specialized model into the base model under a fixed input. We instantiate the method on instruction following and introduce a framework centered on a vector-anchor behavioral interface that provides a shared internal criterion for whether a task-relevant control state has been formed or recovered in open-ended generation. Under this framework, the analysis reveals a hierarchy from shallow candidate source-side carriers to aggregation and routing modules, and further to downstream execution circuits. The recovered component scores can also guide mechanism-aware model merging, improving selective fusion across the evaluated expert combinations and providing additional external validation.

Performance debugging in WebAssembly (Wasm) runtimes is essential for ensuring the robustness of Wasm, especially since performance issues have frequently occurred in Wasm runtimes, which can significantly degrade the capabilities of hosted services. Many performance issues in Wasm runtimes result from suboptimal compilation of input Wasm programs, for which existing performance debugging methods primarily designed for application-level inefficiencies are not well-suited. In this paper, we present WarpL, a novel mutation-based approach that aims to identify the exact suboptimal instruction sequences responsible for the performance issues in Wasm runtimes, thereby narrowing down the root causes. Specifically, WarpL obtains a functionally similar mutant in which the performance issue does not manifest, and isolates the exact suboptimal instructions by comparing the machine code of the original and mutated programs. We implement WarpL as an open-source tool and evaluate it on 12 real-world performance issues across three widely used Wasm runtimes. WarpL identified the exact causes in 10 out of 12 issues. Notably, we have used WarpL to successfully diagnose six previously unknown performance issues in Wasmtime.

As large language models (LLMs) generate text that increasingly resembles human writing, the subtle cues that distinguish AI-generated content from human-written content become increasingly challenging to capture. Reliance on generator-specific artifacts is inherently unstable, since new models emerge rapidly and reduce the robustness of such shortcuts. This generalizes unseen generators as a central and challenging problem for AI-text detection. To tackle this challenge, we propose a progressively structured framework that disentangles AI-detection semantics from generator-aware artifacts. This is achieved through a compact latent encoding that encourages semantic minimality, followed by perturbation-based regularization to reduce residual entanglement, and finally a discriminative adaptation stage that aligns representations with task objectives. Experiments on MAGE benchmark, covering 20 representative LLMs across 7 categories, demonstrate consistent improvements over state-of-the-art methods, achieving up to 24.2% accuracy gain and 26.2% F1 improvement. Notably, performance continues to improve as the diversity of training generators increases, confirming strong scalability and generalization in open-set scenarios. Our source code will be publicly available at https://github.com/PuXiao06/DRGD.

3D editing refers to the ability to apply local or global modifications to 3D assets. Effective 3D editing requires maintaining semantic consistency by performing localized changes according to prompts, while also preserving local invariance so that unchanged regions remain consistent with the original. However, existing approaches have significant limitations: multi-view editing methods incur losses when projecting back to 3D, while voxel-based editing is constrained in both the regions that can be modified and the scale of modifications. Moreover, the lack of sufficiently large editing datasets for training and evaluation remains a challenge. To address these challenges, we propose a Beyond Voxel 3D Editing (BVE) framework with a self-constructed large-scale dataset specifically tailored for 3D editing. Building upon this dataset, our model enhances a foundational image-to-3D generative architecture with lightweight, trainable modules, enabling efficient injection of textual semantics without the need for expensive full-model retraining. Furthermore, we introduce an annotation-free 3D masking strategy to preserve local invariance, maintaining the integrity of unchanged regions during editing. Extensive experiments demonstrate that BVE achieves superior performance in generating high-quality, text-aligned 3D assets, while faithfully retaining the visual characteristics of the original input.

Glitches frequently contaminate data in gravitational-wave detectors, complicating the observation and analysis of astrophysical signals. This work introduces VIGILant, an automatic pipeline for classification and visualization of glitches in the Virgo detector. Using a curated dataset of Virgo O3b glitches, two machine learning approaches are evaluated: tree-based models (Decision Tree, Random Forest and XGBoost) using structured Omicron parameters, and Convolutional Neural Networks (ResNet) trained on spectrogram images. While tree-based models offer higher interpretability and fast training, the ResNet34 model achieved superior performance, reaching a F1 score of 0.9772 and accuracy of 0.9833 in the testing set, with inference times of tens of milliseconds per glitch. The pipeline has been deployed for daily operation at the Virgo site since observing run O4c, providing the Virgo collaboration with an interactive dashboard to monitor glitch populations and detector behavior. This allows to identify low-confidence predictions, highlighting glitches requiring further attention.

While Large Language Models (LLMs) have significantly advanced Text-to-SQL performance, existing benchmarks predominantly focus on Western contexts and simplified schemas, leaving a gap in real-world, non-Western applications. We present IndicDB, a multilingual Text-to-SQL benchmark for evaluating cross-lingual semantic parsing across diverse Indic languages. The relational schemas are sourced from open-data platforms, including the National Data and Analytics Platform (NDAP) and the India Data Portal (IDP), ensuring realistic administrative data complexity. IndicDB comprises 20 databases across 237 tables. To convert denormalized government data into rich relational structures, we employ an iterative three-agent framework (Architect, Auditor, Refiner) to ensure structural rigor and high relational density (11.85 tables per database; join depths up to six). Our pipeline is value-aware, difficulty-calibrated, and join-enforced, generating 15,617 tasks across English, Hindi, and five Indic languages. We evaluate cross-lingual semantic parsing performance of state-of-the-art models (DeepSeek v3.2, MiniMax 2.7, LLaMA 3.3, Qwen3) across seven linguistic variants. Results show a 9.00% performance drop from English to Indic languages, revealing an "Indic Gap" driven by harder schema linking, increased structural ambiguity, and limited external knowledge. IndicDB serves as a rigorous benchmark for multilingual Text-to-SQL. Code and data: https://anonymous.4open.science/r/multilingualText2Sql-Indic--DDCC/

Deep learning-based surface electromyography (sEMG) gesture recognition is frequently bottlenecked by data scarcity and limited subject diversity. While synthetic data generation via Generative Adversarial Networks (GANs) and diffusion models has emerged as a promising augmentation strategy, these approaches often face challenges regarding training stability or inference efficiency. To bridge this gap, we propose EMGFlow, a conditional sEMG generation framework. To the best of our knowledge, this is the first study to investigate the application of Flow Matching (FM) and continuous-time generative modeling in the sEMG domain. To validate EMGFlow across three benchmark sEMG datasets, we employ a unified evaluation protocol integrating feature-based fidelity, distributional geometry, and downstream utility. Extensive evaluations show that EMGFlow outperforms conventional augmentation and GAN baselines, and provides stronger standalone utility than the diffusion baselines considered here under the train-on-synthetic test-on-real (TSTR) protocol. Furthermore, by optimizing generation dynamics through advanced numerical solvers and targeted time sampling, EMGFlow achieves improved quality-efficiency trade-offs. Taken together, these results suggest that Flow Matching is a promising and efficient paradigm for addressing data bottlenecks in myoelectric control systems. Our code is available at: https://github.com/Open-EXG/EMGFlow.

The node2vec random walk is a non-Markovian random walk on the vertex set of a graph, widely used for network embedding and exploration. This random walk model is defined in terms of three parameters which control the probability of, respectively, backtracking moves, moves within triangles, and moves to the remaining neighboring nodes. From a mathematical standpoint, the node2vec random walk is a nontrivial generalization of the non-backtracking random walk and thus belongs to the class of second-order Markov chains. Despite its widespread use in applications, little is known about its long-run behavior. The goal of this paper is to begin exploring its fundamental properties on arbitrary graphs. To this aim, we show how lifting the node2vec random walk to the state spaces of directed edges and directed wedges yields two distinct Markovian representations which are key for its asymptotic analysis. Using these representations, we find mild sufficient conditions on the underlying finite or infinite graph to guarantee ergodicity, reversibility, recurrence and characterization of the invariant measure. As we discuss, the behavior of the node2vec random walk is drastically different compared to the non-backtracking random walk. While the latter simplifies on arbitrary graphs when using its natural edge Markovian representation thanks to bistochasticity, the former simplifies on regular graphs when using its natural wedge Markovian representation. Remarkably, this representation reveals that a graph is regular if and only if a certain weighted Eulerianity condition holds.

The `spotoptim` package implements surrogate-model-based optimization of expensive black-box functions in Python. Building on two decades of Sequential Parameter Optimization (SPO) methodology, it provides a Kriging-based optimization loop with Expected Improvement, support for continuous, integer, and categorical variables, noise-aware evaluation via Optimal Computing Budget Allocation (OCBA), and multi-objective extensions. A steady-state parallelization strategy overlaps surrogate search with objective evaluation on multi-core hardware, and a success-rate-based restart mechanism detects stagnation while preserving the best solution found. The package returns scipy-compatible `OptimizeResult` objects and accepts any scikit-learn-compatible surrogate model. Built-in TensorBoard logging provides real-time monitoring of convergence and surrogate quality. This report describes the architecture and module structure of spotoptim, provides worked examples including neural network hyperparameter tuning, and compares the framework with BoTorch, Optuna, Ray Tune, BOHB, SMAC, and Hyperopt. The package is open-source.

Assessing teachers' geometric content knowledge is essential for geometry instructional quality and student learning, but difficult to scale. The Van Hiele model characterizes geometric reasoning through five hierarchical levels. Traditional Van Hiele assessment relies on manual expert analysis of open-ended responses. This process is time-consuming, costly, and prevents large-scale evaluation. This study develops an automated approach for diagnosing teachers' Van Hiele reasoning levels using large language models grounded in educational theory. Our central hypothesis is that integrating explicit skills information significantly improves Van Hiele classification. In collaboration with mathematics education researchers, we built a structured skills dictionary decomposing the Van Hiele levels into 33 fine-grained reasoning skills. Through a custom web platform, 31 pre-service teachers solved geometry problems, yielding 226 responses. Expert researchers then annotated each response with its Van Hiele level and demonstrated skills from the dictionary. Using this annotated dataset, we implemented two classification approaches: (1) retrieval-augmented generation (RAG) and (2) multi-task learning (MTL). Each approach compared a skills-aware variant incorporating the skills dictionary against a baseline without skills information. Results showed that for both methods, skills-aware variants significantly outperformed baselines across multiple evaluation metrics. This work provides the first automated approach for Van Hiele level classification from open-ended responses. It offers a scalable, theory-grounded method for assessing teachers' geometric reasoning that can enable large-scale evaluation and support adaptive, personalized teacher learning systems.

Tuning of gate-defined semiconductor quantum dots (QDs) is a major bottleneck for scaling spin qubit technologies. We present a deep learning (DL) driven, semantic-segmentation pipeline that performs charge auto-tuning by locating transition lines in full charge stability diagrams (CSDs) and returns gate voltage targets for the single charge regime. We assemble and manually annotate a large, heterogeneous dataset of 1015 experimental CSDs measured from silicon QD devices, spanning nine design geometries, multiple wafers, and fabrication runs. A U-Net style convolutional neural network (CNN) with a MobileNetV2 encoder is trained and validated through five-fold group cross validation. Our model achieves an overall offline tuning success of 80.0% in locating the single-charge regime, with peak performance exceeding 88% for some designs. We analyze dominant failure modes and propose targeted mitigations. Finally, wide-range diagram segmentation also naturally enables scalable physic-based feature extraction that can feed back to fabrication and design workflows and outline a roadmap for real-time integration in a cryogenic wafer prober. Overall, our results show that neural network (NN) based wide-diagram segmentation is a practical step toward automated, high-throughput charge tuning for silicon QD qubits.

Advanced deep learning methods have shown remarkable success in power quality disturbance (PQD) classification. To enhance model transparency, explainable AI (XAI) techniques have been developed to provide instance-specific interpretations of classifier decisions. However, conventional XAI methods yield deterministic explanations, overlooking uncertainty and limiting reliability in safety-critical applications. This paper proposes a Bayesian explanation framework that models explanation uncertainty by generating a relevance attribution distribution for each instance. This method allows experts to select explanations based on confidence percentiles, thereby tailoring interpretability according to specific disturbance types. Extensive experiments on synthetic and real-world power quality datasets demonstrate that the proposed framework improves the transparency and reliability of PQD classifiers through uncertainty-aware explanations.

The statistical essence of the Transformer architecture has long remained elusive: Is it a universal approximator, or a neural network version of known computational algorithms? Through rigorous algebraic proof, we show that the latter better describes Transformer's basic nature: Ordinary Least Squares (OLS) is a special case of the single-layer Linear Transformer. Using the spectral decomposition of the empirical covariance matrix, we construct a specific parameter setting where the attention mechanism's forward pass becomes mathematically equivalent to the OLS closed-form projection. This means attention can solve the problem in one forward pass, not by iterating. Building upon this prototypical case, we further uncover a decoupled slow and fast memory mechanism within Transformers. Finally, the evolution from our established linear prototype to standard Transformers is discussed. This progression facilitates the transition of the Hopfield energy function from linear to exponential memory capacity, thereby establishing a clear continuity between modern deep architectures and classical statistical inference.

Front-end development constitutes a substantial portion of software engineering, yet converting design mockups into production-ready User Interface (UI) code remains tedious and costly. While recent work has explored automating this process with Multimodal Large Language Models (MLLMs), existing approaches typically rely solely on design images. As a result, they must infer complex UI details from images alone, often leading to degraded results. In real-world development workflows, however, design mockups are usually delivered as Figma files, a widely used tool for front-end design, that embed rich multimodal information (e.g., metadata and assets) essential for generating high-quality UI. To bridge this gap, we introduce Figma2Code, a new task that advances design-to-code into a multimodal setting and aims to automate design-to-code in the wild. Specifically, we collect paired design images and their corresponding metadata files from the Figma community. We then apply a series of processing operations, including rule-based filtering, human- and MLLM-based annotation and screening, and metadata refinement. This process yields 3,055 samples, from which designers curate a balanced dataset of 213 high-quality cases. Using this dataset, we benchmark ten state-of-the-art open-source and proprietary MLLMs. Our results show that while proprietary models achieve superior visual fidelity, they remain limited in layout responsiveness and code maintainability. Further experiments across modalities and ablation studies corroborate this limitation, partly due to models' tendency to directly map primitive visual attributes from Figma metadata.

Co-training, which combines limited in-domain real-world data with abundant surrogate data such as simulation or cross-embodiment robot data, is widely used for training generative robot policies. Despite its empirical success, the mechanisms that determine when and why co-training is effective remain poorly understood. We investigate the mechanism of sim-and-real co-training through theoretical analysis and empirical study, and identify two intrinsic effects governing performance. The first, \textbf{``structured representation alignment"}, reflects a balance between cross-domain representation alignment and domain discernibility, and plays a primary role in downstream performance. The second, the \textbf{``importance reweighting effect"}, arises from domain-dependent modulation of action weighting and operates at a secondary level. We validate these effects with controlled experiments on a toy model and extensive sim-and-sim and sim-and-real robot manipulation experiments. Our analysis offers a unified interpretation of recent co-training techniques and motivates a simple method that consistently improves upon prior approaches. More broadly, our aim is to examine the inner workings of co-training and to facilitate research in this direction.

Speculative decoding accelerates autoregressive generation by letting draft tokens bypass full verification, but conventional frameworks suffer from frequent false rejections, particularly when draft models produce semantically correct but lexically divergent outputs. In this paper, we present Calibrated Speculative Decoding (CSD), a training-free framework that recovers valid tokens discarded by standard verification. Guided by the principle of "Frequency-Guided Candidate Selection and Probability-Guarded Acceptance," CSD incorporates two lightweight modules: Online Correction Memory, which aggregates historical rejections to propose recurring divergence patterns as rescue candidates, and Semantic Consistency Gating, which verifies candidate admissibility using probability ratios instead of exact token matching. Our evaluation across diverse large language models demonstrates that CSD outperforms existing methods, achieving a peak throughput speedup of 2.33x. CSD preserves model accuracy across all tasks while further boosting performance on complex reasoning datasets. These results establish CSD as a highly effective, lightweight solution for practical LLM deployments.

The performance of large language model (LLM) agents depends critically on the execution harness, the system layer that orchestrates tool use, context management, and state persistence. Yet this same architectural centrality makes the harness a high-value attack surface: a single compromise at the harness level can cascade through the entire execution pipeline. We observe that existing security approaches suffer from structural mismatch, leaving them blind to harness-internal state and unable to coordinate across the different phases of agent operation. In this paper, we introduce \safeharness{}, a security architecture in which four proposed defense layers are woven directly into the agent lifecycle to address above significant limitations: adversarial context filtering at input processing, tiered causal verification at decision making, privilege-separated tool control at action execution, and safe rollback with adaptive degradation at state update. The proposed cross-layer mechanisms tie these layers together, escalating verification rigor, triggering rollbacks, and tightening tool privileges whenever sustained anomalies are detected. We evaluate \safeharness{} on benchmark datasets across diverse harness configurations, comparing against four security baselines under five attack scenarios spanning six threat categories. Compared to the unprotected baseline, \safeharness{} achieves an average reduction of approximately 38\% in UBR and 42\% in ASR, substantially lowering both the unsafe behavior rate and the attack success rate while preserving core task utility.

Supervised fine-tuning (SFT) is a common first stage of LLM post-training, teaching the model to follow instructions and shaping its behavior as a helpful assistant. At the same time, SFT may harm the fundamental capabilities of an LLM, particularly after long pretraining: a phenomenon known as catastrophic overtraining (Springer et al., 2025). To understand overtraining, we first investigate catastrophic forgetting in finetuning through the lens of implicit regularization of the learning rate. For models trained to the same SFT loss, we identify how the learning rate mediates optimization: finetuning with large and small steps converges to qualitatively different models. Next, we link forgetting to overtraining: learning rate decay increases the sharpness of the pretrained model, which in turn exacerbates catastrophic forgetting during SFT, leading to overtraining. Our findings paint a picture of the overtraining mechanism in LLMs and broadly contribute to the understanding of the interplay between optimization dynamics during pretraining and finetuning.

Self-Organizing Maps (SOM) are a classical method for unsupervised learning, vector quantization, and topographic mapping of high-dimensional data. However, existing SOM formulations often involve a trade-off between computational efficiency and a clearly defined optimization objective. Objective-based variants such as Soft Topographic Vector Quantization (STVQ) provide a principled formulation, but their neighborhood-coupled computations become expensive as the number of latent nodes increases. In this paper, we propose Self-Organizing Maps with Optimized Latent Positions (SOM-OLP), an objective-based topographic mapping method that introduces a continuous latent position for each data point. Starting from the neighborhood distortion of STVQ, we construct a separable surrogate local cost based on its local quadratic structure and formulate an entropy-regularized objective based on it. This yields a simple block coordinate descent scheme with closed-form updates for assignment probabilities, latent positions, and reference vectors, while guaranteeing monotonic non-increase of the objective and retaining linear per-iteration complexity in the numbers of data points and latent nodes. Experiments on a synthetic saddle manifold, scalability studies on the Digits and MNIST datasets, and 16 benchmark datasets show that SOM-OLP achieves competitive neighborhood preservation and quantization performance, favorable scalability for large numbers of latent nodes and large datasets, and the best average rank among the compared methods on the benchmark datasets.

Managing natural dialogue timing is a significant challenge for voice-based chatbots. Most current systems usually rely on simple silence detection, which often fails because human speech patterns involve irregular pauses. This causes bots to interrupt users, breaking the conversational flow. This problem is even more severe for languages like Turkish, which lack high-quality datasets for turn-taking prediction. This paper introduces Syn-TurnTurk, a synthetic Turkish dialogue dataset generated using various Qwen Large Language Models (LLMs) to mirror real-life verbal exchanges, including overlaps and strategic silences. We evaluated the dataset using several traditional and deep learning architectures. The results show that advanced models, particularly BI-LSTM and Ensemble (LR+RF) methods, achieve high accuracy (0.839) and AUC scores (0.910). These findings demonstrate that our synthetic dataset can have a positive affect for models understand linguistic cues, allowing for more natural human-machine interaction in Turkish.

Rubric-augmented verification guides reward models with explicit evaluation criteria, yielding more reliable judgments than single-model verification. However, most existing methods require costly rubric annotations, limiting scalability. Moreover, we find that rubric generation is vulnerable to a failure of cooperation; low-quality rubrics actively mislead reward models rather than help. Inspired by the principle of cooperative communication, we propose Cooperative yet Critical reward modeling (C2), a framework that significantly improves reward model judgments by having the reward model critically collaborate with a rubric generator trained solely from binary preferences. In C2, we synthesize helpful and misleading rubric pairs by measuring how each rubric shifts the reward model toward or away from the correct preference. Using these contrastive pairs, we train a cooperative rubric generator to propose helpful rubrics, and a critical verifier to assess rubric validity before making its judgment, following only rubrics it deems helpful at inference time. C2 outperforms reasoning reward models trained on the same binary preferences, with gains of up to 6.5 points on RM-Bench and 6.0 points length-controlled win rate on AlpacaEval 2.0. Without external rubric annotations, C2 enables an 8B reward model to match performance achieved with rubrics from a 4$\times$ larger model. Overall, our work demonstrates that eliciting deliberate cooperation in rubric-augmented verification makes reward models more trustworthy in a scalable way.

