The Inference Report

Vision-language-action (VLA) models can learn manipulation skills from demonstrations, but their capabilities are bounded by the skills in the training data. We present InSight, a framework that unlocks autonomous skill acquisition by rendering VLAs steerable at the primitive-action level (e.g., "move gripper to the bowl", "lift upward", "pour the bottle"). InSight consists of two primary stages: (1) an automated segmentation pipeline that partitions demonstrations into labeled primitives via VLM plan decomposition and end-effector poses to enable VLA primitive steerability, and (2) a VLM-guided data flywheel that identifies missing primitives required to accomplish a novel task, autonomously attempts demonstrations of the missing primitives with VLM-proposed low-level control, and automatically labels, stores, and integrates successful demonstrations into the VLA training set. We evaluate InSight across simulation and real-world manipulation tasks, including block flipping, drawer closing, sweeping, twisting, and pouring, without any human demonstrations of these target skills. Once learned, these primitives can be composed to execute novel, long-horizon tasks without additional human demonstrations. Our findings demonstrate that primitive steerability provides a practical foundation for continual skill acquisition in VLA policies. Project website: https://insight-vla.github.io.

We study the last iterate of the stochastic subgradient method for one-dimensional convex Lipschitz objectives. For a fixed horizon $n$, we consider the standard fixed stepsizes $η=Θ(1/\sqrt n)$. We prove that, for such stepsize policies, under additive i.i.d. subgradient noise with uniformly bounded variance, the last iterate features an optimization error of order $1/\sqrt n$, thereby removing the extra $(\log n)$ factor present in existing generic bounds. On the other hand, we show that without the i.i.d. assumption, the optimization error can be of order $(\log n)/\sqrt n$. Thus, under the uniformly bounded variance assumption alone, the last iterate of SsGM is suboptimal even in dimension one, resolving negatively an open problem posed in Koren and Segal, COLT, 2020.

Sparse voxel representation has emerged as a scalable foundation for image-to-3D Gaussian Splatting (3DGS) generation, yet current methods struggle to preserve high-frequency visual details of input images due to two structural bottlenecks. First, they adopt discriminative 2D features optimized for semantic abstraction to construct sparse voxel latents, which suppress reconstructive cues and induce a representation bottleneck. Second, in the generation stage, standard diffusion transformers lack effective mechanisms to align dense 2D image tokens with sparse 3D voxel latents, resulting in a cross-modal correspondence bottleneck. To address these issues, we propose FLUX3D, a scalable image-to-3DGS framework that boosts both representation learning and cross-modal alignment during generation. We first revisit 2D feature selection for sparse-voxel-based 3D representation learning, propose Diffusion-Aligned Structured Latents (DA-SLAT) and couple it with a decoder-only architecture to improve 3DGS reconstruction fidelity. We also design a sparse-structure-aware diffusion framework, which integrates the Sparse-structure Multimodal Diffusion Transformer (SMDiT) and Modal-Aware Rotary Positional Embedding (MARoPE) to achieve geometry-agnostic 2D-3D alignment. Extensive benchmark experiments demonstrate that FLUX3D yields substantial improvements in appearance fidelity and significantly outperforms all state-of-the-art (SOTA) methods in generating high-quality 3DGS assets.

Agentic language models dramatically expand the applications of AI yet little is publicly known about how to curate training data for broadly capable agents. Existing open efforts such as SWE-Smith, SERA, and Nemotron-Terminal typically target a single benchmark, leaving open the question of how to train models that generalize across diverse agentic tasks. The OpenThoughts-Agent (OT-Agent) project addresses this gap with a fully open data curation pipeline for training agentic models. We conduct more than 100 controlled ablation experiments to systematically investigate each stage of the pipeline, yielding insights on the importance of task sources and diversity. We then assemble a training set of 100K examples from our pipeline and fine-tune Qwen3-32B on this dataset, which yields an average accuracy of 44.8% across seven agentic benchmarks and a 3.9 percentage point improvement over the strongest existing open data agentic model (Nemotron-Terminal-32B, 40.9%). Moreover, our training data exhibits strong scaling properties, outperforming alternative open datasets at every training set size in compute-controlled comparisons. We publicly release our training sets, data pipeline, experimental data, and models at openthoughts.ai to support future open research on agentic model training.

Artificial intelligence (AI) can enhance what people who use augmentative and alternative communication (AAC) are able to do with their systems. However, evaluating AI-powered AAC interfaces can be difficult. People are intersectional beings and current evaluation metrics can struggle to capture the multifaceted and nuanced desires people may have for their AAC. We explore the complicated nature of six AAC problem spaces, explore how AI might be used in these spaces, and suggest more robust methods of evaluation that take the intersectional nuances of people into account. We also discuss broader issues that arise across these problem spaces and how they could be addressed using our proposed evaluation methods.