Smart contracts are a critical component of blockchain systems. Due to the large amount of digital assets carried by smart contracts, their security is of critical importance. Although numerous tools have been developed for detecting smart contract vulnerability, their effectiveness remains limited, particularly due to the high false positives included in the reported results. Therefore, developers and auditors are often overwhelmed with manually verifying the reported issues. A fundamental reason behind this is that while a reported vulnerability satisfies specific vulnerable patterns, it may not actually be exploitable, either because the vulnerable code cannot be triggered or it does not result in any financial loss. In this paper, we propose V2E, a new framework for validating whether a reported vulnerability is truly exploitable. The core idea of V2E is to automatically generate executable Proof-of-Concept Exploit (PoC for short), and then assess if the vulnerability could be triggered and incur any real damage (i.e., causing financial loss) by the PoC. While LLMs have shown proficiency in PoC generation, achieving our task is by no means trivial. In detail, it is difficult for LLM to: (1) generate and update PoC to trigger a specific vulnerability, (2) evaluate the PoC's effectiveness to validate exploitable vulnerability. To this end, V2E automates the whole process through a novel combination of PoC generation, validation, and refinement: (1) Firstly, V2E generates targeted PoCs by analyzing potential vulnerability paths. (2) Then, V2E verifies the validity of PoCs through triggerability and profitability analysis. (3) In addition, V2E iteratively refines the generated PoC based on PoC execution feedback, therefore, increasing the chance to confirm the vulnerability. Evaluation on 264 manually labeled contracts shows that V2E outperforms the baseline approach.

Reinforcement learners can attain high reward through novel unintended strategies. We study a Bayesian mitigation for general environments: we expand the agent's subjective reward range to include a large negative value $-L$, while the true environment's rewards lie in $[0,1]$. After observing consistently high rewards, the Bayesian policy becomes risk-averse to novel schemes that plausibly lead to $-L$. We design a simple override mechanism that yields control to a safe mentor whenever the predicted value drops below a fixed threshold. We prove two properties of the resulting agent: (i) Capability: using mentor-guided exploration with vanishing frequency, the agent attains sublinear regret against its best mentor. (ii) Safety: no decidable low-complexity predicate is triggered by the optimizing policy before it is triggered by a mentor.

Hybrid Quantum Neural Networks (HQNNs) have recently emerged as a promising paradigm for near-term quantum machine learning. However, their practical performance strongly depends on design choices such as classical-to-quantum data encoding, quantum circuit architecture, measurement strategy and shots. In this paper, we present a comprehensive design space exploration of HQNNs for Chronic Kidney Disease (CKD) diagnosis. Using a carefully curated and preprocessed clinical dataset, we benchmark 625 different HQNN models obtained by combining five encoding schemes, five entanglement architectures, five measurement strategies, and five different shot settings. To ensure fair and robust evaluation, all models are trained using 10-fold stratified cross-validation and assessed on a test set using a comprehensive set of metrics, including accuracy, area under the curve (AUC), F1-score, and a composite performance score. Our results reveal strong and non-trivial interactions between encoding choices and circuit architectures, showing that high performance does not necessarily require large parameter counts or complex circuits. In particular, we find that compact architectures combined with appropriate encodings (e.g., IQP with Ring entanglement) can achieve the best trade-off between accuracy, robustness, and efficiency. Beyond absolute performance analysis, we also provide actionable insights into how different design dimensions influence learning behavior in HQNNs.

Binary stellar evolution simulations are computationally expensive. Stellar population synthesis relies on these detailed evolution models at a fundamental level. Producing thousands of such models requires hundreds of CPU hours, but stellar track interpolation provides one approach to significantly reduce this computational cost. Although single-star track interpolation is straightforward, stellar interactions in binary systems introduce significant complexity to binary evolution, making traditional single-track interpolation methods inapplicable. Binary tracks present fundamentally different challenges compared to single stars, which possess relatively straightforward evolutionary phases identifiable through distinct physical properties. Binary systems are complicated by mutual interactions that can dramatically alter evolutionary trajectories and introduce discontinuities difficult to capture through standard interpolation. In this work, we introduce a novel approach for track alignment and iterative track averaging based on Dynamic Time Warping to address misalignments between neighboring tracks. Our method computes a single shared warping path across all physical parameters simultaneously, placing them on a consistent temporal grid that preserves the causal relationships between parameters. We demonstrate that this joint-alignment strategy maintains key physical relationships such as the Stefan-Boltzmann law in the interpolated tracks. Our comprehensive evaluation across multiple binary configurations demonstrates that proper temporal alignment is crucial for track interpolation methods. The proposed method consistently outperforms existing approaches and enables the efficient generation of more accurate binary population samples for astrophysical studies.

Reinforcement Learning from Human Feedback (RLHF) and related alignment paradigms have become central to steering large language models (LLMs) and multimodal large language models (MLLMs) toward human-preferred behaviors. However, these approaches introduce a systemic vulnerability: reward hacking, where models exploit imperfections in learned reward signals to maximize proxy objectives without fulfilling true task intent. As models scale and optimization intensifies, such exploitation manifests as verbosity bias, sycophancy, hallucinated justification, benchmark overfitting, and, in multimodal settings, perception--reasoning decoupling and evaluator manipulation. Recent evidence further suggests that seemingly benign shortcut behaviors can generalize into broader forms of misalignment, including deception and strategic gaming of oversight mechanisms. In this survey, we propose the Proxy Compression Hypothesis (PCH) as a unifying framework for understanding reward hacking. We formalize reward hacking as an emergent consequence of optimizing expressive policies against compressed reward representations of high-dimensional human objectives. Under this view, reward hacking arises from the interaction of objective compression, optimization amplification, and evaluator--policy co-adaptation. This perspective unifies empirical phenomena across RLHF, RLAIF, and RLVR regimes, and explains how local shortcut learning can generalize into broader forms of misalignment, including deception and strategic manipulation of oversight mechanisms. We further organize detection and mitigation strategies according to how they intervene on compression, amplification, or co-adaptation dynamics. By framing reward hacking as a structural instability of proxy-based alignment under scale, we highlight open challenges in scalable oversight, multimodal grounding, and agentic autonomy.

Recent reinforcement learning (RL) approaches have advanced radiology report generation (RRG), yet two core limitations persist: (1) report-level rewards offer limited evidence-grounded guidance for clinical faithfulness; and (2) current methods lack an explicit self-improving mechanism to align with clinical preference. We introduce clinically aligned Evidence-aware Self-Correcting Reinforcement Learning (ESC-RL), comprising two key components. First, a Group-wise Evidence-aware Alignment Reward (GEAR) delivers group-wise, evidence-aware feedback. GEAR reinforces consistent grounding for true positives, recovers missed findings for false negatives, and suppresses unsupported content for false positives. Second, a Self-correcting Preference Learning (SPL) strategy automatically constructs a reliable, disease-aware preference dataset from multiple noisy observations and leverages an LLM to synthesize refined reports without human supervision. ESC-RL promotes clinically faithful, disease-aligned reward and supports continual self-improvement during training. Extensive experiments on two public chest X-ray datasets demonstrate consistent gains and state-of-the-art performance.

Binary droplet collisions are ubiquitous in dense sprays. Traditional deterministic models cannot adequately represent transitional and stochastic behaviors of binary droplet collision. To bridge this gap, we developed a probabilistic model by using a machine learning approach, the Light Gradient-Boosting Machine (LightGBM). The model was trained on a comprehensive dataset of 33,540 experimental cases covering eight collision regimes across broad ranges of Weber number, Ohnesorge number, impact parameter, size ratio, and ambient pressure. The resulting machine learning classifier captures highly nonlinear regime boundaries with 99.2% accuracy and retains sensitivity in transitional regions. To facilitate its implementation in spray simulation, the model was translated into a probabilistic form, a multinomial logistic regression, which preserves 93.2% accuracy and maps continuous inter-regime transitions. A biased-dice sampling mechanism then converts these probabilities into definite yet stochastic outcomes. This work presents the first probabilistic, high-dimensional droplet collision model derived from experimental data, offering a physically consistent, comprehensive, and user-friendly solution for spray simulation.

Reasoning capabilities in large language models (LLMs) have generally advanced significantly. However, it is still challenging for existing reasoning-based LLMs to perform effective decision-making abilities in multi-agent environments, due to the absence of explicit foresight modeling. To this end, strategic reasoning, the most fundamental capability to anticipate the counterpart's behaviors and foresee its possible future actions, has been introduced to alleviate the above issues. Strategic reasoning is fundamental to effective decision-making in multi-agent environments, yet existing reasoning enhancement methods for LLMs do not explicitly capture its foresight nature. In this work, we introduce Foresight Policy Optimization (FoPO) to enhance strategic reasoning in LLMs, which integrates opponent modeling principles into policy optimization, thereby enabling explicit consideration of both self-interest and counterpart influence. Specifically, we construct two curated datasets, namely Cooperative RSA and Competitive Taboo, equipped with well-designed rules and moderate difficulty to facilitate a systematic investigation of FoPO in a self-play framework. Our experiments demonstrate that FoPO significantly enhances strategic reasoning across LLMs of varying sizes and origins. Moreover, models trained with FoPO exhibit strong generalization to out-of-domain strategic scenarios, substantially outperforming standard LLM reasoning optimization baselines.

Evaluating large language models (LLMs) for legal reasoning requires workflows that span task design, expert annotation, model execution, and metric-based evaluation. In practice, these steps are split across platforms and scripts, limiting transparency, reproducibility, and participation by non-technical legal experts. We present the BenGER (Benchmark for German Law) framework, an open-source web platform that integrates task creation, collaborative annotation, configurable LLM runs, and evaluation with lexical, semantic, factual, and judge-based metrics. BenGER supports multi-organization projects with tenant isolation and role-based access control, and can optionally provide formative, reference-grounded feedback to annotators. We will demonstrate a live deployment showing end-to-end benchmark creation and analysis.

Conventional Retrieval-Augmented Generation (RAG) systems often struggle with complex multi-hop queries over long documents due to their single-pass retrieval. We introduce MM-Doc-R1, a novel framework that employs an agentic, vision-aware workflow to address long document visual question answering through iterative information discovery and synthesis. To incentivize the information seeking capabilities of our agents, we propose Similarity-based Policy Optimization (SPO), addressing baseline estimation bias in existing multi-turn reinforcement learning (RL) algorithms like GRPO. Our core insight is that in multi-turn RL, the more semantically similar two trajectories are, the more accurate their shared baseline estimation becomes. Leveraging this, SPO calculates a more precise baseline by similarity-weighted averaging of rewards across multiple trajectories, unlike GRPO which inappropriately applies the initial state's baseline to all intermediate states. This provides a more stable and accurate learning signal for our agents, leading to superior training performance that surpasses GRPO. Our experiments on the MMLongbench-Doc benchmark show that MM-Doc-R1 outperforms previous baselines by 10.4%. Furthermore, SPO demonstrates superior performance over GRPO, boosting results by 5.0% with Qwen3-8B and 6.1% with Qwen3-4B. These results highlight the effectiveness of our integrated framework and novel training algorithm in advancing the state-of-the-art for complex, long-document visual question answering.

Brain digital twins aim to provide faithful, individualized computational representations of brains as dynamical systems, enabling mechanistic understanding and supporting prediction of clinical interventions. Yet current approaches remain fragmented across data pipelines, model classes, temporal scales, and computing platforms, which prevents the preservation of execution semantics across the end-toend workflow. This survey introduces physically constrained executability as a unifying perspective for comparing approaches at the level of execution: whether an execution state is persistent, which events are permitted to update it (simulation, measurement, actuation), and how strongly execution is temporally and causally coupled to neurobiological dynamics. Building on modeling and simulation theory, I propose a taxonomy of execution regimes ranging from isolated offline models to coordinated co-simulation, to continuously executing digital twins sustained by online data assimilation, and ultimately to neuro-neuromorphic physical systems in which biological and computational dynamics are co-executed under shared physical constraints. The executability concept clarifies why accuracy alone is insufficient, and motivates an agenda centered on semantic interoperability, hybrid-time correctness, evaluation protocols, scalable reproducible workflows, and safe closed-loop validation. This survey adopts a systems and runtime-oriented perspective, enabling comparison of heterogeneous approaches based on their execution semantics rather than on model form or application domain alone.

Heart sound signals, phonocardiography (PCG) signals, allow for the automatic diagnosis of potential cardiovascular pathology. Such classification task can be tackled using the bidirectional long short-term memory (biLSTM) network, trained on features extracted from labeled PCG signals. Regarding the non-stationarity of PCG signals, it is recommended to extract the features from multiple short-length segments of the signals using a sliding window of certain shape and length. However, some window contains unfavorable spectral side lobes, which distort the features. Accordingly, it is preferable to adapt the window shape and length in terms of classification performance. We propose an experimental evaluation for three window shapes, each with three window lengths. The biLSTM network is trained and tested on statistical features extracted, and the performance is reported in terms of the window shapes and lengths. Results show that the best performance is obtained when the Gaussian window is used for splitting the signals, and the triangular window competes with the Gaussian window for a length of 75 ms. Although the rectangular window is a commonly offered option, it is the worst choice for splitting the signals. Moreover, the classification performance obtained with a 75 ms Gaussian window outperforms that of a baseline method.

Ultra-high-resolution (UHR) remote sensing imagery couples kilometer-scale context with query-critical evidence that may occupy only a few pixels. Such vast spatial scale leads to a quadratic explosion of visual tokens and hinders the extraction of information from small objects. Previous works utilize direct downsampling, dense tiling, or global top-k pruning, which either compromise query-critical image details or incur unpredictable compute. In this paper, we propose UHR-BAT, a query-guided and region-faithful token compression framework to efficiently select visual tokens under a strict context budget. Specifically, we leverage text-guided, multi-scale importance estimation for visual tokens, effectively tackling the challenge of achieving precise yet low-cost feature extraction. Furthermore, by introducing region-wise preserve and merge strategies, we mitigate visual token redundancy, further driving down the computational budget. Experimental results show that UHR-BAT achieves state-of-the-art performance across various benchmarks. Code will be available at https://github.com/Yunkaidang/UHR.

Vision-language models trained with contrastive learning on paired medical images and reports show strong zero-shot diagnostic capabilities, yet the effect of training batch composition on learned representations remains unexplored for 3D medical imaging. We reproduce Merlin, a dual-encoder model that aligns 3D abdominal CT volumes with radiology reports using symmetric InfoNCE loss, achieving a zero-shot macro F1 of 74.45% across 30 findings (original: 73.00%). We then investigate two axes of variation. First, we control the normal-to-abnormal ratio within training batches at 25:75, 50:50, and 75:25 using section-level balanced sampling on the full dataset. All three configurations underperform the unbalanced baseline by 2.4 to 2.8 points, with 75:25 achieving the best result (72.02%) among balanced variants. Second, we conduct data scaling ablations on a 4,362-study subset, training with 20%, 40%, and 100% of the data. Performance scales sub-linearly from 65.26% to 71.88%, with individual findings varying dramatically in data sensitivity. Enforcing 50:50 balanced sampling on the same subset further degrades performance to 68.01%, confirming that explicit class balancing hurts regardless of dataset or balancing granularity. Our results indicate that the stochastic diversity of random sampling, combined with Merlin's alternating batching over anatomical subsections, provides more effective regularization than engineered class ratios at the small batch sizes required by 3D medical volumes.

Multi-task learning (MTL) improves generalization and data efficiency by jointly learning related tasks through shared representations. In the widely used hard-parameter-sharing setting, a shared backbone is combined with task-specific prediction heads. However, task-specific parameters can grow rapidly with the number of tasks. Therefore, designing multi-task heads that preserve task specialization while improving parameter efficiency remains a key challenge. In Quantum Machine Learning (QML), variational quantum circuits (VQCs) provide a compact mechanism for mapping classical data to quantum states residing in high-dimensional Hilbert spaces, enabling expressive representations within constrained parameter budgets. We propose a parameter-efficient quantum multi-task learning (QMTL) framework that replaces conventional task-specific linear heads with a fully quantum prediction head in a hybrid architecture. The model consists of a VQC with a shared, task-independent quantum encoding stage, followed by lightweight task-specific ansatz blocks enabling localized task adaptation while maintaining compact parameterization. Under a controlled and capacity-matched formulation where the shared representation dimension grows with the number of tasks, our parameter-scaling analysis demonstrates that a standard classical head exhibits quadratic growth, whereas the proposed quantum head parameter cost scales linearly. We evaluate QMTL on three multi-task benchmarks spanning natural language processing, medical imaging, and multimodal sarcasm detection, where we achieve performance comparable to, and in some cases exceeding, classical hard-parameter-sharing baselines while consistently outperforming existing hybrid quantum MTL models with substantially fewer head parameters. We further demonstrate QMTL's executability on noisy simulators and real quantum hardware, illustrating its feasibility.

Scenario testing is an important technique for detecting errors in web systems. Testers draft test scenarios and convert them into test scripts for execution. Early methods relied on testers to convert test scenarios into test scripts. Recent LLM-based scenario testing methods can generate test scripts from natural language descriptions of test scenarios. However, these methods are not only limited by the incompleteness of descriptions but also overlook test adequacy criteria, making it difficult to detect potential errors. To address these limitations, this paper proposes WebMAC, a multi-agent collaborative framework for scenario testing of web systems. WebMAC can complete natural language descriptions of test scenarios through interactive clarification and transform adequate instantiated test scenarios via equivalence class partitioning. WebMAC consists of three multi-agent modules, responsible respectively for completing natural language descriptions of test scenarios, transforming test scenarios, and converting test scripts. We evaluated WebMAC on four web systems. Compared with the SOTA method, WebMAC improves the execution success rate of generated test scripts by 30%-60%, increases testing efficiency by 29%, and reduces token consumption by 47.6%. Furthermore, WebMAC can effectively detect more errors in web systems.

Cross-layer key-value (KV) compression has been found to be effective in efficient inference of large language models (LLMs). Although they reduce the memory consumption of the KV cache, such methods usually introduce non-negligible performance degradation. In this work, we aim to enhance the performance of YOCO, a cross-layer KV compression method that shares the KVs of the middle layer with the top-half layers. We propose YOCO++, an enhanced YOCO that incorporates a weighted residual connection between the KVs of each bottom-half layer and the bottom layer. Compared to YOCO, YOCO++ increases model capacity while maintaining the same training and inference efficiency. Our experiments show that YOCO++ achieves state-of-the-art performance among the cross-layer KV compression methods at a 50% KV cache compression rate, outperforming the standard Transformer.

Large language models (LLMs) demonstrate strong reasoning capabilities, but their performance often degrades under distribution shift. Existing test-time adaptation (TTA) methods rely on gradient-based updates that require white-box access and need substantial overhead, while training-free alternatives are either static or depend on external guidance. In this paper, we propose Training-Free Test-Time Contrastive Learning TF-TTCL, a training-free adaptation framework that enables a frozen LLM to improve online by distilling supervision from its own inference experiences. Specifically, TF-TTCL implements a dynamic "Explore-Reflect-Steer" loop through three core modules: 1) Semantic Query Augmentation first diversifies problem views via multi-agent role-playing to generate different reasoning trajectories; 2) Contrastive Experience Distillation then captures the semantic gap between superior and inferior trajectories, distilling them into explicit textual rules; and 3) Contextual Rule Retrieval finally activates these stored rules during inference to dynamically steer the frozen LLM toward robust reasoning patterns while avoiding observed errors. Extensive experiments on closed-ended reasoning tasks and open-ended evaluation tasks demonstrate that TF-TTCL consistently outperforms strong zero-shot baselines and representative TTA methods under online evaluation. Code is available at https://github.com/KevinSCUTer/TF-TTCL.

Entity alignment (EA) aims to identify entities referring to the same real-world object across different knowledge graphs (KGs). Recent approaches based on large language models (LLMs) typically obtain entity embeddings through knowledge representation learning and use embedding similarity to identify an alignment-uncertain entity set. For each uncertain entity, a candidate entity set (CES) is then retrieved based on embedding similarity to support subsequent alignment reasoning and decision making. However, the reliability of the CES and the reasoning capability of LLMs critically affect the effectiveness of subsequent alignment decisions. To address this issue, we propose AgentEA, a reliable EA framework based on multi-agent debate. AgentEA first improves embedding quality through entity representation preference optimization, and then introduces a two-stage multi-role debate mechanism consisting of lightweight debate verification and deep debate alignment to progressively enhance the reliability of alignment decisions while enabling more efficient debate-based reasoning. Extensive experiments on public benchmarks under cross-lingual, sparse, large-scale, and heterogeneous settings demonstrate the effectiveness of AgentEA.

Conventional neural networks strictly separate learning and inference because if parameters are updated during inference, outputs become unstable and even the inference function itself is not well defined [1, 2, 3]. This paper shows that DynamicGate MLP structurally permits learning inference concurrency [4, 5]. The key idea is to separate routing (gating) parameters from representation (prediction) parameters, so that the gate can be adapted online while inference stability is preserved, or weights can be selectively updated only within the inactive subspace [4, 5, 6, 7]. We mathematically formalize sufficient conditions for concurrency and show that even under asynchronous or partial updates, the inference output at each time step can always be interpreted as a forward computation of a valid model snapshot [8, 9, 10]. This suggests that DynamicGate MLP can serve as a practical foundation for online adaptive and on device learning systems [11, 12].