Fourier Neural Operators (FNO) learn solution operators of partial differential equations by parameterizing global convolutions in the complex Fourier domain. For real-valued PDE solutions, the complex FFT carries representational redundancy through conjugate symmetry. We introduce the Hartley Neural Operator (HNO), the exact real-valued mirror of FNO: it replaces the FFT with the purely real Discrete Hartley Transform and learns a single real multiplier per retained spectral mode, with no complex arithmetic. Because the real Hartley spectrum is not halved by conjugate symmetry, HNO retains twice as many frequency corners as FNO but one real weight where FNO carries a complex pair, so the two operators are iso-parametric at equal width and differ only in spectral basis. Our central thesis is that the best basis is a property of the operator. Self-adjoint elliptic operators (Poisson, biharmonic) have real, symmetric Green's functions that the real Hartley multiplier diagonalizes exactly, and HNO is favored there. Time-dependent operators carry phase, from oscillation in the wave equation to transport in advection, Burgers, and Navier-Stokes, which a real diagonal multiplier cannot represent, so FNO is favored there, and increasingly so with the operator's phase content, leaving the phaseless heat equation as the borderline case. Training both operators identically and benchmarking across PDE classes, initial-condition families, and boundary conditions, we find an elliptic-versus-time-dependent split that is monotone in operator phase content and matches the Green's-function theory we develop. Rather than a universal winner, our findings give a predictive rule: match the spectral basis to the symmetry of the solution operator.

Unified multi-modal large language models (MLLMs) have achieved strong text-to-image generation quality, but still struggle with structure-aware prompt following, where object counts, spatial relations, attribute bindings, and coarse layouts must be preserved. We attribute this limitation in part to the entanglement of structural planning and appearance rendering within a single conditioning stream. To address this issue, we propose Implicit Visual Chain-of-Thought (IV-CoT), a latent visual reasoning framework for query-conditioned image generation. IV-CoT decomposes the visual conditioning queries into a structural-to-semantic cascade, where structural queries first form a latent visual plan and semantic queries then render appearance conditioned on this plan. To guide the structural queries, we introduce training-only sketch supervision, which encourages them to capture structure from sketches without requiring sketch extraction or intermediate decoding at inference time. IV-CoT performs implicit CoT reasoning in a single forward pass and achieves superior results on GenEval and T2I-CompBench. Visualizations and analyses demonstrate that the learned structural and semantic queries play complementary roles in structure-aware generation.

In the big-world regime, agents cannot be universally capable and their ability is inevitably specialized across a world model in pieces. Consequently, standard uniform guarantees fail to distinguish between the understanding of critical bottlenecks and irrelevant failures. We first formalize this limitation by proving that general agents are not universal, rendering standard worst-case analysis uninformative. To overcome this, we introduce structural certification, a transition-local framework that maps bounded goal-conditioned performance to entry-wise guarantees on the agent's internal world model. Our main contribution is constructive. We provide algorithms that filter specific transitions using deep compositional goals and prove that a general agent on these goals has a structural world model with a $\mathcal{O}(1/n) + \mathcal{O}(δ)$ error bound. Conversely, this bound is tight in the small-$δ$ regime, whose existence is explicitly guaranteed by our certification. These results enable the certifiable deployment of general agents by localizing the specific transitions where long-horizon planning is reliable.

Prompt-based learning has emerged as a dominant paradigm in natural language processing. This study explores the impact of diverse pre-training objectives on the performance of encoder-decoder pre-trained language models across generation and question answering tasks, with a focus on commonsense knowledge retrieval and completion. We highlight the benefits of incorporating multiple objectives during both pre-training and fine-tuning stages. We introduce the Match Task to Objective (MTO) framework and methods for determining the appropriate objective for a given task. This framework offers automated methods to prepare task-related data for adaptation through unsupervised training, based on the identified objective. In the fine-tuning stage, we design novel templates that align with the objectives of the pre-training and adaptation stages. When aligned with task requirements, these strategies can achieve a performance gain of over 120\% compared to conventional methods in few-shot settings. They significantly outperform related works in few-shot settings and exceed the baseline even in full-dataset scenarios. Furthermore, we extend this approach to include prompt-tuning methodologies, providing guidance for more effective soft prompt engineering and optimization. Our strategies significantly enhance prompt-tuning performance as well. These insights hold substantial value, precisely guiding the selection and optimization of models customized for specific tasks. Code is available at https://github.com/puraminy/MTO/