Long Short-Term Memory (LSTM) neural networks have penetrated healthcare applications where real-time requirements and edge computing capabilities are essential. Gait analysis that detects abnormal steps to prevent patients from falling is a prominent problem for such applications. Given the extremely stringent design requirements in performance, power dissipation, and area, an Application-Specific Integrated Circuit (ASIC) enables an efficient real-time exploitation of LSTMs for gait analysis, achieving high accuracy. To the best of our knowledge, this work presents the first cross-layer co-optimized LSTM accelerator for real-time gait analysis, targeting an ASIC design. We conduct a comprehensive design space exploration from software down to layout design. We carry out a bit-width optimization at the software level with hardware-aware quantization to reduce the hardware complexity, explore various designs at the register-transfer level, and generate alternative layouts to find efficient realizations of the LSTM accelerator in terms of hardware complexity and accuracy. The physical synthesis results show that, using the 65 nm technology, the die size of the accelerator's layout optimized for the highest accuracy is 0.325 mm^2, while the alternative design optimized for hardware complexity with a slightly lower accuracy occupies 15.4% smaller area. Moreover, the designed accelerators achieve accurate gait abnormality detection 4.05x faster than the given application requirement.

The growth of compute-intensive AI tasks highlights the need to mitigate the processing costs and improve performance and energy efficiency. This necessitates the integration of intelligent agents as architectural adaptation supervisors tasked with adaptive scaling of the infrastructure and efficient offloading of computation within the continuum. This paper presents a self-adaptation approach for an efficient computing system of a mixed human-robot environment. The computation task is associated with a Neural Network algorithm that leverages sensory data to predict human mobility behaviors, to enhance mobile robots' proactive path planning, and ensure human safety. To streamline neural network processing, we built a distributed edge offloading system with heterogeneous processing units, orchestrated by Kubernetes. By monitoring response times and power consumption, the MAPE-K-based adaptation supervisor makes informed decisions on scaling and offloading. Results show notable improvements in service quality over traditional setups, demonstrating the effectiveness of the proposed approach for AI-driven systems.

Unified Multimodal Models (UMMs) aim to integrate visual understanding and generation within a single structure. However, these models exhibit a notable capability mismatch, where their understanding capability significantly outperforms their generation. This mismatch indicates that the model's rich internal knowledge, while effective for understanding tasks, remains underactivated during generation. To address this, we draw inspiration from the human ``Thinking-While-Drawing'' paradigm, where humans continuously reflect to activate their knowledge and rectify intermediate results. In this paper, we propose UniRect-CoT, a training-free unified rectification chain-of-thought framework. Our approach unlocks the ``free lunch'' hidden in the UMM's powerful inherent understanding to continuously reflect, activating its internal knowledge and rectifying intermediate results during generation.We regard the diffusion denoising process in UMMs as an intrinsic visual reasoning process and align the intermediate results with the target instruction understood by the model, serving as a self-supervisory signal to rectify UMM generation.Extensive experiments demonstrate that UniRect-CoT can be easily integrated into existing UMMs, significantly enhancing generation quality across diverse complex tasks.

Using responses generated by high-performing large language models (LLMs) for instruction tuning has become a widely adopted approach. However, the existing literature overlooks a property of LLM-generated responses: they conflate world knowledge acquired during pre-training with instruction-following capabilities acquired during post-training. We hypothesize that disentangling the instruction-following capabilities from pre-trained knowledge improves the effectiveness of instruction tuning. To this end, we propose CoDIT, a method that applies contrastive decoding between a post-trained model and its pre-trained counterpart during response generation. The method suppresses pre-trained knowledge shared between the two models while amplifying the instruction-following behavior acquired via post-training, resulting in responses that more purely reflect instruction-following capabilities. Experiment results demonstrate that models trained on datasets constructed via CoDIT consistently outperform those trained on directly generated responses. Training on our datasets also yields better performance than on existing publicly available instruction-tuning datasets across multiple benchmarks. Furthermore, we theoretically and empirically show that CoDIT can be interpreted as distilling the chat vector from parameter space to text space, enabling the transfer of instruction-tuning capabilities across models of different architectures.

Graphical User Interface (GUI) agents show strong capabilities for automating web tasks, but existing interactive benchmarks primarily target benign, predictable consumer environments. Their effectiveness in high-stakes, investigative domains such as authentic e-commerce risk management remains underexplored. To bridge this gap, we present RiskWebWorld, the first highly realistic interactive benchmark for evaluating GUI agents in e-commerce risk management. RiskWebWorld features 1,513 tasks sourced from production risk-control pipelines across 8 core domains, and captures the authentic challenges of risk operations on uncooperative websites, partially environmental hijackments. To support scalable evaluation and agentic reinforcement learning (RL), we further build a Gymnasium-compliant infrastructure that decouples policy planning from environment mechanics. Our evaluation across diverse models reveals a dramatic capability gap: top-tier generalist models achieve 49.1% success, while specialized open-weights GUI models lag at near-total failure. This highlights that foundation model scale currently matters more than zero-shot interface grounding in long-horizon professional tasks. We also demonstrate the viability of our infrastructure through agentic RL, which improves open-source models by 16.2%. These results position RiskWebWorld as a practical testbed for developing robust digital workers.

The robust low-rank tensor completion problem addresses the challenge of recovering corrupted high-dimensional tensor data with missing entries, outliers, and sparse noise commonly found in real-world applications. Existing methodologies have encountered fundamental limitations due to their reliance on uniform regularization schemes, particularly the tensor nuclear norm and $\ell_1$ norm regularization approaches, which indiscriminately apply equal shrinkage to all singular values and sparse components, thereby compromising the preservation of critical tensor structures. The proposed tensor weighted correlated total variation (TWCTV) regularizer addresses these shortcomings through an $M$-product framework that combines a weighted Schatten-$p$ norm on gradient tensors for low-rankness with smoothness enforcement and weighted sparse components for noise suppression. The proposed weighting scheme adaptively reduces the thresholding level to preserve both dominant singular values and sparse components, thus improving the reconstruction of critical structural elements and nuanced details in the recovered signal. Through a systematic algorithmic approach, we introduce an enhanced alternating direction method of multipliers (ADMM) that offers both computational efficiency and theoretical substantiation, with convergence properties comprehensively analyzed within the $M$-product framework.Comprehensive numerical evaluations across image completion, denoising, and background subtraction tasks validate the superior performance of this approach relative to established benchmark methods.

Existing multi-source root cause analysis (RCA) methods for microservice systems assume all services have traces to construct a service call graph. However, this assumption is not practical as microservice systems evolve rapidly and may contain blackbox services without traces, such as compiled software or unsupported services. We refer to these services as blind spots. In the presence of blind spots, the performance of existing multi-source RCA methods may be affected, as they only diagnose visible services on the call graph. To overcome this limitation, we propose TORAI, a novel unsupervised approach that effectively pinpoints fine-grained root causes without relying on the service call graph. Instead, TORAI first measures anomaly severity using available multi-source telemetry data. It then performs clustering to group services based on their severity symptoms and conducts causal analysis to rank services within each severity cluster. Finally, TORAI aggregates the cluster rankings and uses hypothesis testing to identify fine-grained root causes. TORAI provides an unsupervised approach that leverages available multi-source telemetry data for RCA without requiring a constructed service call graph or further intrusive actions, thus addressing the limitations of existing methods. Our experiments on three benchmark systems demonstrate that TORAI outperforms state-of-the-art baselines remarkably in the presence of blind spots. Performance on real-world failures further shows that TORAI can accurately pinpoint the root causes in top-3 recommendations.

Neural network models with latent recurrent processing, where identical layers are recursively applied to the latent state, have gained attention as promising models for performing reasoning tasks. A strength of such models is that they enable test-time scaling, where the models can enhance their performance in the test phase without additional training. Models such as the Hierarchical Reasoning Model (HRM) and Artificial Kuramoto Oscillatory Neurons (AKOrN) can facilitate deeper reasoning by increasing the number of recurrent steps, thereby enabling the completion of challenging tasks, including Sudoku, Maze solving, and AGI benchmarks. In this work, we introduce confidence-based voting (C-voting), a test-time scaling strategy designed for recurrent models with multiple latent candidate trajectories. Initializing the latent state with multiple candidates using random variables, C-voting selects the one maximizing the average of top-1 probabilities of the predictions, reflecting the model's confidence. Additionally, it yields 4.9% higher accuracy on Sudoku-hard than the energy-based voting strategy, which is specific to models with explicit energy functions. An essential advantage of C-voting is its applicability: it can be applied to recurrent models without requiring an explicit energy function. Finally, we introduce a simple attention-based recurrent model with randomized initial values named ItrSA++, and demonstrate that when combined with C-voting, it outperforms HRM on Sudoku-extreme (95.2% vs. 55.0%) and Maze (78.6% vs. 74.5%) tasks.

Metal-organic frameworks (MOFs) are highly promising for carbon capture, yet navigating their vast design space remains challenging. Recent deep generative models enable de novo MOF design but primarily act as feed-forward structure generators. By heavily relying on predefined building block libraries and non-differentiable post-optimization, they fundamentally sever the information flow required for continuous structural editing. Here, we propose a target-driven generative framework focused on continuous structural manipulation. At its core is LinkerVAE, which maps discrete 3D chemical graphs into a continuous, SE(3)-equivariant latent space. This smooth manifold unlocks geometry-aware manipulations, including implicit chemical style transfer and zero-shot isoreticular expansion. Building upon this, we introduce a test-time optimization (TTO) strategy, utilizing an accurate surrogate model to continuously optimize the latent graphs of existing MOFs toward desired properties. This approach systematically enhances carbon capture performance, achieving a striking average relative boost of 147.5% in pure CO2 uptake while strictly preserving structural validity. Integrated with a latent diffusion model and rigid-body assembly for full MOF construction, our framework establishes a scalable, fully differentiable pathway for both the automated discovery, targeted optimization and editing of functional materials.

Tool calling has greatly expanded the practical utility of large language models (LLMs) by enabling them to interact with external applications. As LLM capabilities advance, effective tool use increasingly involves multi-step, multi-turn interactions to solve complex tasks. However, the resulting growth in tool interactions incurs substantial latency, posing a key challenge for real-time LLM serving. Through empirical analysis, we find that tool-calling traces are highly structured, conform to constrained schemas, and often exhibit recurring invocation patterns. Motivated by this, we propose ToolSpec, a schema-aware, retrieval-augmented speculative decoding method for accelerating tool calling. ToolSpec exploits predefined tool schemas to generate accurate drafts, using a finite-state machine to alternate between deterministic schema token filling and speculative generation for variable fields. In addition, ToolSpec retrieves similar historical tool invocations and reuses them as drafts to further improve efficiency. ToolSpec presents a plug-and-play solution that can be seamlessly integrated into existing LLM workflows. Experiments across multiple benchmarks demonstrate that ToolSpec achieves up to a 4.2x speedup, substantially outperforming existing training-free speculative decoding methods.

Self-supervised learning has emerged as a major technique for the task of learning from unlabeled data, where the current methods mostly revolve around alignment of representations and input recon struction. Although such approaches have demonstrated excellent performance in practice, their scope remains mostly confined to learning from observed data and does not provide much help in terms of a learning structure that is predictive of the data distribution. In this paper, we study some of the recent developments in the realm of self-supervised learning. We define a new category called Predictive Representation Learning (PRL), which revolves around the latent prediction of unobserved components of data based on the observation. We propose a common taxonomy that classifies PRL along with alignment and reconstruction-based learning approaches. Furthermore, we argue that Joint-Embedding Predictive Architecture(JEPA) can be considered as an exemplary member of this new paradigm. We further discuss theoretical perspectives and open challenges, highlighting predictive representation learning as a promising direction for future self-supervised learning research. In this study, we implemented Bootstrap Your Own Latent (BYOL), Masked Autoencoders (MAE), and Image-JEPA (I-JEPA) for comparative analysis. The results indicate that MAE achieves perfect similarity of 1.00, but exhibits relatively weak robustness of 0.55. In contrast, BYOL and I-JEPA attain accuracies of 0.98 and 0.95, with robustness scores of 0.75 and 0.78, respectively.

Temporal credit assignment in reinforcement learning has long been a central challenge. Inspired by the multi-timescale encoding of the dopamine system in neurobiology, recent research has sought to introduce multiple discount factors into Actor-Critic architectures, such as Proximal Policy Optimization (PPO), to balance short-term responses with long-term planning. However, this paper reveals that blindly fusing multi-timescale signals in complex delayed-reward tasks can lead to severe algorithmic pathologies. We systematically demonstrate that exposing a temporal attention routing mechanism to policy gradients results in surrogate objective hacking, while adopting gradient-free uncertainty weighting triggers irreversible myopic degeneration, a phenomenon we term the Paradox of Temporal Uncertainty. To address these issues, we propose a Target Decoupling architecture: on the Critic side, we retain multi-timescale predictions to enforce auxiliary representation learning, while on the Actor side, we strictly isolate short-term signals and update the policy based solely on long-term advantages. Rigorous empirical evaluations across multiple independent random seeds in the LunarLander-v2 environment demonstrate that our proposed architecture achieves statistically significant performance improvements. Without relying on hyperparameter hacking, it consistently surpasses the ''Environment Solved'' threshold with minimal variance, completely eliminates policy collapse, and escapes the hovering local optima that trap single-timescale baselines.

Supervised fine-tuning (SFT) followed by Group Relative Policy Optimization (GRPO) is a common post-training recipe. We conduct a controlled ablation over SFT-GRPO data overlap, evaluating Qwen3-8B (thinking disabled) post-trained for Lean 4 autoformalization under six conditions that differ solely in training recipe: a base model, SFT-only, GRPO-only, and three SFT+GRPO configurations where 0 percent, 30 percent, or 100 percent of the GRPO prompts coincide with the SFT corpus. Keeping SFT and GRPO data disjoint consistently outperforms full overlap at zero additional compute cost. Evaluating on Gaokao-Formal and PutnamBench under both compile pass at k and semantic pass at k assessed by an LLM judge, we find that lower overlap is monotonically associated with higher compilation and semantic accuracy. At 0 percent overlap, GRPO yields a 10.4 percentage point semantic gain over SFT alone on Gaokao, while at 100 percent overlap both metrics remain flat, rendering the GRPO stage effectively redundant. We further show that dual-metric evaluation reveals compile semantic gaps exceeding 30 percentage points for the highest compiling models, a disparity invisible under compile-only benchmarking. To our knowledge, this is the first controlled investigation of SFT-GRPO data overlap as a post-training hyperparameter, demonstrating how model behavior varies based on the degree of data sharing between training stages.

Designing effective reward functions is a cornerstone of reinforcement learning (RL), yet it remains a challenging and labor-intensive process due to the inefficiencies and inconsistencies inherent in traditional methods. Existing methods often rely on extensive manual design and evaluation steps, which are prone to redundancy and overlook local uncertainties at intermediate decision points. To address these challenges, we propose the Chain of Uncertain Rewards (CoUR), a novel framework that integrates large language models (LLMs) to streamline reward function design and evaluation in RL environments. Specifically, our CoUR introduces code uncertainty quantification with a similarity selection mechanism that combines textual and semantic analyses to identify and reuse the most relevant reward function components. By reducing redundant evaluations and leveraging Bayesian optimization on decoupled reward terms, CoUR enables a more efficient and robust search for optimal reward feedback. We comprehensively evaluate CoUR across nine original environments from IsaacGym and all 20 tasks from the Bidexterous Manipulation benchmark. The experimental results demonstrate that CoUR not only achieves better performance but also significantly lowers the cost of reward evaluations.

Social Determinants of Health (SDOH) refer to environmental, behavioral, and social conditions that influence how individuals live, work, and age. SDOH have a significant impact on personal health outcomes, and their systematic identification and management can yield substantial improvements in patient care. However, SDOH information is predominantly captured in unstructured clinical notes within electronic health records, which limits its direct use as machine-readable entities. To address this issue, researchers have employed Natural Language Processing (NLP) techniques using pre-trained BERT-based models, demonstrating promising performance but requiring sophisticated implementation and extensive computational resources. In this study, we investigated prompt engineering strategies for extracting structured SDOH events utilizing LLMs with advanced reasoning capabilities. Our method consisted of four modules: 1) developing concise and descriptive prompts integrated with established guidelines, 2) applying few-shot learning with carefully curated examples, 3) using a self-consistency mechanism to ensure robust outputs, and 4) post-processing for quality control. Our approach achieved a micro-F1 score of 0.866, demonstrating competitive performance compared to the leading models. The results demonstrated that LLMs with reasoning capabilities are effective solutions for SDOH event extraction, offering both implementation simplicity and strong performance.

Autonomous Graphical User Interface (GUI) agents powered by Multimodal Large Language Models (MLLMs) enable digital automation on end-user devices. While scaling both parameters and data has yielded substantial gains, advanced methods still suffer from prohibitive deployment costs on resource-constrained devices. When facing complex in-the-wild scenarios, lightweight GUI agents are bottlenecked by limited capacity and poor task scalability under end-to-end episodic learning, impeding adaptation to multi-agent systems (MAS), while training multiple skill-specific experts remains costly. Can we strike an effective trade-off in this cost-scalability dilemma, enabling lightweight MLLMs to participate in realistic GUI workflows? To address these challenges, we propose the LAMO framework, which endows a lightweight MLLM with GUI-specific knowledge and task scalability, allowing multi-role orchestration to expand its capability boundary for GUI automation. LAMO combines role-oriented data synthesis with a two-stage training recipe: (i) supervised fine-tuning with Perplexity-Weighted Cross-Entropy optimization for knowledge distillation and visual perception enhancement, and (ii) reinforcement learning for role-oriented cooperative exploration. With LAMO, we develop a task-scalable native GUI agent, LAMO-3B, supporting monolithic execution and MAS-style orchestration. When paired with advanced planners as a plug-and-play policy executor, LAMO-3B can continuously benefit from planner advances, enabling a higher performance ceiling. Extensive static and online evaluations validate the effectiveness of our design.

Clustering and dimensionality reduction have been crucial topics in machine learning and computer vision. Clustering high-dimensional data has been challenging for a long time due to the curse of dimensionality. For that reason, a more promising direction is the joint learning of dimension reduction and clustering. In this work, we propose a Manifold Learning Framework that learns dimensionality reduction and clustering simultaneously. The proposed framework is able to jointly learn the parameters of a dimension reduction technique (e.g. linear projection or a neural network) and cluster the data based on the resulting features (e.g. under a Gaussian Mixture Model framework). The framework searches for the dimension reduction parameters and the optimal clusters by traversing a manifold,using Gradient Manifold Optimization. The obtained The proposed framework is exemplified with a Gaussian Mixture Model as one simple but efficient example, in a process that is somehow similar to unsupervised Linear Discriminant Analysis (LDA). We apply the proposed method to the unsupervised training of simulated data as well as a benchmark image dataset (i.e. MNIST). The experimental results indicate that our algorithm has better performance than popular clustering algorithms from the literature.

Here, we describe Monthly Diffusion at 1.5-degree grid spacing (MD-1.5 version 0.9), a climate emulator that leverages a spherical Fourier neural operator (SFNO)-inspired Conditional Variational Auto-Encoder (CVAE) architecture to model the evolution of low-frequency internal atmospheric variability using latent diffusion. MDv0.9 was designed to forward-step at monthly mean timesteps in a data-sparse regime, using modest computational requirements. This work describes the motivation behind the architecture design, the MDv0.9 training procedure, and initial results.

We propose a novel end-to-end privacy-preserving framework, instantiated by three efficient protocols for different deployment scenarios, covering both input and output privacy, for the vertically split scenario in federated learning (FL), where features are split across clients and labels are not shared by all parties. We do so by distributing the role of the aggregator in FL into multiple servers and having them run secure multiparty computation (MPC) protocols to perform model and feature aggregation and apply differential privacy (DP) to the final released model. While a naive solution would have the clients delegating the entirety of training to run in MPC between the servers, our optimized solution, which supports purely global and also global-local models updates with privacy-preserving, drastically reduces the amount of computation and communication performed using multiparty computation. The experimental results also show the effectiveness of our protocols.

Cooperative multi-agent reinforcement learning (MARL) is widely used to address large joint observation and action spaces by decomposing a centralized control problem into multiple interacting agents. However, such decomposition often introduces additional challenges, including non-stationarity, unstable training, weak coordination, and limited theoretical guarantees. In this paper, we propose the Consensus Multi-Agent Transformer (CMAT), a centralized framework that bridges cooperative MARL to a hierarchical single-agent reinforcement learning (SARL) formulation. CMAT treats all agents as a unified entity and employs a Transformer encoder to process the large joint observation space. To handle the extensive joint action space, we introduce a hierarchical decision-making mechanism in which a Transformer decoder autoregressively generates a high-level consensus vector, simulating the process by which agents reach agreement on their strategies in latent space. Conditioned on this consensus, all agents generate their actions simultaneously, enabling order-independent joint decision making and avoiding the sensitivity to action-generation order in conventional Multi-Agent Transformers (MAT). This factorization allows the joint policy to be optimized using single-agent PPO while preserving expressive coordination through the latent consensus. To evaluate the proposed method, we conduct experiments on benchmark tasks from StarCraft II, Multi-Agent MuJoCo, and Google Research Football. The results show that CMAT achieves superior performance over recent centralized solutions, sequential MARL methods, and conventional MARL baselines. The code for this paper is available at:https://github.com/RS2002/CMAT .

In silico tools are important for generating novel hypotheses and exploring alternatives in de novo metabolic pathway design. However, while many computational frameworks have been proposed for retrobiosynthesis, few successful examples of algorithm-guided xenobiotic biochemical retrosynthesis have been reported in the literature. Deep learning has improved the quality of synthesis and retrosynthesis in organic chemistry applications. Inspired by this progress, we explored combining deep learning of biochemical transformations with the traditional retrobiosynthetic workflow to improve in silico synthetic metabolic pathway designs. To develop our computational biosynthetic pathway design framework, we assembled metabolic reaction and enzymatic template data from public databases. A data augmentation procedure, adapted from literature, was carried out to enrich the assembled reaction dataset with artificial metabolic reactions generated by enzymatic reaction templates. Two neural network-based pathway ranking models were trained as binary classifiers to distinguish assembled reactions from artificial counterparts; each model output a scalar quantifying the plausibility of a 1-step or 2-step pathway. Combining these two models with enzymatic templates, we built a multistep retrobiosynthesis pipeline and validated it by reproducing some natural and non-natural pathways computationally.

We prove that conditional diffusion models whose reverse kernels are finite Gaussian mixtures with ReLU-network logits can approximate suitably regular target distributions arbitrarily well in context-averaged conditional KL divergence, up to an irreducible terminal mismatch that typically vanishes with increasing diffusion horizon. A path-space decomposition reduces the output error to this mismatch plus per-step reverse-kernel errors; assuming each reverse kernel factors through a finite-dimensional feature map, each step becomes a static conditional density approximation problem, solved by composing Norets' Gaussian-mixture theory with quantitative ReLU bounds. Under exact terminal matching the resulting neural reverse-kernel class is dense in conditional KL.

Generative information retrieval (GenIR) formulates the retrieval process as a text-to-text generation task, leveraging the vast knowledge of large language models. However, existing works primarily optimize for relevance while often overlooking document trustworthiness. This is critical in high-stakes domains like healthcare and finance, where relying solely on semantic relevance risks retrieving unreliable information. To address this, we propose an Authority-aware Generative Retriever (AuthGR), the first framework that incorporates authority into GenIR. AuthGR consists of three key components: (i) Multimodal Authority Scoring, which employs a vision-language model to quantify authority from textual and visual cues; (ii) a Three-stage Training Pipeline to progressively instill authority awareness into the retriever; and (iii) a Hybrid Ensemble Pipeline for robust deployment. Offline evaluations demonstrate that AuthGR successfully enhances both authority and accuracy, with our 3B model matching a 14B baseline. Crucially, large-scale online A/B tests and human evaluations conducted on the commercial web search platform confirm significant improvements in real-world user engagement and reliability.

Ultrasonic metal welding (UMW) is widely used in industrial applications but is sensitive to tool wear, surface contamination, and material variability, which can lead to unexpected process faults and unsatisfactory weld quality. Conventional monitoring systems typically rely on supervised learning models that assume all fault types are known in advance, limiting their ability to handle previously unseen process faults. To address this challenge, this paper proposes an adaptive condition monitoring approach that enables unknown fault detection and few-shot continual learning for UMW. Unknown faults are detected by analyzing hidden-layer representations of a multilayer perceptron and leveraging a statistical thresholding strategy. Once detected, the samples from unknown fault types are incorporated into the existing model through a continual learning procedure that selectively updates only the final layers of the network, which enables the model to recognize new fault types while preserving knowledge of existing classes. To accelerate the labeling process, cosine similarity transformation combined with a clustering algorithm groups similar unknown samples, thereby reducing manual labeling effort. Experimental results using a multi-sensor UMW dataset demonstrate that the proposed method achieves 96% accuracy in detecting unseen fault conditions while maintaining reliable classification of known classes. After incorporating a new fault type using only five labeled samples, the updated model achieves 98% testing classification accuracy. These results demonstrate that the proposed approach enables adaptive monitoring with minimal retraining cost and time. The proposed approach provides a scalable solution for continual learning in condition monitoring where new process conditions may constantly emerge over time and is extensible to other manufacturing processes.

Mobile apps often suffer from functional bugs that do not cause crashes but instead manifest as incorrect behaviors under specific user interactions. Such bugs are difficult to detect automatically because they often lack explicit test oracles. Property-based testing can effectively expose them by checking intended behavioral properties under diverse interactions. However, its use largely depends on manually written properties, whose construction is difficult and expensive, limiting its practical use for mobile apps. To address this limitation, we propose PropGen, an automated approach for generating properties for Android apps. However, this task is challenging for two reasons: app functionalities are often hard to systematically uncover and execute, and properties are difficult to derive accurately from observed behaviors. To this end, PropGen performs functionality-guided exploration to collect behavioral evidence from app executions, synthesizes properties from the collected evidence, and refines imprecise properties based on testing feedback. We implemented PropGen and evaluated it on 12 real-world Android apps. The results show that PropGen can effectively identify and execute valid app functionalities, generate valid properties, and repair most imprecise ones. Across all apps, PropGen identified 1,210 valid functionalities and correctly executed 977 of them, compared with 491 and 187 for the baseline. It generated 985 properties, 912 of which were valid, and repaired 118 of 127 imprecise ones exposed during testing. With the resulting properties, we found 25 previously unknown functional bugs in the latest versions of the subject apps, many of which were missed by existing functional testing techniques.

Effective IT change management is important for businesses that depend on software and services, particularly in highly regulated sectors such as finance, where operational reliability, auditability, and explainability are essential. A significant portion of IT incidents are caused by changes, making it important to identify high-risk changes before deployment. This study presents a predictive incident risk scoring approach at a large international bank. The approach supports engineers during the assessment and planning phases of change deployments by predicting the potential of inducing incidents. To satisfy regulatory constraints, we built the model with auditability and explainability in mind, applying SHAP values to provide feature-level insights and ensure decisions are traceable and transparent. Using a one-year real-world dataset, we compare the existing rule-based process with three machine learning models: HGBC, LightGBM, and XGBoost. LightGBM achieved the best performance, particularly when enriched with aggregated team metrics that capture organisational context. Our results show that data-driven, interpretable models can outperform rule-based approaches while meeting compliance needs, enabling proactive risk mitigation and more reliable IT operations.

A central challenge in continual learning is forgetting, the loss of performance on previously learned tasks induced by sequential adaptation to new ones. While forgetting has been extensively studied empirically, rigorous theoretical characterizations remain limited. A notable step in this direction is \citet{evron2022catastrophic}, which analyzes forgetting under random orderings of a fixed task collection in overparameterized linear regression. We shift the perspective from order to distribution. Rather than asking how a fixed task collection behaves under random orderings, we study an exact-fit linear regime in which tasks are sampled i.i.d.\ from a task distribution~$Π$, and ask how the generating distribution itself governs forgetting. In this setting, we derive an exact operator identity for the forgetting quantity, revealing a recursive spectral structure. Building on this identity, we establish an unconditional upper bound, identify the leading asymptotic term, and, in generic nondegenerate cases, characterize the convergence rate up to constants. We further relate this rate to geometric properties of the task distribution, clarifying what drives slow or fast forgetting in this model.

Turbofan engine degradation under sustained operational stress necessitates robust prognostic systems capable of accurately estimating the Remaining Useful Life (RUL) of critical components. Existing deep learning approaches frequently fail to simultaneously capture multi-sensor spatial correlations and long-range temporal dependencies, while standard symmetric loss functions inadequately penalize the safety-critical error of over-estimating residual life. This study proposes a hybrid architecture integrating Twin-Stage One-Dimensional Convolutional Neural Networks (1D-CNN), a Bidirectional Long Short-Term Memory (BiLSTM) network, and a custom Bahdanau Additive Attention mechanism. The model was trained and evaluated on the NASA Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) FD001 sub-dataset employing a zero-leakage preprocessing pipeline, piecewise-linear RUL labeling capped at 130 cycles, and the NASA-specified asymmetric exponential loss function that disproportionately penalizes over-estimation to enforce industrial safety constraints. Experiments on 100 test engines achieved a Root Mean Squared Error (RMSE) of 17.52 cycles and a NASA S-Score of 922.06. Furthermore, extracted attention weight heatmaps provide interpretable, per-engine insights into the temporal progression of degradation, supporting informed maintenance decision-making. The proposed framework demonstrates competitive performance against established baselines and offers a principled approach to safe, interpretable prognostics in industrial settings.

Woody Breast (WB) and Spaghetti Meat (SM) myopathies significantly impact poultry meat quality, yet current detection methods rely either on subjective manual evaluation or costly laboratory-grade imaging systems. We address the problem of low-cost, non-destructive multi-class myopathy classification using consumer smartphones. MyoVision is introduced as a mobile transillumination imaging framework in which 14-bit RAW images are captured and structural texture descriptors indicative of internal tissue abnormalities are extracted. To classify three categories (Normal, Woody Breast, Spaghetti Meat), we propose a NEATBoost-Attention Ensemble model, which is a neuroevolution-optimized weighted fusion of LightGBM and attention-based MLP models. Hyperparameters are automatically discovered using NeuroEvolution of Augmenting Topologies (NEAT), eliminating manual tuning and enabling architecture diversity for small tabular datasets. On a dataset of 336 fillets collected from a commercial processing facility, our method achieves 82.4% test accuracy (F1 = 0.83), outperforming conventional machine learning and deep learning baselines and matching performance reported by hyperspectral imaging systems costing orders of magnitude more. Beyond classification performance, MyoVision establishes a reproducible mobile RGB-D acquisition pipeline for multimodal meat quality research, demonstrating that consumer-grade imaging can support scalable internal tissue assessment.

Accurate Global Horizontal Irradiance (GHI) forecasting is critical for grid stability, particularly in arid regions characterized by rapid aerosol fluctuations. While recent trends favor computationally expensive Transformer-based architectures, this paper challenges the prevailing "complexity-first" paradigm. We propose a lightweight, Physics-Informed Hybrid CNN-BiLSTM framework that prioritizes domain knowledge over architectural depth. The model integrates a Convolutional Neural Network (CNN) for spatial feature extraction with a Bi-Directional LSTM for capturing temporal dependencies. Unlike standard data-driven approaches, our model is explicitly guided by a vector of 15 engineered features including Clear-Sky indices and Solar Zenith Angle - rather than relying solely on raw historical data. Hyperparameters are rigorously tuned using Bayesian Optimization to ensure global optimality. Experimental validation using NASA POWER data in Sudan demonstrates that our physics-guided approach achieves a Root Mean Square Error (RMSE) of 19.53 W/m^2, significantly outperforming complex attention-based baselines (RMSE 30.64 W/m^2). These results confirm a "Complexity Paradox": in high-noise meteorological tasks, explicit physical constraints offer a more efficient and accurate alternative to self-attention mechanisms. The findings advocate for a shift towards hybrid, physics-aware AI for real-time renewable energy management.

Traffic forecasting requires modeling complex temporal dynamics and long-range spatial dependencies over large sensor networks. Existing methods typically face a trade-off between expressiveness and efficiency: Transformer-based models capture global dependencies well but suffer from quadratic complexity, while recent selective state-space models are computationally efficient yet less effective at modeling spatial interactions in graph-structured traffic data. We propose FAST, a unified framework that combines attention and state-space modeling for scalable spatiotemporal traffic forecasting. FAST adopts a Temporal-Spatial-Temporal architecture, where temporal attention modules capture both short- and long-term temporal patterns, and a Mamba-based spatial module models long-range inter-sensor dependencies with linear complexity. To better represent heterogeneous traffic contexts, FAST further introduces a learnable multi-source spatiotemporal embedding that integrates historical traffic flow, temporal context, and node-level information, together with a multi-level skip prediction mechanism for hierarchical feature fusion. Experiments on PeMS04, PeMS07, and PeMS08 show that FAST consistently outperforms strong baselines from Transformer-, GNN-, attention-, and Mamba-based families. In particular, FAST achieves the best MAE and RMSE on all three benchmarks, with up to 4.3\% lower RMSE and 2.8\% lower MAE than the strongest baseline, demonstrating a favorable balance between accuracy, scalability, and generalization.

Long-form visual storytelling requires maintaining continuity across shots, including consistent characters, stable environments, and smooth scene transitions. While existing generative models can produce strong individual frames, they fail to preserve such continuity, leading to appearance changes, inconsistent backgrounds, and abrupt scene shifts. We introduce CANVAS (Continuity-Aware Narratives via Visual Agentic Storyboarding), a multi-agent framework that explicitly plans visual continuity in multi-shot narratives. CANVAS enforces coherence through character continuity, persistent background anchors, and location-aware scene planning for smooth transitions within the same setting We evaluate CANVAS on two storyboard generation benchmarks ST-BENCH and ViStoryBench and introduce a new challenging benchmark HardContinuityBench for long-range narrative consistency. CANVAS consistently outperforms the best-performing baseline, improving background continuity by 21.6%, character consistency by 9.6% and props consistency by 7.6%.

Human-object interaction (HOI) detection aims to detect interactions between humans and objects in images. While recent advances have improved performance on existing benchmarks, their evaluations mainly focus on overall prediction accuracy and provide limited insight into the underlying causes of model failures. In particular, modern models often struggle in complex scenes involving multiple people and rare interaction combinations. In this work, we present a study to better understand the failure modes of two-stage HOI models, which form the basis of many current HOI detection approaches. Rather than constructing a large-scale benchmark, we instead decompose HOI detection into multiple interpretable perspectives and analyze model behavior across these dimensions to study different types of failure patterns. We curate a subset of images from an existing HOI dataset organized by human-object-interaction configurations (e.g., multi-person interactions and object sharing), and analyze model behavior under these configurations to examine different failure modes. This design allows us to analyze how these HOI models behave under different scene compositions and why their predictions fail. Importantly, high overall benchmark performance does not necessarily reflect robust visual reasoning about human-object relationships. We hope that this study can provide useful insights into the limitations of HOI models and offer observations for future research in this area.

Deploying Large Language Models (LLMs) on edge devices faces severe computational and memory constraints, limiting real-time processing and on-device intelligence. Hybrid architectures combining Structured State Space Models (SSMs) with transformer-based LLMs offer a balance of efficiency and performance. Aggressive quantization can drastically cut model size and speed up inference, but its uneven effects on different components require careful management. In this work, we propose a lightweight, backpropagation-free, surrogate-based sensitivity analysis framework to identify hybrid SSM-Transformer components most susceptible to quantization-induced degradation. Relying solely on forward-pass metrics, our method avoids expensive gradient computations and retraining, making it suitable for situations where access to in-domain data is limited due to proprietary restrictions or privacy constraints. We also provide a formal analysis showing that the Kullback-Leibler (KL) divergence metric better captures quantization sensitivity for Language modeling tasks than widely adopted alternatives such as mean squared error (MSE) and signal-to-quantization-noise ratio (SQNR). Through extensive experiments on SSM and hybrid architectures, our ablation studies confirm that KL-based rankings align with observed performance drops and outperform alternative metrics. This framework enables the practical deployment of advanced hybrid models on resource-constrained edge devices with minimal accuracy loss. We further validate our approach with real-world on-device profiling on Intel Lunar Lake hardware, demonstrating that KL-guided mixed-precision achieves near-FP16 perplexity with model sizes and throughput competitive with Uniform INT4 on both CPU and GPU execution modes. Code is available at https://github.com/jasonkongie/kl-ssm-quant.

Privacy concerns in LLMs have led to the rapidly growing need to enforce a data's "right to be forgotten". Machine unlearning addresses precisely this task, namely the removal of the influence of some specific data, i.e., the forget set, from a trained model. The gold standard for unlearning is to produce the model that would have been learned on only the rest of the training data, i.e., the retain set. Most existing unlearning methods rely on direct access to the retained data, which may not be practical due to privacy or cost constraints. We propose WIN-U, a retained-data free unlearning framework that requires only second order information for the originally trained model on the full data. The unlearning is performed using a single Newton-style step. Using the Woodbury matrix identity and a generalized Gauss-Newton approximation for the forget set curvature, the WIN-U update recovers the closed-form linear solution and serves as a local second-order approximation to the gold-standard retraining optimum. Extensive experiments on various vision and language benchmarks demonstrate that WIN-U achieves SOTA performance in terms of unlearning efficacy and utility preservation, while being more robust against relearning attacks compared to existing methods. Importantly, WIN-U does not require access to the retained data.

Token compression is crucial for mitigating the quadratic complexity of self-attention mechanisms in Vision Transformers (ViTs), which often involve numerous input tokens. Existing methods, such as ToMe, rely on GPU-inefficient operations (e.g., sorting, scattered writes), introducing overheads that limit their effectiveness. We introduce MaMe, a training-free, differentiable token merging method based entirely on matrix operations, which is GPU-friendly to accelerate ViTs. Additionally, we present MaRe, its inverse operation, for token restoration, forming a MaMe+MaRe pipeline for image synthesis. When applied to pre-trained models, MaMe doubles ViT-B throughput with a 2% accuracy drop. Notably, fine-tuning the last layer with MaMe boosts ViT-B accuracy by 1.0% at 1.1x speed. In SigLIP2-B@512 zero-shot classification, MaMe provides 1.3x acceleration with negligible performance degradation. In video tasks, MaMe accelerates VideoMAE-L by 48.5% on Kinetics-400 with only a 0.84% accuracy loss. Furthermore, MaMe achieves simultaneous improvements in both performance and speed on some tasks. In image synthesis, the MaMe+MaRe pipeline enhances quality while reducing Stable Diffusion v2.1 generation latency by 31%. Collectively, these results demonstrate MaMe's and MaRe's effectiveness in accelerating vision models. The code is available at https://github.com/cominder/mame}{https://github.com/cominder/mame.

Text-driven motion editing and intra-structural retargeting, where source and target share topology but may differ in bone lengths, are traditionally handled by fragmented pipelines with incompatible inputs and representations: editing relies on specialized generative steering, while retargeting is deferred to geometric post-processing. We present a unifying perspective where both tasks are cast as instances of conditional transport within a single generative framework. By leveraging recent advances in flow matching, we demonstrate that editing and retargeting are fundamentally the same generative task, distinguished only by which conditioning signal, semantic or structural, is modulated during inference. We implement this vision via a rectified-flow motion model jointly conditioned on text prompts and target skeletal structures. Our architecture extends a DiT-style transformer with per-joint tokenization and explicit joint self-attention to strictly enforce kinematic dependencies, while a multi-condition classifier-free guidance strategy balances text adherence with skeletal conformity. Experiments on SnapMoGen and a multi-character Mixamo subset show that a single trained model supports text-to-motion generation, zero-shot editing, and zero-shot intra-structural retargeting. This unified approach simplifies deployment and improves structural consistency compared to task-specific baselines.

Existing Vision Mamba-based RGB-Event(RGBE) tracking methods suffer from using static state transition matrices, which fail to adapt to variations in event sparsity. This rigidity leads to imbalanced modeling-underfitting sparse event streams and overfitting dense ones-thus degrading cross-modal fusion robustness. To address these limitations, we propose MambaTrack, a multimodal and efficient tracking framework built upon a Dynamic State Space Model(DSSM). Our contributions are twofold. First, we introduce an event-adaptive state transition mechanism that dynamically modulates the state transition matrix based on event stream density. A learnable scalar governs the state evolution rate, enabling differentiated modeling of sparse and dense event flows. Second, we develop a Gated Projection Fusion(GPF) module for robust cross-modal integration. This module projects RGB features into the event feature space and generates adaptive gates from event density and RGB confidence scores. These gates precisely control the fusion intensity, suppressing noise while preserving complementary information. Experiments show that MambaTrack achieves state-of-the-art performance on the FE108 and FELT datasets. Its lightweight design suggests potential for real-time embedded deployment.

Motivated by the underspecified, multi-hop nature of search queries and the multimodal, heterogeneous, and often conflicting nature of real-world web results, we introduce MERRIN (Multimodal Evidence Retrieval and Reasoning in Noisy Web Environments), a human-annotated benchmark for evaluating search-augmented agents. MERRIN measures AI agents' ability to identify relevant modalities, retrieve multimodal evidence, and perform multi-hop reasoning over noisy web sources. It differs from prior work in three important aspects: (1) using natural language queries without explicit modality cues, (2) incorporating underexplored modalities such as video and audio, and (3) requiring the retrieval of complex, often noisy or conflicting multimodal evidence during web search. We evaluate diverse search agents powered by ten models, including strong closed-source models (e.g., GPT-5.4-mini, Gemini 3/3.1 Flash/Pro) and open-weight models (Qwen3-4B/30B/235B), across three search settings (no search, native search, and agentic search). Our results show that MERRIN is highly challenging: the average accuracy across all agents is 22.3%, with the best-performing agent reaching only 40.1%. We further observe that while stronger agents like Gemini Deep Research achieve higher performance, gains are modest due to over-exploration; they take more steps and use more tools, but are often distracted by conflicting or partially relevant web content, leading to incorrect answers. Compared to humans, these agents consume more resources yet achieve lower accuracy, largely due to inefficient source selection and an overreliance on text modalities. These findings highlight the need for search agents capable of robust search and reasoning across diverse modalities in noisy web environments, making MERRIN a valuable testbed for evaluating such capabilities.