Agentic data analysis systems produce rich outputs, including code, numerical results, and verbal diagnostics. This makes them more challenging to evaluate than single-turn LLM responses. It is therefore necessary to distinguish genuine disagreement between an agent's output and a ground-truth answer from grading artifacts. We investigate how reliably automated graders assess such a system and what strategies improve grading quality by applying LAMBDA, a multi-agent data-analysis system, on 153 numerical QRData tasks from DSGym. We develop and evaluate a three-layer human-AI grading cascade: strict regex matching, LLM-based lenient grading, and snippet-based human inspection, which combines non-GenAI and GenAI strategies with different failure profiles. Both automated graders achieve 100% observed precision (0/70 false positives). The lenient grader's recall is 97% against human labels. A keyword-anchored extraction pipeline raises the strict grader's recall by 60 percentage points over a last-number heuristic; the lenient grader is architecturally parser-independent. An iterative nudge mechanism raises grading run success from 36% to 97% and lenient-pass rates from 16% to 46%; comparing nudging with and without original-question re-injection shows that re-injection offers no benefit, confirming the nudge as an answer template cue. We further observe in this case study that variable type is the task metadata field most consistently associated with grading pipeline dynamics and observed outcome grades.

LLM-based dialogue assistants have become mainstream tools for software developers, yet current evaluation benchmarks focus exclusively on functional correctness. This leaves a critical gap in assessing the quality and accuracy of these conversations when handling Non-Functional Requirements (NFRs), which are inherently vague, context-dependent, and involve many parts of a program. Evaluating how well these systems support collaborative reasoning about NFRs requires methods that go beyond single-turn accuracy to capture both the correctness of the system's outputs and the quality of the multi-turn interaction. In this paper, we investigate the accuracy and quality of multi-turn conversations between developers and an LLM-based agent in the domain of Health Insurance Portability and Accountability Act (HIPAA) regulatory compliance. We hired 49 programmers to interact with GitHub Copilot to assess 148 HIPAA-derived NFRs against the iTrust codebase, a system designed to comply with HIPAA regulations, across three dimensions: requirement satisfaction level, reasoning, and code localization. We find that developers tend to agree with LLM assessments, but accuracy against expert ground truth is low. We model user satisfaction and find that longer system responses and more information-providing turns negatively affect user satisfaction, whereas proactive interactions positively affect it. Our findings provide insights for designing LLM-based dialogue systems that support NFR assessment.

Over a series of seven papers, Andreas & Günther have introduced seven definitions of actual causation and have classified them as belonging to three different, competing, types of accounts: factual difference-making, counterfactual difference-making, and regularity-based. I show that their most recent - factual difference-making - definition instantiates all three types, thereby proving that these are distinctions without a difference. I further compare their novel account to the other six accounts on several crucial examples, revealing that this undermines all seven of their accounts.

Scientific long-document summarization datasets commonly treat author-written abstracts as gold reference summaries, although their quality and alignment with the source article vary. At the same time, publicly available scientific summarization datasets remain limited in scale and structure for modern long-context models. In this work, we address both challenges by a) constructing and releasing one of the largest biomedical and life science datasets for long-document summarization, containing 1.88 million PMC articles, and b) analyzing the reference quality of author-written abstracts with source-grounded and model-based metrics. We show that author-written abstracts vary in their alignment with the full article and that these quality signals can guide training-data selection. Training on selected high-quality subsets outperforms random sampling at matched training sizes and can match or exceed larger random subsets on factuality-oriented metrics. Our findings suggest that reference quality is an important factor in scientific summarization and that quality-aware data selection can improve training efficiency.

Part-of-Speech (POS) tagging is a foundational NLP task underpinning machine translation, information extraction, and syntactic parsing. Despite Marathi being spoken by over 83 million people and ranking among the top twenty most spoken languages worldwide, it remains severely under-resourced in annotated corpora and standardised evaluation benchmarks. Marathi presents unique challenges for computational modelling owing to its rich morphology, relatively free word order, lack of capitalisation conventions, and pervasive code-mixing with Hindi and English. We introduce L3Cube-MahaPOS, a gold-standard POS tagging dataset for Marathi comprising 32,354 manually annotated sentences drawn from news text. Annotation was performed entirely manually by a team of Marathi-proficient annotators following a 16-tag Universal Dependencies-aligned scheme. A structured preprocessing pipeline covering Unicode normalisation, Devanagari-aware tokenisation, and noise filtering ensures label consistency across all splits. We benchmark the dataset across six model families spanning HMM, CRF, BiLSTM, BiLSTM+CharCNN, MuRIL, and the Marathi-specific transformer MahaBERT-v2. The best system achieves 88.67\% token-level accuracy and a macro-F1 of 81.67% over 15 evaluated tag classes. We release the dataset, annotation guidelines, and trained model checkpoints to foster further research in Marathi NLP.