As Large Language Models (LLMs) are increasingly deployed in mission-critical software systems, detecting hallucinations and ``faked truthfulness'' has become a paramount engineering challenge. Current reliability architectures rely heavily on post-generation, black-box mechanisms, such as Retrieval-Augmented Generation (RAG) cross-checking or LLM-as-a-judge evaluators. These extrinsic methods introduce unacceptable latency, high computational overhead, and reliance on secondary external API calls, frequently violating standard software engineering Service Level Agreements (SLAs). In this paper, we propose the Cognitive Circuit Breaker, a novel systems engineering framework that provides intrinsic reliability monitoring with minimal latency overhead. By extracting hidden states during a model's forward pass, we calculate the ``Cognitive Dissonance Delta'' -- the mathematical gap between an LLM's outward semantic confidence (softmax probabilities) and its internal latent certainty (derived via linear probes). We demonstrate statistically significant detection of cognitive dissonance, highlight architecture-dependent Out-of-Distribution (OOD) generalization, and show that this framework adds negligible computational overhead to the active inference pipeline.

Advances in radiance fields have enabled photorealistic novel view synthesis. In several domains, large-scale real-world datasets have been developed to support comprehensive benchmarking and to facilitate progress beyond scene-specific reconstruction. However, for distractor-free radiance fields, a large-scale dataset with clean and cluttered images per scene remains lacking, limiting the development. To address this gap, we introduce DF3DV-1K, a large-scale real-world dataset comprising 1,048 scenes, each providing clean and cluttered image sets for benchmarking. In total, the dataset contains 89,924 images captured using consumer cameras to mimic casual capture, spanning 128 distractor types and 161 scene themes across indoor and outdoor environments. A curated subset of 41 scenes, DF3DV-41, is systematically designed to evaluate the robustness of distractor-free radiance field methods under challenging scenarios. Using DF3DV-1K, we benchmark nine recent distractor-free radiance field methods and 3D Gaussian Splatting, identifying the most robust methods and the most challenging scenarios. Beyond benchmarking, we demonstrate an application of DF3DV-1K by fine-tuning a diffusion-based 2D enhancer to improve radiance field methods, achieving average improvements of 0.96 dB PSNR and 0.057 LPIPS on the held-out set (e.g., DF3DV-41) and the On-the-go dataset. We hope DF3DV-1K facilitates the development of distractor-free vision and promotes progress beyond scene-specific approaches.

Majority-vote ensembles achieve variance reduction by averaging over diverse, approximately independent base learners. When training data exhibits Markov dependence, as in time-series forecasting, reinforcement learning (RL) replay buffers, and spatial grids, this classical guarantee degrades in ways that existing theory does not fully quantify. We provide a minimax characterization of this phenomenon for discrete classification in a fixed-dimensional Markov setting, together with an adaptive algorithm that matches the rate on a graph-regular subclass. We first establish an information-theoretic lower bound for stationary, reversible, geometrically ergodic chains in fixed ambient dimension, showing that no measurable estimator can achieve excess classification risk better than $Ω(\sqrt{\Tmix/n})$. We then prove that, on the AR(1) witness subclass underlying the lower-bound construction, dependence-agnostic uniform bagging is provably suboptimal with excess risk bounded below by $Ω(\Tmix/\sqrt{n})$, exhibiting a $\sqrt{\Tmix}$ algorithmic gap. Finally, we propose \emph{adaptive spectral routing}, which partitions the training data via the empirical Fiedler eigenvector of a dependency graph and achieves the minimax rate $\mathcal{O}(\sqrt{\Tmix/n})$ up to a lower-order geometric cut term on a graph-regular subclass, without knowledge of $\Tmix$. Experiments on synthetic Markov chains, 2D spatial grids, the 128-dataset UCR archive, and Atari DQN ensembles validate the theoretical predictions. Consequences for deep RL target variance, scalability via Nyström approximation, and bounded non-stationarity are developed as supporting material in the appendix.

Diffusion language models (DLMs) have emerged as a promising paradigm for large language models (LLMs), yet the non-deterministic behavior of DLMs remains poorly understood. The existing non-determinism evaluations for LLMs predominantly rely on dataset-level metrics under fixed inference configurations, providing limited insight into how model behavior varies across runs and evaluation conditions. In this work, we show that dataset-level metrics systematically attenuate non-determinism in diffusion language models by aggregating sample-level prediction quality across different runs. As a result, configurations with similar aggregate performance can exhibit substantially different behaviors on individual inputs, leaving fine-grained instability and distinct error patterns uncharacterized. To address this limitation, we conduct a fine-grained evaluation of non-determinism based on sample-level prediction differences across a range of model-related factors-including guidance scale, diffusion steps, and Monte Carlo sampling-as well as system-related factors such as batch size, hardware, and numerical precision. Our analysis reveals that non-determinism in DLMs is pervasive and structured, with code generation exhibiting markedly higher sensitivity to factor-level choices than question answering. To attribute sources of non-determinism evaluation, we introduce Factor Variance Attribution (FVA), a cross-factor analysis metric that decomposes observed non-determinism into variance attributable to different evaluation factor settings. Our findings highlight the need for fine-grained, factor-aware evaluation to enable reliable non-determinism assessment of diffusion language models.

We study the problem of estimating the effect function for a continuous treatment, which maps each treatment value to a population-averaged outcome. A central challenge in this setting is confounding: treatment assignment often depends on covariates, creating selection bias that makes direct regression of the response on treatment unreliable. To address this issue, we propose a two-stage kernel ridge regression method. In the first stage, we learn a model for the response as a function of both treatment and covariates; in the second stage, we use this model to construct pseudo-outcomes that correct for distribution shift, and then fit a second model to estimate the treatment effect. Although the response varies with both treatment and covariates, the induced effect function obtained by averaging over covariates is typically much simpler, and our estimator adapts to this structure. Furthermore, we introduce a fully data-driven model selection procedure that achieves provable adaptivity to both the unknown degree of overlap and the regularity (eigenvalue decay) of the underlying kernel.

While Aspect-based Sentiment Analysis (ABSA) systems have achieved high accuracy in identifying sentiment polarities, they often operate as "black boxes," lacking the explicit reasoning capabilities characteristic of human affective cognition. Humans do not merely categorize sentiment; they construct causal explanations for their judgments. To bridge this gap, we propose ABSA-R1, a large language model framework designed to mimic this ``reason-before-predict" cognitive process. By leveraging reinforcement learning (RL), ABSA-R1 learns to articulate the why behind the what, generating natural language justifications that ground its sentiment predictions. We introduce a Cognition-Aligned Reward Model (formerly sentiment-aware reward model) that enforces consistency between the generated reasoning path and the final emotional label. Furthermore, inspired by metacognitive monitoring, we implement a performance-driven rejection sampling strategy that selectively targets hard cases where the model's internal reasoning is uncertain or inconsistent. Experimental results on four benchmarks demonstrate that equipping models with this explicit reasoning capability not only enhances interpretability but also yields superior performance in sentiment classification and triplet extraction compared to non-reasoning baselines.

Large Reasoning Models (LRMs) have recently demonstrated significant improvements in complex reasoning. While quantifying generation uncertainty in LRMs is crucial, traditional methods are often insufficient because they do not provide finite-sample guarantees for reasoning-answer generation. Conformal prediction (CP) stands out as a distribution-free and model-agnostic methodology that constructs statistically rigorous uncertainty sets. However, existing CP methods ignore the logical connection between the reasoning trace and the final answer. Additionally, prior studies fail to interpret the origins of uncertainty coverage for LRMs as they typically overlook the specific training factors driving valid reasoning. Notably, it is challenging to disentangle reasoning quality from answer correctness when quantifying uncertainty, while simultaneously establishing theoretical guarantees for computationally efficient explanation methods. To address these challenges, we first propose a novel methodology that quantifies uncertainty in the reasoning-answer structure with statistical guarantees. Subsequently, we develop a unified example-to-step explanation framework using Shapley values that identifies a provably sufficient subset of training examples and their key reasoning steps to preserve the guarantees. We also provide theoretical analyses of our proposed methods. Extensive experiments on challenging reasoning datasets verify the effectiveness of the proposed methods.

Davis, Drusvyatskiy, and Jiang showed that gradient descent with an adaptive stepsize converges locally at a nearly-linear rate for smooth functions that grow at least quartically away from their minimizers. The argument is intricate, relying on monitoring the performance of the algorithm relative to a certain manifold of slow growth -- called the ravine. In this work, we provide a direct Lyapunov-based argument that bypasses these difficulties when the objective is in addition convex and a has a unique minimizer. As a byproduct of the argument, we obtain a more adaptive variant than the original algorithm with encouraging numerical performance.

Tabular data remains prevalent in high-stakes domains such as healthcare and finance, where predictive models are expected to provide both high accuracy and faithful, human-understandable reasoning. While symbolic models offer verifiable logic, they lack semantic expressiveness. Meanwhile, general-purpose LLMs often require specialized fine-tuning to master domain-specific tabular reasoning. To address the dual challenges of scalable data curation and reasoning consistency, we propose ReSS, a systematic framework that bridges symbolic and neural reasoning models. ReSS leverages a decision-tree model to extract instance-level decision paths as symbolic scaffolds. These scaffolds, alongside input features and labels, guide an LLM to generate grounded natural-language reasoning that strictly adheres to the underlying decision logic. The resulting high-quality dataset is used to fine-tune a pretrained LLM into a specialized tabular reasoning model, further enhanced by a scaffold-invariant data augmentation strategy to improve generalization and explainability. To rigorously assess faithfulness, we introduce quantitative metrics including hallucination rate, explanation necessity, and explanation sufficiency. Experimental results on medical and financial benchmarks demonstrate that ReSS-trained models improve traditional decision trees and standard fine-tuning approaches up to $10\%$ while producing faithful and consistent reasoning

Linear probes can detect when language models produce outputs they "know" are wrong, a capability relevant to both deception and reward hacking. However, single-layer probes are fragile: the best layer varies across models and tasks, and probes fail entirely on some deception types. We show that combining probes from multiple layers into an ensemble recovers strong performance even where single-layer probes fail, improving AUROC by +29% on Insider Trading and +78% on Harm-Pressure Knowledge. Across 12 models (0.5B--176B parameters), we find probe accuracy improves with scale: ~5% AUROC per 10x parameters (R=0.81). Geometrically, deception directions rotate gradually across layers rather than appearing at one location, explaining both why single-layer probes are brittle and why multi-layer ensembles succeed.

Open-pit mine scheduling is a complex real world optimization problem that involves uncertain economic values and dynamically changing resource capacities. Evolutionary algorithms are particularly effective in these scenarios, as they can easily adapt to uncertain and changing environments. However, uncertainty and dynamic changes are often studied in isolation in real-world problems. In this paper, we study a dynamic chance-constrained open-pit mine scheduling problem in which block economic values are stochastic and mining and processing capacities vary over time. We adopt a bi-objective evolutionary formulation that simultaneously maximizes expected discounted profit and minimizes its standard deviation. To address dynamic changes, we propose a diversity-based change response mechanism that repairs a subset of infeasible solutions and introduces additional feasible solutions whenever a change is detected. We evaluate the effectiveness of this mechanism across four multi-objective evolutionary algorithms and compare it with a baseline re-evaluation-based change-response strategy. Experimental results on six mining instances demonstrate that the proposed approach consistently outperforms the baseline methods across different uncertainty levels and change frequencies.

Conversational generative artificial intelligence agents (or genAI chatbots) could benefit youth mental health, yet young people's perspectives remain underexplored. We examined the Mental health Intelligence Agent (Mia), a genAI chatbot originally designed for professionals in Australian youth services. Following co-design, 32 young people participated in online workshops exploring their perceptions of genAI chatbots in youth mental health and to develop recommendations for reconceptualising Mia for consumers and integrating it into services. Four themes were developed: (1) Humanising AI without dehumanising care, (2) I need to know what's under the hood, (3) Right tool, right place, right time?, and (4) Making it mine on safe ground. This study offers insights into young people's attitudes, needs, and requirements regarding genAI chatbots in youth mental health, with key implications for service integration. Additionally, by co-designing system requirements, this work informs the ethics, design, development, implementation, and governance of genAI chatbots in youth mental health contexts.

Large Language Models (LLMs) are increasingly described as possessing strong reasoning capabilities, supported by high performance on mathematical, logical, and planning benchmarks. However, most existing evaluations rely on aggregate accuracy over fixed datasets, obscuring how reasoning behavior evolves as task complexity increases. In this work, we introduce a controlled benchmarking framework to systematically evaluate the robustness of reasoning in Large Reasoning Models (LRMs) under progressively increasing problem complexity. We construct a suite of nine classical reasoning tasks: Boolean Satisfiability, Cryptarithmetic, Graph Coloring, River Crossing, Tower of Hanoi, Water Jug, Checker Jumping, Sudoku, and Rubik's Cube, each parameterized to precisely control complexity while preserving underlying semantics. Using deterministic validators, we evaluate multiple open and proprietary LRMs across low, intermediate, and high complexity regimes, ensuring that only fully valid solutions are accepted. Our results reveal a consistent phase transition like behavior: models achieve high accuracy at low complexity but degrade sharply beyond task specific complexity thresholds. We formalize this phenomenon as reasoning collapse. Across tasks, we observe substantial accuracy declines, often exceeding 50%, accompanied by inconsistent reasoning traces, constraint violations, loss of state tracking, and confidently incorrect outputs. Increased reasoning length does not reliably improve correctness, and gains in one problem family do not generalize to others. These findings highlight the need for evaluation methodologies that move beyond static benchmarks and explicitly measure reasoning robustness under controlled complexity.

Surrogate modeling of body-driven fluid flows where immersed moving boundaries couple structural dynamics to chaotic, unsteady fluid phenomena remains a fundamental challenge for both computational physics and machine learning. We present AeTHERON, a heterogeneous graph neural operator whose architecture directly mirrors the structure of the sharp-interface immersed boundary method (IBM): a dual-graph representation separating fluid and structural domains, coupled through sparse cross-attention that reflects the compact support of IBM interpolation stencils. This physics-informed inductive bias enables AeTHERON to learn nonlinear fluid-structure coupling in a shared high-dimensional latent space, with continuous sinusoidal time embeddings providing temporal generalization across lead times. We evaluate AeTHERON on direct numerical simulations of a flapping flexible caudal fin, a canonical FSI benchmark featuring leading-edge vortex formation, large membrane deformation, and chaotic wake shedding across a 4x5 parameter grid of membrane thickness (h* = 0.01-0.04) and Strouhal number (St = 0.30-0.50). As a proof-of-concept, we train on the first 150 timesteps of a representative case using a 70/30 train/validation split and evaluate on the fully unseen extrapolation window t=150-200. AeTHERON captures large-scale vortex topology and wake structure with qualitative fidelity, achieving a mean extrapolation MAE of 0.168 without retraining, with error peaking near flapping half-cycle transitions where flow reorganization is most rapid -- a physically interpretable pattern consistent with the nonlinear fluid-membrane coupling. Inference requires milliseconds per timestep on a single GPU versus hours for equivalent DNS computation. This is a continuously developing preprint; results and figures will be updated in subsequent versions.

Fine-tuning large language models (LLMs) aims to adapt pre-trained models to specific tasks using relatively small and domain-specific datasets. Among Parameter-Efficient Fine-Tuning (PEFT) methods, Low-Rank Adaptation (LoRA) stands out by matching the performance of full fine-tuning while avoiding additional inference latency. In this paper, we propose a novel PEFT method that incorporates the TLoRA+ optimizer into the weight matrices of pre-trained models. The proposed approach not only preserves the efficiency of low-rank adaptation but also further enhances performance without significantly increasing computational cost. We conduct experiments on the GLUE benchmark across diverse model architectures. Numerical experiments consistently demonstrate the effectiveness and robustness of our proposed method.

Radiotherapy-induced normal tissue injury is a clinically important complication, and accurate segmentation of injury regions from medical images could facilitate disease assessment, treatment planning, and longitudinal monitoring. However, automatic segmentation of these lesions remains largely unexplored because of limited voxel-level annotations and substantial heterogeneity across injury types, lesion size, and imaging modality. To address this gap, we curate a dedicated head-and-neck radiotherapy-induced normal tissue injury dataset covering three manifestations: osteoradionecrosis (ORN), cerebral edema (CE), and cerebral radiation necrosis (CRN). We further propose a 3D SAM-based progressive prompting framework for multi-task segmentation in limited-data settings. The framework progressively incorporates three complementary prompts: text prompts for task-aware adaptation, dose-guided box prompts for coarse localization, and click prompts for iterative refinement. A small-target focus loss is introduced to improve local prediction and boundary delineation for small and sparse lesions. Experiments on ORN, CE, and CRN demonstrate that the proposed method achieves reliable segmentation performance across diverse injury types and outperforms state-of-the-art methods.

Accurate modeling of robot dynamics is essential for model-based control, yet remains challenging under distributional shifts and real-time constraints. In this work, we formulate system identification as an in-context meta-learning problem and compare deterministic and generative sequence models for forward dynamics prediction. We take a Transformer-based meta-model, as a strong deterministic baseline, and introduce to this setting two complementary diffusion-based approaches: (i) inpainting diffusion (Diffuser), which learns the joint input-observation distribution, and (ii) conditioned diffusion models (CNN and Transformer), which generate future observations conditioned on control inputs. Through large-scale randomized simulations, we analyze performance across in-distribution and out-of-distribution regimes, as well as computational trade-offs relevant for control. We show that diffusion models significantly improve robustness under distribution shift, with inpainting diffusion achieving the best performance in our experiments. Finally, we demonstrate that warm-started sampling enables diffusion models to operate within real-time constraints, making them viable for control applications. These results highlight generative meta-models as a promising direction for robust system identification in robotics.

Biosignals exhibit substantial cross-subject and cross-session variability, inducing severe domain shifts that degrade post-deployment performance for small, edge-oriented AI models. On-device adaptation is therefore essential to both preserve user privacy and ensure system reliability. However, existing sub-100 mW MCU-based wearable platforms can only support shallow or sparse adaptation schemes due to the prohibitive memory footprint and computational cost of full backpropagation (BP). In this paper, we propose BioTrain, a framework enabling full-network fine-tuning of state-of-the-art biosignal models under milliwatt-scale power and sub-megabyte memory constraints. We validate BioTrain using both offline and on-device benchmarks on EEG and EOG datasets, covering Day-1 new-subject calibration and longitudinal adaptation to signal drift. Experimental results show that full-network fine-tuning achieves accuracy improvements of up to 35% over non-adapted baselines and outperforms last-layer updates by approximately 7% during new-subject calibration. On the GAP9 MCU platform, BioTrain enables efficient on-device training throughput of 17 samples/s for EEG and 85 samples/s for EOG models within a power envelope below 50 mW. In addition, BioTrain's efficient memory allocator and network topology optimization enable the use of a large batch size, reducing peak memory usage. For fully on-chip BP on GAP9, BioTrain reduces the memory footprint by 8.1x, from 5.4 MB to 0.67 MB, compared to conventional full-network fine-tuning using batch normalization with batch size 8.

Mechanisms for continued self-improvement of language models without external supervision remain an open challenge. We propose Peer-Predictive Self-Training (PST), a label-free fine-tuning framework in which multiple language models improve collaboratively by leveraging a cross-model aggregated response as an internal training signal. Given a prompt question, the models generate responses sequentially; the final aggregated answer, often more reliable than individual responses in practice, serves as an internal target for learning. We measure how informative each intermediate response is about the aggregate using pointwise mutual information (PMI), and use this signal to scale self-training updates. Responses already aligned with the aggregate are updated less, while less informative or misaligned responses are updated more. On mathematical reasoning benchmarks (SimulEq, Math500, and MultiArith), PST improves exact-match accuracy by 2.2 to 4.3 percentage points across Gemma-2-2B, LLaMA-3.2-1B, and Qwen-2.5-1.5B, and reduces the average generator-verifier gap (GV-Gap) by 26 to 40 percent, while requiring no external supervision or teacher-student hierarchy and relying solely on cross-model interactions. These results suggest that cross-model generations and peer-predictive feedback can serve as an effective approach for self-supervised training.

The discovery of inorganic crystal structures with targeted properties is a significant challenge in materials science. Generative models, especially state-of-the-art diffusion models, offer the promise of modeling complex data distributions and proposing novel, realistic samples. However, current generative AI models still struggle to produce diverse, original, and reliable structures of experimentally achievable materials suitable for high-stakes applications. In this work, we propose a generative machine learning framework based on diffusion models with adaptive constraint guidance, which enables the incorporation of user-defined physical and chemical constraints during the generation process. This approach is designed to be practical and interpretable for human experts, allowing transparent decision-making and expert-driven exploration. To ensure the robustness and validity of the generated candidates, we introduce a multi-step validation pipeline that combines graph neural network estimators trained to achieve DFT-level accuracy and convex hull analysis for assessing thermodynamic stability. Our approach has been tested and validated on several classical examples of inorganic families of compounds, as case studies. As a consequence, these preliminary results demonstrate our framework's ability to generate thermodynamically plausible crystal structures that satisfy targeted geometric constraints across diverse inorganic chemical systems.

Communication in Large Language Model (LLM)-based multi-agent systems is moving beyond discrete tokens to preserve richer context. Recent work such as LatentMAS enables agents to exchange latent messages through full key-value (KV) caches. However, full KV relay incurs high memory and communication cost. We adapt eviction-style KV compression to this setting and introduce Orthogonal Backfill (OBF) to mitigate information loss from hard eviction. OBF injects a low-rank orthogonal residual from discarded KV states into the retained KV states. We evaluate proposed method against full KV relay on nine standard benchmarks spanning mathematical reasoning, coding, and knowledge-intensive QA. It achieves performance comparable to full KV relay while reducing communication cost by 79.8%--89.4%. OBF further improves the performance and achieves the best results on 7 of the 9 benchmarks. This suggests that more information does not necessarily lead to better communication; preserving the most useful information matters more. Our codebase is publicly available on https://github.com/markli404/When-Less-Latent-Leads-to-Better-Relay.

We introduce CONCORD, a privacy-aware asynchronous assistant-to-assistant (A2A) framework that leverages collaboration between proactive speech-based AI. As agents evolve from reactive to always-listening assistants, they face a core privacy risk (of capturing non-consenting speakers), which makes their social deployment a challenge. To overcome this, we implement CONCORD, which enforces owner-only speech capture via real-time speaker verification, producing a one-sided transcript that incurs missing context but preserves privacy. We demonstrate that CONCORD can safely recover necessary context through (1) spatio-temporal context resolution, (2) information gap detection, and (3) minimal A2A queries governed by a relationship-aware disclosure. Instead of hallucination-prone inferring, CONCORD treats context recovery as a negotiated safe exchange between assistants. Across a multi-domain dialogue dataset, CONCORD achieves 91.4% recall in gap detection, 96% relationship classification accuracy, and 97% true negative rate in privacy-sensitive disclosure decisions. By reframing always-listening AI as a coordination problem between privacy-preserving agents, CONCORD offers a practical path toward socially deployable proactive conversational agents.

Language-model agent systems commonly rely on reactive prompting, in which a single instruction guides the model through an open-ended sequence of reasoning and tool-use steps, leaving control flow and intermediate state implicit and making agent behavior potentially difficult to control. Orchestration frameworks such as LangGraph, DSPy, and CrewAI impose greater structure through explicit workflow definitions, but tightly couple workflow logic with Python, making agents difficult to maintain and modify. In this paper, we introduce AgentSPEX, an Agent SPecification and EXecution Language for specifying LLM-agent workflows with explicit control flow and modular structure, along with a customizable agent harness. AgentSPEX supports typed steps, branching and loops, parallel execution, reusable submodules, and explicit state management, and these workflows execute within an agent harness that provides tool access, a sandboxed virtual environment, and support for checkpointing, verification, and logging. Furthermore, we provide a visual editor with synchronized graph and workflow views for authoring and inspection. We include ready-to-use agents for deep research and scientific research, and we evaluate AgentSPEX on 7 benchmarks. Finally, we show through a user study that AgentSPEX provides a more interpretable and accessible workflow-authoring paradigm than a popular existing agent framework.

Identifying meaningful feature interactions is a central challenge in building accurate and interpretable models for tabular data. Generalized additive models (GAMs) have shown great success at modeling tabular data, but often rely on heuristic procedures to select interactions, potentially missing higher-order or context-dependent effects. To meet this challenge, we propose TabDistill, a method that leverages tabular foundation models and post-hoc distillation methods. Our key intuition is that tabular foundation models implicitly learn rich, adaptive feature dependencies through large-scale representation learning. Given a dataset, TabDistill first fits a tabular foundation model to the dataset, and then applies a post-hoc interaction attribution method to extract salient feature interactions from it. We evaluate these interactions by then using them as terms in a GAM. Across tasks, we find that interactions identified by TabDistill lead to consistent improvements in downstream GAMs' predictive performance. Our results suggest that tabular foundation models can serve as effective, data-driven guides for interaction discovery, bridging high-capacity models and interpretable additive frameworks.

In electronic health record (EHR) mining, learning high-quality representations of medical concepts (e.g., standardized diagnosis, medication, and procedure codes) is fundamental for downstream clinical prediction. However, robust concept representation learning is hindered by two key challenges: (i) clinically important cross-type dependencies (e.g., diagnosis-medication and medication-procedure relations) are often missing or incomplete in existing ontology resources, limiting the ability to model complex EHR patterns; and (ii) rich clinical semantics are often missing from structured resources, and even when available as text, are difficult to integrate with KG structure for representation learning. To address these challenges, we present CoMed, an LLM-empowered graph learning framework for medical concept representation. CoMed first builds a global knowledge graph (KG) over medical codes by combining statistically reliable associations mined from EHRs with type-constrained LLM prompting to infer semantic relations. It then utilizes LLMs to enrich the KG into a text-attributed graph by generating node descriptions and edge rationales, providing semantic signals for both concepts and their relationships. Finally, CoMed jointly trains a LoRA-tuned LLaMA text encoder with a heterogeneous GNN, fusing text semantics and graph structure into unified concept embeddings. Extensive experiments on MIMIC-III and MIMIC-IV show that CoMed consistently improves prediction performance and serves as an effective plug-in concept encoder for standard EHR pipelines.

Pathology reports serve as the definitive record for breast cancer staging, yet their unstructured format impedes large-scale data curation. While Large Language Models (LLMs) offer semantic reasoning, their deployment is often limited by high computational costs and hallucination risks. This study introduces a parameter-efficient, multi-task framework for automating the extraction of Tumor-Node-Metastasis (TNM) staging, histologic grade, and biomarkers. We fine-tune a Llama-3-8B-Instruct encoder using Low-Rank Adaptation (LoRA) on a curated, expert-verified dataset of 10,677 reports. Unlike generative approaches, our architecture utilizes parallel classification heads to enforce consistent schema adherence. Experimental results demonstrate that the model achieves a Macro F1 score of 0.976, successfully resolving complex contextual ambiguities and heterogeneous reporting formats that challenge traditional extraction methods including rule-based natural language processing (NLP) pipelines, zero-shot LLMs, and single-task LLM baselines. The proposed adapter-efficient, multi-task architecture enables reliable, scalable pathology-derived cancer staging and biomarker profiling, with the potential to enhance clinical decision support and accelerate data-driven oncology research.

Modern GPU workloads, especially large language model (LLM) inference, suffer from kernel launch overheads and coarse synchronization that limit inter-kernel parallelism. Recent megakernel techniques fuse multiple operators into a single persistent kernel to eliminate launch gaps and expose inter-kernel parallelism, but struggle to handle dynamic shapes and data-dependent computation in real workloads. We present Event Tensor, a unified compiler abstraction for dynamic megakernels. Event Tensor encodes dependencies between tiled tasks, and enables first-class support for both shape and data-dependent dynamism. Built atop this abstraction, our Event Tensor Compiler (ETC) applies static and dynamic scheduling transformations to generate high-performance persistent kernels. Evaluations show that ETC achieves state-of-the-art LLM serving latency while significantly reducing system warmup overhead.

Autonomous web agents powered by large language models (LLMs) have shown promise in completing complex browser tasks, yet they still struggle with long-horizon workflows. A key bottleneck is the grounding gap in existing skill formulations: textual workflow skills provide natural language guidance but cannot be directly executed, while code-based skills are executable but opaque to the agent, offering no step-level understanding for error recovery or adaptation. We introduce WebXSkill, a framework that bridges this gap with executable skills, each pairing a parameterized action program with step-level natural language guidance, enabling both direct execution and agent-driven adaptation. WebXSkill operates in three stages: skill extraction mines reusable action subsequences from readily available synthetic agent trajectories and abstracts them into parameterized skills, skill organization indexes skills into a URL-based graph for context-aware retrieval, and skill deployment exposes two complementary modes, grounded mode for fully automated multi-step execution and guided mode where skills serve as step-by-step instructions that the agent follows with its native planning. On WebArena and WebVoyager, WebXSkill improves task success rate by up to 9.8 and 12.9 points over the baseline, respectively, demonstrating the effectiveness of executable skills for web agents. The code is publicly available at https://github.com/aiming-lab/WebXSkill.

Neural operators have emerged as fast surrogate models for physics simulations, yet they remain acutely vulnerable to adversarial perturbations, a critical liability for safety-critical digital twin deployments. We present a synergistic defense that combines active learning-based data generation with an input denoising architecture. The active learning component adaptively probes model weaknesses using differential evolution attacks, then generates targeted training data at discovered vulnerability locations while an adaptive smooth-ratio safeguard preserves baseline accuracy. The input denoising component augments the operator architecture with a learnable bottleneck that filters adversarial noise while retaining physics-relevant features. On the viscous Burgers' equation benchmark, the combined approach achieves a 2.04% combined error (1.21% baseline + 0.83% robustness), representing an 87% reduction relative to standard training (15.42% combined) and outperforming both active learning alone (3.42%) and input denoising alone (5.22%). More broadly, our results, combined with cross-architecture vulnerability analysis from prior work, suggest that optimal training data for neural operators is architecture-dependent: because different architectures concentrate sensitivity in distinct input subspaces, uniform sampling cannot adequately cover the vulnerability landscape of all models. These findings have potential implications for the deployment of neural operators in safety-critical energy systems including nuclear reactor monitoring.

High-resolution data in spatial and temporal contexts is imperative for developing climate resilient cities. Current datasets for monitoring urban parameters are developed primarily using manual inspections, embedded-sensing, remote sensing, or standard street-view imagery (RGB). These methods and datasets are often constrained respectively by poor scalability, inconsistent spatio-temporal resolutions, overhead views or low spectral information. We present a novel method and its open implementation: a multi-spectral terrestrial-view dataset that circumvents these limitations. This dataset consists of 17,718 street level multi-spectral images captured with RGB, Near-infrared, and Thermal imaging sensors on bikes, across diverse urban morphologies (village, town, small city, and big urban area) in the Netherlands. Strict emphasis is put on data calibration and quality while also providing the details of our data collection methodology (including the hardware and software details). To the best of our knowledge, Spectrascapes is the first open-access dataset of its kind. Finally, we demonstrate two downstream use-cases enabled using this dataset and provide potential research directions in the machine learning, urban planning and remote sensing domains.

Vision-Language Models demonstrate remarkable capabilities but often struggle with compositional reasoning, exhibiting vulnerabilities regarding word order and attribute binding. This limitation arises from a scarcity of informative samples needed to differentiate subtle semantic variations during contrastive pretraining. Although hard negative mining offers a promising remedy, existing methods lack explicit mechanisms to dictate which linguistic elements undergo modification. Instead of engineering generative architectures, this study establishes lexical concreteness as a fundamental determinant of negative sample efficacy. Modifying highly concrete terms generates more pronounced structural and visual discrepancies, providing a substantially stronger learning signal. Leveraging this principle, ConcretePlant is proposed to systematically isolate and manipulate perceptually grounded concepts. Analyses of the InfoNCE further reveals a severe gradient imbalance, where easily distinguishable pairs disproportionately overwhelm the optimization process and restrict the bandwidth available for nuanced learning. To resolve this degradation, the Cement loss is formulated utilizing a margin-based approach. By correlating psycholinguistic scores with sample difficulty, this objective dynamically calibrates the penalization applied to individual training pairs. Comprehensive evaluations substantiate these theoretical claims. The integrated framework, designated as Slipform, achieves state-of-the-art accuracy across diverse compositional evaluation benchmarks, general cross-modal retrieval, single and multi label linear probing.

An unsupervised framework for hyperspectral image (HSI) clustering is proposed that incorporates masked deep representation learning with diffusion-based clustering, extending the Spatially-Regularized Superpixel-based Diffusion Learning ($S^2DL$) algorithm. Initially, a denoised latent representation of the original HSI is learned via an unsupervised masked autoencoder (UMAE) model with a Vision Transformer backbone. The UMAE takes spatial context and long-range spectral correlations into account and incorporates an efficient pretraining process via masking that utilizes only a small subset of training pixels. In the next stage, the entropy rate superpixel (ERS) algorithm is used to segment the image into superpixels, and a spatially regularized diffusion graph is constructed using Euclidean and diffusion distances within the compressed latent space instead of the HSI space. The proposed algorithm, Deep Spatially-Regularized Superpixel-based Diffusion Learning ($DS^2DL$), leverages more faithful diffusion distances and subsequent diffusion graph construction that better reflect the intrinsic geometry of the underlying data manifold, improving labeling accuracy and clustering quality. Experiments on Botswana and KSC datasets demonstrate the efficacy of $DS^2DL$.

Understanding the internal activations of Vision Transformers (ViTs) is critical for building interpretable and trustworthy models. While Sparse Autoencoders (SAEs) have been used to extract human-interpretable features, they operate on individual layers and fail to capture the cross-layer computational structure of Transformers, as well as the relative significance of each layer in forming the last-layer representation. Alternatively, we introduce the adoption of Cross-Layer Transcoders (CLTs) as reliable, sparse, and depth-aware proxy models for MLP blocks in ViTs. CLTs use an encoder-decoder scheme to reconstruct each post-MLP activation from learned sparse embeddings of preceding layers, yielding a linear decomposition that transforms the final representation of ViTs from an opaque embedding into an additive, layer-resolved construction that enables faithful attribution and process-level interpretability. We train CLTs on CLIP ViT-B/32 and ViT-B/16 across CIFAR-100, COCO, and ImageNet-100. We show that CLTs achieve high reconstruction fidelity with post-MLP activations while preserving and even improving, in some cases, CLIP zero-shot classification accuracy. In terms of interpretability, we show that the cross-layer contribution scores provide faithful attribution, revealing that the final representation is concentrated in a smaller set of dominant layer-wise terms whose removal degrades performance and whose retention largely preserves it. These results showcase the significance of adopting CLTs as an alternative interpretable proxy of ViTs in the vision domain.

This paper investigates the problem of data-driven modeling of port-Hamiltonian systems while preserving their intrinsic Hamiltonian structure and stability properties. We propose a novel neural-network-based port-Hamiltonian modeling technique that relaxes the convexity constraint commonly imposed by neural network-based Hamiltonian approximations, thereby improving the expressiveness and generalization capability of the model. By removing this restriction, the proposed approach enables the use of more general non-convex Hamiltonian representations to enhance modeling flexibility and accuracy. Furthermore, the proposed method incorporates information about stable equilibria into the learning process, allowing the learned model to preserve the stability of multiple isolated equilibria rather than being restricted to a single equilibrium as in conventional methods. Two numerical experiments are conducted to validate the effectiveness of the proposed approach and demonstrate its ability to achieve more accurate structure- and stability-preserving learning of port-Hamiltonian systems compared with a baseline method.

t-SNE has gained popularity as a dimension reduction technique, especially for visualizing data. It is well-known that all dimension reduction techniques may lose important features of the data. We provide a mathematical framework for understanding this loss for t-SNE by establishing a number of results in different scenarios showing how important features of data are lost by using t-SNE.

Accurate characterization of subsurface heterogeneity is challenging but essential for applications such as reservoir pressure management, geothermal energy extraction and CO$_2$, H$_2$, and wastewater injection operations. This challenge becomes especially acute in extreme pressure events, which are rarely observed but can strongly affect operational risk. Traditional history matching and inversion techniques rely on expensive full-physics simulations, making it infeasible to handle uncertainty and extreme events at scale. Purely data-driven models often struggle to maintain physics consistency when dealing with sparse observations, complex geology, and extreme events. To overcome these limitations, we introduce a physics-informed machine learning method that embeds a differentiable subsurface flow simulator directly into neural network training. The network infers heterogeneous permeability fields from limited pressure observations, while training minimizes both permeability and pressure losses through the simulator, enforcing physical consistency. Because the simulator is used only during training, inference remains fast once the model is learned. In an initial test, the proposed method reduces the pressure inference error by half compared with a purely data-driven approach. We then extend the test over eight distinct data scenarios, and in every case, our method produces significantly lower pressure inference errors than the purely data-driven model. We also evaluate our method on extreme events, which represent high-consequence data in the tail of the sample distribution. Similar to the bulk distribution, the physics-informed model maintains higher pressure inference accuracy in the extreme event regimes. Overall, the proposed method enables rapid, physics-consistent subsurface inversion for real-time reservoir characterization and risk-aware decision-making.

We present a unified pipeline for synthesizing high-quality Quechua and Spanish speech for the Peruvian Constitution using three state-of-the-art text-to-speech (TTS) architectures: XTTS v2, F5-TTS, and DiFlow-TTS. Our models are trained on independent Spanish and Quechua speech datasets with heterogeneous sizes and recording conditions, and leverage bilingual and multilingual TTS capabilities to improve synthesis quality in both languages. By exploiting cross-lingual transfer, our framework mitigates data scarcity in Quechua while preserving naturalness in Spanish. We release trained checkpoints, inference code, and synthesized audio for each constitutional article, providing a reusable resource for speech technologies in indigenous and multilingual contexts. This work contributes to the development of inclusive TTS systems for political and legal content in low-resource settings.

Weight pruning is a common technique for compressing large neural networks. We focus on the challenging post-training one-shot setting, where a pre-trained model is compressed without any retraining. Existing one-shot pruning methods typically optimize a single objective, such as a layer-wise reconstruction loss or a second-order Taylor approximation of the training loss. We highlight that neither objective alone is consistently the most effective across architectures and sparsity levels. Motivated by this insight, we propose MOONSHOT, a general and flexible framework that extends any single-objective pruning method into a multi-objective formulation by jointly optimizing both the layer-wise reconstruction error and second-order Taylor approximation of the training loss. MOONSHOT acts as a wrapper around existing pruning algorithms. To enable this integration while maintaining scalability to billion-parameter models, we propose modeling decisions and introduce an efficient procedure for computing the inverse Hessian, preserving the efficiency of state-of-the-art one-shot pruners. When combined with state-of-the-art pruning methods on Llama-3.2 and Llama-2 models, MOONSHOT reduces C4 perplexity by up to 32.6% at 2:4 sparsity and improves zero-shot mean accuracy across seven classification benchmarks by up to 4.9 points. On Vision Transformers, it improves accuracy on ImageNet-1k by over 5 points at 70% sparsity, and on ResNet-50, it yields a 4-point gain at 90% sparsity.

Despite the widespread multilingual deployment of large language models, post-training pipelines remain predominantly English-centric, contributing to performance disparities across languages. We present a systematic, controlled study of the interplay between training language coverage, model scale, and task domain, based on 220 supervised fine-tuning runs on parallel translated multilingual data mixtures spanning mathematical reasoning and API calling tasks, with models up to 8B parameters. We find that increasing language coverage during post-training is largely beneficial across tasks and model scales, with low-resource languages benefiting the most and high-resource languages plateauing rather than degrading. Even minimal multilinguality helps: incorporating a single non-English language improves both English performance and cross-lingual generalization, making English-only post-training largely suboptimal. Moreover, at sufficient language diversity, zero-shot cross-lingual transfer can match or exceed the effects of direct language inclusion in a low-diversity setting, although gains remain limited for typologically distant, low-resource languages.

Clinical text classification requires choosing between specialized fine-tuned models (BERT variants) and general-purpose large language models (LLMs), yet neither dominates across all instances. We introduce Learning to Defer for clinical text (L2D-Clinical), a framework that learns when a BERT classifier should defer to an LLM based on uncertainty signals and text characteristics. Unlike prior L2D work that defers to human experts assumed universally superior, our approach enables adaptive deferral-improving accuracy when the LLM complements BERT. We evaluate on two English clinical tasks: (1) ADE detection (ADE Corpus V2), where BioBERT (F1=0.911) outperforms the LLM (F1=0.765), and (2) treatment outcome classification (MIMIC-IV with multi-LLM consensus ground truth), where GPT-5-nano (F1=0.967) outperforms ClinicalBERT (F1=0.887). On ADE, L2D-Clinical achieves F1=0.928 (+1.7 points over BERT) by selectively deferring 7% of instances where the LLM's high recall compensates for BERT's misses. On MIMIC, L2D-Clinical achieves F1=0.980 (+9.3 points over BERT) by deferring only 16.8\% of cases to the LLM. The key insight is that L2D-Clinical learns to selectively leverage LLM strengths while minimizing API costs.

Earth Observation (EO) satellite scheduling (deciding which imaging tasks to perform and when) is a well-studied combinatorial optimization problem. Existing methods typically assume that the operational constraint model is fully specified in advance. In practice, however, constraints governing separation between observations, power budgets, and thermal limits are often embedded in engineering artefacts or high-fidelity simulators rather than in explicit mathematical models. We study EO scheduling under \emph{unknown constraints}: the objective is known, but feasibility must be learned interactively from a binary oracle. Working with a simplified model restricted to pairwise separation and global capacity constraints, we introduce Conservative Constraint Acquisition~(CCA), a domain-specific procedure designed to identify justified constraints efficiently in practice while limiting unnecessary tightening of the learned model. Embedded in the \textsc{Learn\&Optimize} framework, CCA supports an interactive search process that alternates optimization under a learned constraint model with targeted oracle queries. On synthetic instances with up to 50~tasks and dense constraint networks, L\&O improves over a no-knowledge greedy baseline and uses far fewer main oracle queries than a two-phase acquire-then-solve baseline (FAO). For $n\leq 30$, the average gap drops from 65--68\% (Priority Greedy) to 17.7--35.8\% using L\&O. At $n{=}50$, where the CP-SAT reference is the best feasible solution found in 120~s, L\&O improves on FAO on average (17.9\% vs.\ 20.3\%) while using 21.3 main queries instead of 100 and about $5\times$ less execution time.