Modeling chaotic systems is crucial yet challenging. Inverse problems in chaotic dynamics, namely inferring initial conditions from final states, remain largely unsolved because of ill-posedness, non-uniqueness, instability, and potentially chaotic time-reverse dynamics. We address this open problem with Bidirectional Conditional Flow Matching (Bi-CFM), which learns bidirectional mappings between distributions of initial and final states to capture the stochasticity of chaotic evolution and mitigate exponential error accumulation over time. Furthermore, for systems with conservation laws, we extend it to Conservation-constrained Bi-CFM (CBi-CFM). Across the classic Lorenz, Circuit, and high-dimensional Lorenz 96 systems, Bi-CFM improves five distribution-level metrics over baselines while achieving a speedup of more than two orders of magnitude. In the three-body planet-planet scattering problem in planetary dynamics, CBi-CFM better respects conservation laws, with conservation errors comparable to those of the ground truth. Finally, on real observations of globular clusters, collisional million-body systems shaped by $\sim 10^{10}$ years (10 Gyr) of evolution, our method represents an advance in accuracy, establishing a scalable route to solving inverse problems of long-timescale real-world chaotic dynamics.

LLM agents solve repository-level coding tasks through multi-turn tool use, but utilize half their budget on locating faults before editing. Dedicated localization frameworks have emerged, yet are still evaluated as file retrieval rather than actionable diagnosis, producing locations without the diagnostic context a repair agent needs. We introduce SHERLOC (Structured Hypothesis-driven Exploration and Reasoning for Localization), a training-free framework pairing a reasoning LLM with compact repository tools and self-recovery, without fine-tuning or multi-agent orchestration. SHERLOC reaches state-of-the-art localization across model scales: 84.33% accuracy@1 on SWE-Bench Lite and 81.27% recall@1 on SWE-Bench Verified; at ~30B parameters, it matches or outperforms other agentic methods. Injecting our locations and diagnostic findings into repair agents yields, on average, +5.95 pp resolve rate on SWE-Bench Verified while cutting localization and total tokens by 36.7% and 23.1%.

Vision-Language-Action (VLA) models are emerging robotic control systems that integrate perception, language understanding, and action generation in a unified architecture. Existing testing approaches for VLA-enabled robots rely on manually constructed symbolic test oracles that determine task success from final environment states. These oracles are costly to construct, require domain expertise, and are often tightly coupled to specific tasks and environments, limiting scalability and reuse. Furthermore, they provide only end-state assessments of task outcomes, offering limited insight into intermediate behavior and fault localization. To address these limitations, we introduce MANGO, a multi-agent framework that automatically generates fine-grained oracles from natural-language descriptions of robotic tasks. MANGO first generates a reusable library of atomic tasks, then generates simulator-grounded oracle definitions for each atomic task, and finally produces executable fine-grained oracles by decomposing complex instructions into ordered sequences of atomic actions and corresponding oracles. The framework uses collaborative Generator, Assessor, and Judge agents that iteratively refine generated artifacts through structured feedback. We evaluate MANGO on the LIBERO_10 and RoboCasa Humanoid Tabletop benchmarks. Results show that MANGO generates executable, fine-grained oracles that detect a similar number of failures as symbolic oracles while accurately localizing them and providing richer diagnostic information. Through ablation studies, we further analyzed component contributions and the effect of initial task set, while preserving oracle quality. Overall, the results show the feasibility and effectiveness of test oracle generation for VLA-enabled robots testing.

Quantum computers could outperform classical machines on important problems, but only if the errors that pervade quantum hardware can be corrected at scale. Quantum low-density parity-check (qLDPC) codes offer a promising route to this goal by combining sparse parity checks with finite encoding rate and growing distance, but their construction remains a challenging discrete design problem. Here we introduce structured concept evolution (SCE), a search framework that pairs a large language model with a structured algebraic mutation grammar to discover lifted-product code families, a class of CSS qLDPC codes. Instead of asking the LLM to design codes from first principles, SCE evolves structured concepts consisting of algebraic specifications paired with executable programs that realize them, using hierarchical mutations that modify the group algebra, protograph geometry, or base space. Running SCE, we discover a diverse set of competitive code families, ranging from abelian constructions to families over non-abelian groups beyond those underlying standard designs such as bivariate-bicycle codes, and characterize them under code-capacity depolarizing noise with BP+OSD decoding. These results are obtained with lightweight models (GPT-5.4-mini and GPT-5.4-nano).