Fall detection in elderly care requires not only accurate classification but also reliable explanations that clinicians can trust. However, existing post-hoc explainability methods, when applied frame-by-frame to sequential data, produce temporally unstable attribution maps that clinicians cannot reliably act upon. To address this issue, we propose a lightweight and explainable framework for skeleton-based fall detection that combines an efficient LSTM model with T-SHAP, a temporally aware post-hoc aggregation strategy that stabilizes SHAP-based feature attributions over contiguous time windows. Unlike standard SHAP, which treats each frame independently, T-SHAP applies a linear smoothing operator to the attribution sequence, reducing high-frequency variance while preserving the theoretical guarantees of Shapley values, including local accuracy and consistency. Experiments on the NTU RGB+D Dataset demonstrate that the proposed framework achieves 94.3% classification accuracy with an end-to-end inference latency below 25 milliseconds, satisfying real-time constraints on mid-range hardware and indicating strong potential for deployment in clinical monitoring scenarios. Quantitative evaluation using perturbation-based faithfulness metrics shows that T-SHAP improves explanation reliability compared to standard SHAP (AUP: 0.89 vs. 0.91) and Grad-CAM (0.82), with consistent improvements observed across five-fold cross-validation, indicating enhanced explanation reliability. The resulting attributions consistently highlight biomechanically relevant motion patterns, including lower-limb instability and changes in spinal alignment, aligning with established clinical observations of fall dynamics and supporting their use as transparent decision aids in long-term care environments

Aerial object detection in UAV imagery presents unique challenges due to the high prevalence of tiny objects, adverse environmental conditions, and strict computational constraints. Standard YOLO-based detectors fail to address these jointly: their minimum detection stride of 8 pixels renders sub-32px objects nearly undetectable, their CIoU loss produces zero gradients for non-overlapping tiny boxes, and their architectures contain significant filter redundancy. We propose DroneScan-YOLO, a holistic system contribution that addresses these limitations through four coordinated design choices: (1) increased input resolution of 1280x1280 to maximize spatial detail for tiny objects, (2) RPA-Block, a dynamic filter pruning mechanism based on lazy cosine-similarity updates with a 10-epoch warm-up period, (3) MSFD, a lightweight P2 detection branch at stride 4 adding only 114,592 parameters (+1.1%), and (4) SAL-NWD, a hybrid loss combining Normalized Wasserstein Distance with size-adaptive CIoU weighting, integrated into YOLOv8's TaskAligned assignment pipeline. Evaluated on VisDrone2019-DET, DroneScan-YOLO achieves 55.3% mAP@50 and 35.6% mAP@50-95, outperforming the YOLOv8s baseline by +16.6 and +12.3 points respectively, improving recall from 0.374 to 0.518, and maintaining 96.7 FPS inference speed with only +4.1% parameters. Gains are most pronounced on tiny object classes: bicycle AP@50 improves from 0.114 to 0.328 (+187%), and awning-tricycle from 0.156 to 0.237 (+52%).

Generative Artificial Intelligence (GenAI) tools (e.g., ChatGPT, Calude) have rapidly become integral to software development. These tools are especially attractive to students, as they can reduce cognitive load. However, their adoption also introduces a socio-cognitive risk: the accumulation of Comprehension Debt (CD). CD refers to the growing gap between what a development team knows about its codebase and what it actually needs to understand in order to maintain and modify it effectively. This qualitative study investigate how GenAI tools contribute to CD in the context of an undergraduate software engineering project. Our study is based on 621 reflective diaries from 207 students over eight weeks. We identify four CD accumulation patterns and one mitigating pattern in students' use of GenAI tools. The four accumulation patterns include: (1) AI-as-black-box code acceptance, (2) context-mismatch debt, (3) dependency-induced atrophy, and (4) verification-bypass. In contrast, the mitigating pattern involves students using GenAI as a comprehension scaffold, allowing them to build a deeper understanding of the code. We argue that CD is distinct from traditional technical debt because it resides in the collective cognition of development teams rather than in the codebase itself. Our findings highlight the need for explicit pedagogical strategies to mitigate CD in software engineering education, emphasizing verification practices, structured retrospectives, and active learning assessments.

Larger language models become simultaneously better and worse at handling contextual information -- better at ignoring false claims, worse at ignoring irrelevant tokens. We formalize this apparent paradox through the first scaling laws for contextual entrainment, the tendency of models to favor tokens that appeared in context regardless of relevance. Analyzing the Cerebras-GPT (111M-13B) and Pythia (410M-12B) model families, we find entrainment follows predictable power-law scaling, but with opposite trends depending on context type: semantic contexts show decreasing entrainment with scale, while non-semantic contexts show increasing entrainment. Concretely, the largest models are four times more resistant to counterfactual misinformation than the smallest, yet simultaneously twice as prone to copying arbitrary tokens. These diverging trends, which replicate across model families, suggest that semantic filtering and mechanical copying are functionally distinct behaviors that scale in opposition -- scaling alone does not resolve context sensitivity, it reshapes it.

Large Language Models (LLMs) are increasingly applied to complex telecommunications tasks, including 3GPP specification analysis and O-RAN network troubleshooting. However, a critical limitation remains: LLM-generated confidence scores are often biased and unreliable, frequently exhibiting systematic overconfidence. This lack of trustworthy self-assessment makes it difficult to verify model outputs and safely rely on them in practice. In this paper, we study confidence calibration in telecom-domain LLMs using the representative Gemma-3 model family (4B, 12B, and 27B parameters), evaluated on TeleQnA, ORANBench, and srsRANBench. We show that standard single-pass, verbalized confidence estimates fail to reflect true correctness, often assigning high confidence to incorrect predictions. To address this, we propose a novel Twin-Pass Chain of Thought (CoT)-Ensembling methodology for improving confidence estimation by leveraging multiple independent reasoning evaluations and aggregating their assessments into a calibrated confidence score. Our approach reduces Expected Calibration Error (ECE) by up to 88% across benchmarks, significantly improving the reliability of model self-assessment. These results highlight the limitations of current confidence estimation practices and demonstrate a practical path toward more trustworthy evaluation of LLM outputs in telecommunications.

Multimodal Large Language Models (MLLMs) have demonstrated strong cross-modal reasoning capabilities, yet their potential for vision-only tasks remains underexplored. We investigate MLLMs as training-free similarity estimators for instance-level image-to-image retrieval. Our approach prompts the model with paired images and converts next-token probabilities into similarity scores, enabling zero-shot re-ranking within large-scale retrieval pipelines. This design avoids specialized architectures and fine-tuning, leveraging the rich visual discrimination learned during multimodal pre-training. We address scalability by combining MLLMs with memory-efficient indexing and top-$k$ candidate re-ranking. Experiments across diverse benchmarks show that MLLMs outperform task-specific re-rankers outside their native domains and exhibit superior robustness to clutter, occlusion, and small objects. Despite strong results, we identify failure modes under severe appearance changes, highlighting opportunities for future research. Our findings position MLLMs as a promising alternative for open-world large-scale image retrieval.

Meta-learning offers a principled framework leveraging \emph{task-invariant} priors from related tasks, with which \emph{task-specific} models can be fine-tuned on downstream tasks, even with limited data records. Gradient-based meta-learning (GBML) relies on gradient descent (GD) to adapt the prior to a new task. Albeit effective, these methods incur high computational overhead that scales linearly with the number of GD steps. To enhance efficiency and scalability, existing methods approximate the gradient of prior parameters (meta-gradient) via truncated backpropagation, yet suffer large approximation errors. Targeting accurate approximation, this work puts forth binomial GBML (BinomGBML), which relies on a truncated binomial expansion for meta-gradient estimation. This novel expansion endows more information in the meta-gradient estimation via efficient parallel computation. As a running paradigm applied to model-agnostic meta-learning (MAML), the resultant BinomMAML provably enjoys error bounds that not only improve upon existing approaches, but also decay super-exponentially under mild conditions. Numerical tests corroborate the theoretical analysis and showcase boosted performance with slightly increased computational overhead.

In medical image segmentation, uncertainty estimates are often reported but rarely used to guide decisions. We study the missing step: how uncertainty maps are converted into actionable policies such as accepting, flagging, or deferring predictions. We formulate segmentation as a two-stage pipeline, estimation followed by decision, and show that optimizing uncertainty alone fails to capture most of the achievable safety gains. Using retinal vessel segmentation benchmarks (DRIVE, STARE, CHASE_DB1), we evaluate two uncertainty sources (Monte Carlo Dropout and Test-Time Augmentation) combined with three deferral strategies, and introduce a simple confidence-aware deferral rule that prioritizes uncertain and low-confidence predictions. Our results show that the best method and policy combination removes up to 80 percent of segmentation errors at only 25 percent pixel deferral, while achieving strong cross-dataset robustness. We further show that calibration improvements do not translate to better decision quality, highlighting a disconnect between standard uncertainty metrics and real-world utility. These findings suggest that uncertainty should be evaluated based on the decisions it enables, rather than in isolation.

Attribution methods seek to explain language model predictions by quantifying the contribution of input tokens to generated outputs. However, most existing techniques are designed for encoder-based architectures and rely on linear approximations that fail to capture the causal and semantic complexities of autoregressive generation in decoder-only models. To address these limitations, we propose Hessian-Enhanced Token Attribution (HETA), a novel attribution framework tailored for decoder-only language models. HETA combines three complementary components: a semantic transition vector that captures token-to-token influence across layers, Hessian-based sensitivity scores that model second-order effects, and KL divergence to measure information loss when tokens are masked. This unified design produces context-aware, causally faithful, and semantically grounded attributions. Additionally, we introduce a curated benchmark dataset for systematically evaluating attribution quality in generative settings. Empirical evaluations across multiple models and datasets demonstrate that HETA consistently outperforms existing methods in attribution faithfulness and alignment with human annotations, establishing a new standard for interpretability in autoregressive language models.

Neural models for TCR-pMHC binding prediction are susceptible to shortcut learning: they exploit spurious correlations in training data -- such as peptide length bias or V-gene co-occurrence -- rather than the physical binding interface. This renders predictions brittle under family-held-out and distance-aware evaluation, where such shortcuts do not transfer. We introduce \emph{Counterfactual Invariant Prediction} (CIP), a training framework that generates biologically constrained counterfactual peptide edits and enforces invariance to edits at non-anchor positions while amplifying sensitivity at MHC anchor residues. CIP augments the base classifier with two auxiliary objectives: (1) an invariance loss penalizing prediction changes under conservative non-anchor substitutions, and (2) a contrastive loss encouraging large prediction changes under anchor-position disruptions. Evaluated on a curated VDJdb-IEDB benchmark under family-held-out, distance-aware, and random splits, CIP achieves AUROC 0.831 and counterfactual consistency (CFC) 0.724 under the challenging family-held-out protocol -- a 39.7\% reduction in shortcut index relative to the unconstrained baseline. Ablations confirm that anchor-aware edit generation is the dominant driver of OOD gains, providing a practical recipe for causally-grounded TCR specificity modeling.

Adaptive Conformal Inference (ACI) provides distribution-free prediction intervals with asymptotic coverage guarantees for time series under distribution shift. However, ACI only adapts the quantile threshold -- it cannot shift the interval center. When a base forecaster develops persistent bias after a regime change, ACI compensates by widening intervals symmetrically, producing unnecessarily conservative bands. We propose Bias-Corrected ACI (BC-ACI), which augments standard ACI with an online exponentially weighted moving average (EWM) estimate of forecast bias. BC-ACI corrects nonconformity scores before quantile computation and re-centers prediction intervals, addressing the root cause of miscalibration rather than its symptom. An adaptive dead-zone threshold suppresses corrections when estimated bias is indistinguishable from noise, ensuring no degradation on well-calibrated data. In controlled experiments across 688 runs spanning two base models, four synthetic regimes, and three real datasets, BC-ACI reduces Winkler interval scores by 13--17% under mean and compound distribution shifts (Wilcoxon p < 0.001) while maintaining equivalent performance on stationary data (ratio 1.002x). We provide finite-sample analysis showing that coverage guarantees degrade gracefully with bias estimation error.

Anomaly detection aims to identify observations that deviate from expected behavior. Because anomalous events are inherently sparse, most frameworks are trained exclusively on normal data to learn a single reference model of normality. This implicitly assumes that normal behavior can be captured by a single, unconditional reference distribution. In practice, however, anomalies are often context-dependent: A specific observation may be normal under one operating condition, yet anomalous under another. As machine learning systems are deployed in dynamic and heterogeneous environments, these fixed-context assumptions introduce structural ambiguity, i.e., the inability to distinguish contextual variation from genuine abnormality under marginal modeling, leading to unstable performance and unreliable anomaly assessments. While modern sensing systems frequently collect multimodal data capturing complementary aspects of both system behavior and operating conditions, existing methods treat all data streams equally, without distinguishing contextual information from anomaly-relevant signals. As a result, abnormality is often evaluated without explicitly conditioning on operating conditions. We argue that multimodal anomaly detection should be reframed as a cross-modal contextual inference problem, in which modalities play asymmetric roles, separating context from observation, to define abnormality conditionally rather than relative to a single global reference. This perspective has implications for model design, evaluation protocols, and benchmark construction, and outline open research challenges toward robust, context-aware multimodal anomaly detection.

Analog optical computers promise large efficiency gains for machine learning inference, yet no demonstration has moved beyond small-scale image benchmarks. We benchmark the analog optical computer (AOC) digital twin on mortgage approval classification from 5.84 million U.S. HMDA records and separate three sources of accuracy loss. On the original 19 features, the AOC reaches 94.6% balanced accuracy with 5,126 parameters (1,024 optical), compared with 97.9% for XGBoost; the 3.3 percentage-point gap narrows by only 0.5pp when the optical core is widened from 16 to 48 channels, suggesting an architectural rather than hardware limitation. Restricting all models to a shared 127-bit binary encoding drops every model to 89.4--89.6%, with an encoding cost of 8pp for digital models and 5pp for the AOC. Seven calibrated hardware non-idealities impose no measurable penalty. The three resulting layers of limitation (encoding, architecture, hardware fidelity) locate where accuracy is lost and what to improve next.

Remote medical response systems are increasingly being deployed to support emergency care in disaster-affected and infrastructure-limited environments. Enabled by GeoVision capabilities, this paper presents a Digital Twin architecture for hybrid autonomous-teleoperated medical response systems. The proposed framework integrates perception and adaptive navigation with a Digital Twin, synchronized in real-time, that mirrors system states, environmental dynamics, patient conditions, and mission objectives. Unlike traditional ground control interfaces, the Digital Twin provides remote clinical and operational users with an intuitive, continuously updated virtual representation of the platform and its operational context, enabling enhanced situational awareness and informed decision-making.

The 4th Workshop on Maritime Computer Vision (MaCVi) is organized as part of CVPR 2026. This edition features five benchmark challenges with emphasis on both predictive accuracy and embedded real-time feasibility. This report summarizes the MaCVi 2026 challenge setup, evaluation protocols, datasets, and benchmark tracks, and presents quantitative results, qualitative comparisons, and cross-challenge analyses of emerging method trends. We also include technical reports from top-performing teams to highlight practical design choices and lessons learned across the benchmark suite. Datasets, leaderboards, and challenge resources are available at https://macvi.org/workshop/cvpr26.

Gaze event detection is fundamental to vision science, human-computer interaction, and applied analytics. However, current workflows often require specialized programming knowledge and careful handling of heterogeneous raw data formats. Classical detectors such as I-VT and I-DT are effective but highly sensitive to preprocessing and parameterization, limiting their usability outside specialized laboratories. This work introduces a code-free, large language model (LLM)-driven pipeline that converts natural language instructions into an end-to-end analysis. The system (1) inspects raw eye-tracking files to infer structure and metadata, (2) generates executable routines for data cleaning and detector implementation from concise user prompts, (3) applies the generated detector to label fixations and saccades, and (4) returns results and explanatory reports, and allows users to iteratively optimize their code by editing the prompt. Evaluated on public benchmarks, the approach achieves accuracy comparable to traditional methods while substantially reducing technical overhead. The framework lowers barriers to entry for eye-tracking research, providing a flexible and accessible alternative to code-intensive workflows.

Large language models (LLMs), particularly when integrated into agentic systems, have demonstrated human- and even superhuman-level performance across multiple domains. Whether these systems can truly be considered creative, however, remains a matter of debate, as conclusions heavily depend on the definitions, evaluation methods, and specific use cases employed. In this paper, we analyse creativity along two complementary macro-level perspectives. The first is a functionalist perspective, focusing on the observable characteristics of creative outputs. The second is an ontological perspective, emphasising the underlying processes, as well as the social and personal dimensions involved in creativity. We focus on LLM agents and we argue that they exhibit functionalist creativity, albeit not at its most sophisticated levels, while they continue to lack key aspects of ontological creativity. Finally, we discuss whether it is desirable for agentic systems to attain both forms of creativity, evaluating potential benefits and risks, and proposing pathways toward artificial creativity that can enhance human society.

Mapping the spatial distribution of species is essential for conservation policy and invasive species management. Species distribution models (SDMs) are the primary tools for this task, serving two purposes: achieving robust predictive performance while providing ecological insights into the driving factors of distribution. However, the increasing complexity of deep learning SDMs has made extracting these insights more challenging. To reconcile these objectives, we propose the first implementation of concept-based Explainable AI (XAI) for SDMs. We leverage the Robust TCAV (Testing with Concept Activation Vectors) methodology to quantify the influence of landscape concepts on model predictions. To enable this, we provide a new open-access landscape concept dataset derived from high-resolution multispectral and LiDAR drone imagery. It includes 653 patches across 15 distinct landscape concepts and 1,450 random reference patches, designed to suit a wide range of species. We demonstrate this approach through a case study of two aquatic insects, Plecoptera and Trichoptera, using two Convolutional Neural Networks and one Vision Transformer. Results show that concept-based XAI helps validate SDMs against expert knowledge while uncovering novel associations that generate new ecological hypotheses. Robust TCAV also provides landscape-level information, useful for policy-making and land management. Code and datasets are publicly available.

Semiconductor failure analysis (FA) requires engineers to examine inspection images, correlate equipment telemetry, consult historical defect records, and write structured reports, a process that can consume several hours of expert time per case. We present SemiFA, an agentic multi-modal framework that autonomously generates structured FA reports from semiconductor inspection images in under one minute. SemiFA decomposes FA into a four-agent LangGraph pipeline: a DefectDescriber that classifies and narrates defect morphology using DINOv2 and LLaVA-1.6, a RootCauseAnalyzer that fuses SECS/GEM equipment telemetry with historically similar defects retrieved from a Qdrant vector database, a SeverityClassifier that assigns severity and estimates yield impact, and a RecipeAdvisor that proposes corrective process adjustments. A fifth node assembles a PDF report. We introduce SemiFA-930, a dataset of 930 annotated semiconductor defect images paired with structured FA narratives across nine defect classes, drawn from procedural synthesis, WM-811K, and MixedWM38. Our DINOv2-based classifier achieves 92.1% accuracy on 140 validation images (macro F1 = 0.917), and the full pipeline produces complete FA reports in 48 seconds on an NVIDIA A100-SXM4-40 GB GPU. A GPT-4o judge ablation across four modality conditions demonstrates that multi-modal fusion improves root cause reasoning by +0.86 composite points (1-5 scale) over an image-only baseline, with equipment telemetry as the more load-bearing modality. To our knowledge, SemiFA is the first system to integrate SECS/GEM equipment telemetry into a vision-language model pipeline for autonomous FA report generation.

This discussion paper re-examines SemEval-2020 Task 1, the most influential shared benchmark for lexical semantic change detection, through a three-part evaluative framework: operationalisation, data quality, and benchmark design. First, at the level of operationalisation, we argue that the benchmark models semantic change mainly as gain, loss, or redistribution of discrete senses. While practical for annotation and evaluation, this framing is too narrow to capture gradual, constructional, collocational, and discourse-level change. Also, the gold labels are outcomes of annotation decisions, clustering procedures, and threshold settings, which could potentially limit the validity of the task. Second, at the level of data quality, we show that the benchmark is affected by substantial corpus and preprocessing problems, including OCR noise, malformed characters, truncated sentences, inconsistent lemmatisation, POS-tagging errors, and missed targets. These issues can distort model behaviour, complicate linguistic analysis, and reduce reproducibility. Third, at the level of bench-mark design, we argue the small curated target sets and limited language coverage reduce realism and increase statistical uncertainty. Taken together, these limitations suggest that the benchmark should be treated as a useful but partial test bed rather than a definitive measure of progress. We therefore call for future datasets and shared tasks to adopt broader theories of semantic change, document pre-processing transparently, expand cross-linguistic coverage, and use more realistic evaluation settings. Such steps are necessary for more valid, interpretable, and generalisable progress in lexical semantic change detection

Exploratory Landscape Analysis (ELA) provides numerical features for characterizing black-box optimization problems. In high-dimensional settings, however, ELA suffers from sparsity effects, high estimator variance, and the prohibitive cost of computing several feature classes. Dimensionality reduction has therefore been proposed as a way to make ELA applicable in such settings, but it remains unclear whether features computed in reduced spaces still reflect intrinsic properties of the original landscape. In this work, we investigate the robustness of ELA features under dimensionality reduction via Random Gaussian Embeddings (RGEs). Starting from the same sampled points and objective values, we compute ELA features in projected spaces and compare them to those obtained in the original search space across multiple sample budgets and embedding dimensions. Our results show that linear random projections often alter the geometric and topological structure relevant to ELA, yielding feature values that are no longer representative of the original problem. While a small subset of features remains comparatively stable, most are highly sensitive to the embedding. Moreover, robustness under projection does not necessarily imply informativeness, as apparently robust features may still reflect projection-induced artifacts rather than intrinsic landscape characteristics.

Large Language Models (LLMs) rely heavily on Key-Value (KV) caching to minimize inference latency. However, standard KV caches are context-dependent: reusing a cached document in a new context requires recomputing KV states to account for shifts in attention distribution. Existing solutions such as CacheBlend, EPIC, and SAM-KV mitigate this issue by selectively recomputing a subset of tokens; however, they still incur non-negligible computational overhead (FLOPs) and increased Time-to-First-Token (TTFT) latency. In this paper, we propose KV Packet, a recomputation-free cache reuse framework that treats cached documents as immutable ``packets'' wrapped in light-weight trainable soft-token adapters, which are trained via self-supervised distillation to bridge context discontinuities. Experiments on Llama-3.1 and Qwen2.5 demonstrate that the proposed KV Packet method achieves near-zero FLOPs and lower TTFT than recomputation-based baselines, while retaining F1 scores comparable to those of the full recomputation baseline.

Causal representation learning (CRL) aims to identify the underlying latent variables from high-dimensional observations, even when variables are dependent with each other. We study this problem for latent variables that follow a potentially degenerate Gaussian mixture distribution and that are only observed through the transformation via a piecewise affine mixing function. We provide a series of progressively stronger identifiability results for this challenging setting in which the probability density functions are ill-defined because of the potential degeneracy. For identifiability up to permutation and scaling, we leverage a sparsity regularization on the learned representation. Based on our theoretical results, we propose a two-stage method to estimate the latent variables by enforcing sparsity and Gaussianity in the learned representations. Experiments on synthetic and image data highlight our method's effectiveness in recovering the ground-truth latent variables.

Reliable evaluation of blastocyst quality is critical for the success of in vitro fertilization (IVF) treatments. Current embryo grading practices primarily rely on visual assessment of morphological features, which introduces subjectivity, inter-embryologist variability, and challenges in standardizing quality assurance. In this study, we propose a multitask embedding-based approach for the automated analysis and prediction of key blastocyst components, including the trophectoderm (TE), inner cell mass (ICM), and blastocyst expansion (EXP). The method leverages biological and physical characteristics extracted from images of day-5 human embryos. A pretrained ResNet-18 architecture, enhanced with an embedding layer, is employed to learn discriminative representations from a limited dataset and to automatically identify TE and ICM regions along with their corresponding grades, structures that are visually similar and inherently difficult to distinguish. Experimental results demonstrate the promise of the multitask embedding approach and potential for robust and consistent blastocyst quality assessment.

Rare events such as conformational changes in biomolecules, phase transitions, and chemical reactions are central to the behavior of many physical systems, yet they are extremely difficult to study computationally because unbiased simulations seldom produce them. Transition Path Theory (TPT) provides a rigorous statistical framework for analyzing such events: it characterizes the ensemble of reactive trajectories between two designated metastable states (reactant and product), and its central object--the committor function, which gives the probability that the system will next reach the product rather than the reactant--encodes all essential kinetic and thermodynamic information. We introduce a framework that casts committor estimation as a stochastic optimal control (SOC) problem. In this formulation the committor defines a feedback control--proportional to the gradient of its logarithm--that actively steers trajectories toward the reactive region, thereby enabling efficient sampling of reactive paths. To solve the resulting hitting-time control problem we develop two complementary objectives: a direct backpropagation loss and a principled off-policy Value Matching loss, for which we establish first-order optimality guarantees. We further address metastability, which can trap controlled trajectories in intermediate basins, by introducing an alternative sampling process that preserves the reactive current while lowering effective energy barriers. On benchmark systems, the framework yields markedly more accurate committor estimates, reaction rates, and equilibrium constants than existing methods.

As Large Language Models (LLMs) are increasingly integrated into agentic workflows, their unpredictability stemming from numerical instability has emerged as a critical reliability issue. While recent studies have demonstrated the significant downstream effects of these instabilities, the root causes and underlying mechanisms remain poorly understood. In this paper, we present a rigorous analysis of how unpredictability is rooted in the finite numerical precision of floating-point representations, tracking how rounding errors propagate, amplify, or dissipate through Transformer computation layers. Specifically, we identify a chaotic "avalanche effect" in the early layers, where minor perturbations trigger binary outcomes: either rapid amplification or complete attenuation. Beyond specific error instances, we demonstrate that LLMs exhibit universal, scale-dependent chaotic behaviors characterized by three distinct regimes: 1) a stable regime, where perturbations fall below an input-dependent threshold and vanish, resulting in constant outputs; 2) a chaotic regime, where rounding errors dominate and drive output divergence; and 3) a signal-dominated regime, where true input variations override numerical noise. We validate these findings extensively across multiple datasets and model architectures.

Mild Cognitive Impairment (MCI) affects 15-20% of adults aged 65 and older, often making kitchen navigation and independent living difficult, particularly in lower-income communities with limited access to professional design help. This study created an AI system that converts standard kitchen photos into MCI-friendly designs using the Home Design Guidelines (HDG). Stable Diffusion models, enhanced with DreamBooth LoRA and ControlNet, were trained on 100 kitchen images to produce realistic visualizations with open layouts, transparent cabinetry, better lighting, non-slip flooring, and less clutter. The models achieved moderate to high semantic alignment (normalized CLIP scores 0.69-0.79) and improved visual realism (GIQA scores 0.45-0.65). In a survey of 33 participants (51.5% caregivers, 36.4% older adults with MCI), the AI-modified kitchens were strongly preferred as more cognitively friendly (87.4% of 198 choices, p < .001). Participants reported high confidence in their kitchen choice selections (M = 5.92/7) and found the visualizations very helpful for home modifications (M = 6.27/7). Thematic analysis emphasized improved visibility, lower cognitive load, and greater independence. Overall, this AI tool provides a low-cost, scalable way for older adults and caregivers to visualize and implement DIY kitchen changes, supporting aging in place and resilience for those with MCI.

Large language models are emerging as scientific assistants, but evaluating their ability to reason from empirical data remains challenging. Benchmarks derived from published studies and human annotations inherit publication bias, known-knowledge bias, label noise, and substantial storage requirements. We present InfiniteScienceGym, a procedurally generated benchmark of scientific repositories paired with a verifiable question-answering task. From a seed, the simulator deterministically generates a self-contained repository with realistic directory structure, files, and tabular data, and a privileged QA generator produces both answerable and unanswerable questions with exact ground truth. This makes it possible to evaluate evidence-grounded reasoning, abstention, and tool-mediated analysis in a controlled setting without distributing a large static corpus. InfiniteScienceGym complements real scientific benchmarks by targeting blind spots and failure modes that are hard to evaluate using published datasets alone. Evaluating both proprietary and open-weight models, we find that none achieve more than 45% accuracy overall, that recognizing unanswerable questions remains a major weakness, and that stronger models tend to use tools more effectively rather than simply consuming more tokens.

Process reward models (PRMs) provide fine-grained reward signals along the reasoning process, but training reliable PRMs often requires step annotations or heavy verification pipelines, making them expensive to scale and refresh during online RL. Implicit PRMs mitigate this cost by learning decomposable token- or step-level rewards from trajectory-level outcome labels. However, they suffer from a train-inference mismatch: training only constrains a sequence-level aggregate, whereas inference requires token-level scores to reflect local step quality. As a result, token-level credits are weakly identified and may fail to faithfully reflect which reasoning steps are actually correct. This unreliability undermines a key promise of implicit PRMs: scoring many candidate tokens. In practice, noisy per-token advantages may systematically reinforce incorrect continuations. We address this problem with a novel Implicit Prefix-Value Reward Model (IPVRM), which directly learns a prefix-conditioned value function estimating the probability of eventual correctness, and derives step signals via temporal-difference (TD) differences. IPVRM substantially improves step-verification F1 on ProcessBench. Building on these calibrated prefix values, we further propose Distribution-Level RL (DistRL), which computes TD advantages for both sampled tokens and high-probability candidate tokens, enabling dense counterfactual updates without additional rollouts. While DistRL offers limited gains when powered by miscalibrated implicit rewards, it consistently improves downstream reasoning once paired with IPVRM.

Many materials show anisotropic light scattering patterns due to the shape and local alignment of their underlying micro structures: surfaces with small elements such as fibers, or the ridges of a brushed metal, are very sparse and require a high spatial resolution to be properly represented as a volume. The acquisition of voxel data from such objects is a time and memory-intensive task, and most rendering approaches require an additional Level-of-Detail (LoD) data structure to aggregate the visual appearance, as observed from multiple distances, in order to reduce the number of samples computed per pixel (E.g.: MIP mapping). In this work we introduce first, an efficient parallel voxelization method designed to facilitate fast data aggregation at multiple resolution levels, and second, a novel representation based on hierarchical SGGX clustering that provides better accuracy than baseline methods. We validate our approach with a CUDA-based implementation of the voxelizer, tested both on triangle meshes and volumetric fabrics modeled with explicit fibers. Finally, we show the results generated with a path tracer based on the proposed LoD rendering model.

Recent advances in agentic AI have enabled increasingly autonomous workflows, but existing systems still face substantial challenges in achieving reliable deployment in real-world scientific research. In this work, we present a safe, lightweight, and user-friendly agentic framework for the autonomous execution of well-defined scientific tasks. The framework combines an isolated execution environment, a three-layer agent loop, and a self-assessing do-until mechanism to ensure safe and reliable operation while effectively leveraging large language models of varying capability levels. By focusing on structured tasks with clearly defined context and stopping criteria, the framework supports end-to-end automation with minimal human intervention, enabling researchers to offload routine workloads and devote more effort to creative activities and open-ended scientific inquiry.

This paper presents HUANet, a constrained deep neural network architecture that unrolls the iterations of the Alternating Direction Method of Multipliers (ADMM) into a trainable neural network for solving constrained convex optimization problems. Existing end-to-end learning methods operate as black-box mappings from parameters to solutions, often lacking explicit optimality principles and failing to enforce constraints. To address this limitation, we unroll ADMM and embed a hard-constrained neural network at each iteration to accelerate the algorithm, where equality constraints are enforced via a differentiable correction stage at the network output. Furthermore, we incorporate first-order optimality conditions as soft constraints during training to promote the convergence of the proposed unrolled algorithm. Extensive numerical experiments are conducted to validate the effectiveness of the proposed architecture for constrained optimization problems.

Large language models can be aligned with human preferences through offline reinforcement learning (RL) on small labeled datasets. While single-objective alignment is well-studied, many real-world applications demand the simultaneous optimization of multiple conflicting rewards, e.g. optimizing both catalytic activity and specificity in protein engineering, or helpfulness and harmlessness for chatbots. Prior work has largely relied on linear reward scalarization, but this approach provably fails to recover non-convex regions of the Pareto front. In this paper, instead of scalarizing the rewards directly, we frame multi-objective RL itself as an optimization problem to be scalarized via smooth Tchebysheff scalarization, a recent technique that overcomes the shortcomings of linear scalarization. We use this formulation to derive Smooth Tchebysheff Optimization of Multi-Objective Preferences (STOMP), a novel offline RL algorithm that extends direct preference optimization to the multi-objective setting in a principled way by standardizing the individual rewards based on their observed distributions. We empirically validate STOMP on a range of protein engineering tasks by aligning three autoregressive protein language models on three laboratory datasets of protein fitness. Compared to state-of-the-art baselines, STOMP achieves the highest hypervolumes in eight of nine settings according to both offline off-policy and generative evaluations. We thus demonstrate that STOMP is a powerful, robust multi-objective alignment algorithm that can meaningfully improve post-trained models for multi-attribute protein optimization and beyond.

3D Gaussian Splatting (3DGS) has recently enabled highly photorealistic 3D reconstruction from casually captured multi-view images. However, this accessibility raises a privacy concern: publicly available images or videos can be exploited to reconstruct detailed 3D models of scenes or objects without the owner's consent. We present PatchPoison, a lightweight dataset-poisoning method that prevents unauthorized 3D reconstruction. Unlike global perturbations, PatchPoison injects a small high-frequency adversarial patch, a structured checkerboard, into the periphery of each image in a multi-view dataset. The patch is designed to corrupt the feature-matching stage of Structure-from-Motion (SfM) pipelines such as COLMAP by introducing spurious correspondences that systematically misalign estimated camera poses. Consequently, downstream 3DGS optimization diverges from the correct scene geometry. On the NeRF-Synthetic benchmark, inserting a 12 X 12 pixel patch increases reconstruction error by 6.8x in LPIPS, while the poisoned images remain unobtrusive to human viewers. PatchPoison requires no pipeline modifications, offering a practical, "drop-in" preprocessing step for content creators to protect their multi-view data.

Language Model (LM) agents are increasingly used in complex open-ended decision-making tasks, from AI coding to physical AI. A core requirement in these settings is the ability to both explore the problem space and exploit acquired knowledge effectively. However, systematically distinguishing and quantifying exploration and exploitation from observed actions without access to the agent's internal policy remains challenging. To address this, we design controllable environments inspired by practical embodied AI scenarios. Each environment consists of a partially observable 2D grid map and an unknown task Directed Acyclic Graph (DAG). The map generation can be programmatically adjusted to emphasize exploration or exploitation difficulty. To enable policy-agnostic evaluation, we design a metric to quantify exploration and exploitation errors from agent's actions. We evaluate a variety of frontier LM agents and find that even state-of-the-art models struggle on our task, with different models exhibiting distinct failure modes. We further observe that reasoning models solve the task more effectively and show both exploration and exploitation can be significantly improved through minimal harness engineering. We release our code \href{https://github.com/jjj-madison/measurable-explore-exploit}{here}.

On-policy distillation (OPD) has become a core technique in the post-training of large language models, yet its training dynamics remain poorly understood. This paper provides a systematic investigation of OPD dynamics and mechanisms. We first identify that two conditions govern whether OPD succeeds or fails: (i) the student and teacher should share compatible thinking patterns; and (ii) even with consistent thinking patterns and higher scores, the teacher must offer genuinely new capabilities beyond what the student has seen during training. We validate these findings through weak-to-strong reverse distillation, showing that same-family 1.5B and 7B teachers are distributionally indistinguishable from the student's perspective. Probing into the token-level mechanism, we show that successful OPD is characterized by progressive alignment on high-probability tokens at student-visited states, a small shared token set that concentrates most of the probability mass (97%-99%). We further propose two practical strategies to recover failing OPD: off-policy cold start and teacher-aligned prompt selection. Finally, we show that OPD's apparent free lunch of dense token-level reward comes at a cost, raising the question of whether OPD can scale to long-horizon distillation.

This paper studies continuous-time stochastic control problems whose controlled states are fully non-Markovian and depend on unknown model parameters. Such problems arise naturally in path-dependent stochastic differential equations, rough-volatility hedging, and systems driven by fractional Brownian motion. Building on the discrete skeleton approach developed in earlier work, we propose a Monte Carlo learning methodology for the associated embedded backward dynamic programming equation. Our main contribution is twofold. First, we construct explicit dominating training laws and Radon--Nikodym weights for several representative classes of non-Markovian controlled systems. This yields an off-model training architecture in which a fixed synthetic dataset is generated under a reference law, while the dynamic programming operators associated with a target model are recovered by importance sampling. Second, we use this structure to design an adaptive update mechanism under parametric model uncertainty, so that repeated recalibration can be performed by reweighting the same training sample rather than regenerating new trajectories. For fixed parameters, we establish non-asymptotic error bounds for the approximation of the embedded dynamic programming equation via deep neural networks. For adaptive learning, we derive quantitative estimates that separate Monte Carlo approximation error from model-risk error. Numerical experiments illustrate both the off-model training mechanism and the adaptive importance-sampling update in structured linear-quadratic examples.

Humans readily recognize objects from sparse line drawings, a capacity that appears early in development and persists across cultures, suggesting neural rather than purely learned origins. Yet the computational mechanism by which the brain transforms high-level semantic knowledge into low-level visual symbols remains poorly understood. Here we propose that ancient pictographic writing emerged from the brain's intrinsic tendency to compress visual input into stable, boundary-based abstractions. We construct a biologically inspired digital twin of the visual hierarchy that encodes an image into low-level features, generates a contour sketch, and iteratively refines it through top-down feedback guided by semantic representations, mirroring the feedforward and recurrent architecture of the human visual cortex. The resulting symbols bear striking structural resemblance to early pictographs across culturally distant writing systems, including Egyptian hieroglyphs, Chinese oracle bone characters, and proto-cuneiform, and offer candidate interpretations for undeciphered scripts. Our findings support a neuro-computational origin of pictographic writing and establish a framework in which AI can recapitulate the cognitive processes by which humans first externalized perception into symbols.

The rapid release of both language models and benchmarks makes it increasingly costly to evaluate every model on every dataset. In practice, models are often evaluated on different samples, making scores difficult to compare across studies. To address this, we propose a framework based on multidimensional Item Response Theory (IRT) that uses anchor items to calibrate new benchmarks to the evaluation suite while holding previously calibrated item parameters fixed. Our approach supports a realistic evaluation setting in which datasets are introduced over time and models are evaluated only on the datasets available at the time of evaluation, while a fixed anchor set for each dataset is used so that results from different evaluation periods can be compared directly. In large-scale experiments on more than $400$ models, our framework predicts full-evaluation performance within 2-3 percentage points using only $100$ anchor questions per dataset, with Spearman $ρ\geq 0.9$ for ranking preservation, showing that it is possible to extend benchmark suites over time while preserving score comparability, at a constant evaluation cost per new dataset. Code available at https://github.com/eliyahabba/growing-pains

Existing Multimodal Large Language Models (MLLMs) suffer from significant performance degradation on the long document understanding task as document length increases. This stems from two fundamental challenges: 1) a low Signal-to-Noise Ratio (SNR), with crucial evidence buried in irrelevant pages; and 2) supervision scarcity, as datasets offering only final short answers provide a weak learning signal. In this paper, we address these challenges by proposing a paradigm that requires the model to execute a structured Analysis, Localization and Reasoning workflow. To instill this capability, we design a two-stage training framework: we first perform Supervised Fine-Tuning on high-quality data generated via an efficient knowledge distillation strategy. Subsequently, we employ an Evidence-aware Group Relative Policy Optimization which jointly optimizes for both evidence localization and answer accuracy. Additionally, we introduce a Evidence-Guided Resolution Allocation strategy to mitigate memory constraints of training on multi-pages documents. Extensive experiments demonstrate that DocSeeker achieves superior performance on both in-domain and out-of-domain tasks. We show it robustly generalizes from short-page training to ultra-long documents and is naturally synergistic with visual Retrieval-Augmented Generation systems, serving as a solid foundation for their implementation.

While Federated Learning (FL) mitigates direct data exposure, the resulting trained models remain susceptible to membership inference attacks (MIAs). This paper presents an empirical evaluation of Differential Privacy (DP) as a defense mechanism against MIAs in FL, leveraging the environment of the 2025 NIST Genomics Privacy-Preserving Federated Learning (PPFL) Red Teaming Event. To improve inference accuracy, we propose a stacking attack strategy that ensembles seven black-box estimators to train a meta-classifier on prediction probabilities and cross-entropy losses. We evaluate this methodology against target models under three privacy configurations: an unprotected convolutional neural network (CNN, $ε=\infty$), a low-privacy DP model ($ε=200$), and a high-privacy DP model ($ε=10$). The attack outperforms all baselines in the No DP and Low Privacy settings and, critically, maintains measurable membership leakage at $ε=200$ where a single-signal LiRA baseline collapses. Evaluated on an independent third-party benchmark, these results provide an empirical characterisation of how stacking-based inference degrades across calibrated DP tiers in FL.

Humans use introspection to evaluate their understanding through private internal states inaccessible to external observers. We investigate whether large language models possess similar privileged knowledge about answer correctness, information unavailable through external observation. We train correctness classifiers on question representations from both a model's own hidden states and external models, testing whether self-representations provide a performance advantage. On standard evaluation, we find no advantage: self-probes perform comparably to peer-model probes. We hypothesize this is due to high inter-model agreement of answer correctness. To isolate genuine privileged knowledge, we evaluate on disagreement subsets, where models produce conflicting predictions. Here, we discover domain-specific privileged knowledge: self-representations consistently outperform peer representations in factual knowledge tasks, but show no advantage in math reasoning. We further localize this domain asymmetry across model layers, finding that the factual advantage emerges progressively from early-to-mid layers onward, consistent with model-specific memory retrieval, while math reasoning shows no consistent advantage at any depth.