Generic text-to-video models can be used as rich open-world scene priors. Despite the high quality of today's generated videos, they do not directly yield reliable 3D assets: camera motion is difficult to control, view coverage is partial, and frames often contain inconsistencies across time. We introduce OrbitForge, an adapter built from frozen video priors and per-prompt Gaussian Splatting reconstruction optimization that converts a single text-generated video into a canonical closed-orbit 3D Gaussian Splatting scene. We use 3D reconstruction as an anchor to improve the 3D consistency of the generated video. We obtain a preliminary 3D reconstruction from a first generated video via Deformable Gaussian Splatting with a robust MedianGS proxy. We render views from a prescribed orbit to detect missing viewpoints. OrbitForge uses the text-to-video model to complete only the missing views, and reconstructs the completed orbit into a final Gaussian Splatting scene. This design requires no task-specific video or multiview fine-tuning, avoids per-prompt score-distillation optimization, and does not progressively generate views one step at a time. We further argue that this setting demands coverage-aware evaluation: local smoothness alone rewards methods that never attempt a full orbit. On a frozen 300-prompt T3Bench-derived audit, OrbitForge reconstruction attains a 359.0-degree measured median span, raises originally unsupported-bin Q10 ImageReward from 8.07 to 16.36 relative to MedianGS-only reconstruction, while remaining competitive with VideoMV on the coverage-quality.

Recent advances in Video Large Language Models (Video-LLMs) have yielded promising performance on video question answering (VideoQA). Nevertheless, existing benchmarks are predominantly evaluated through answer correctness, while the grounding of predictions in relevant video evidence remains largely unexamined. This disconnect between answer generation and evidence understanding motivates the construction of the Evidence-Grounded Video Question Answering Benchmark (EG-VQA), an open-ended evaluation protocol in which each QA pair is explicitly annotated with supporting temporal evidence, thereby requiring joint reasoning and precise evidence localization. EG-VQA is comprised of 2,067 videos and 11,838 QA pairs with fine-grained evidence annotations. To evaluate predicted evidence, Evidence-Grounded F1 (EG-F1) is introduced as a unified metric in which temporal alignment and semantic consistency against ground-truth evidence are jointly measured. Experimental evaluation reveals that even strong proprietary models struggle to accurately ground their predictions, exposing a fundamental discrepancy between answer correctness and faithful evidence localization. To bridge this gap, EG-Reasoner, an evidence-grounded reasoning model trained with explicit supervision, is proposed. State-of-the-art performance is achieved among open-source models, with results competitive against proprietary systems, particularly pronounced gains are observed on reasoning-intensive tasks such as counterfactual questions. These findings demonstrate that scaling alone is insufficient for robust video understanding and that structured evidence supervision is essential for the development of more reliable and interpretable VideoQA systems.

Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse tasks, yet they remain prone to generating hallucinations. Detecting these hallucinations is critical for deploying LLMs reliably in high-stakes applications. We present Grad Detect, a gradient-based approach for predicting hallucinations by analyzing layer-wise gradient patterns from a single forward-backward pass during inference. Our method shows that the internal gradient structure of a model carries rich information about the correctness of its output. This information is not accessible through output-level signals alone. We evaluate Grad Detect on several Q&A benchmarks across both hallucination detection and model abstention prediction, where it consistently outperforms confidence-based and sampling-based baselines. Through comprehensive layer ablation studies across all eleven models from four architectural families, we find that the final five layers concentrate over 97% of the discriminative gradient signal, enabling efficient deployment with minimal performance loss. Grad Detect provides a unified framework for predicting multiple dimensions of LLM reliability, offering strong predictive performance alongside interpretable insights into where and how model failures originate.

Commercial NLP treats the shopping chatbot as a recommender or a conversion tool: its job is to match a user to a catalogue entry and close a sale. We argue that the arrival of agent-native micro-payment rails (e.g., x402, AP2) changes what is scarce. When the buyer is an autonomous agent that can investigate exhaustively, the bottleneck is no longer matching products but acquiring trustworthy, decision-relevant information about them. We envision agentic e-commerce as a micro-transaction market for verified information: buyer agents spend fractions of a cent to progressively unlock seller- and reviewer-supplied data -- service histories, third-party test reports, bills of materials, audited sales and support metrics -- paid for a la carte under a freemium model, with reviewer trust scored reputationally. We sketch the architecture of such a market and argue that it rewards genuine product quality and yields truer competition than ranking-based storefronts. We then translate the vision into concrete NLP problems -- cost-optimal information acquisition, data pricing and negotiation, real-time entity resolution, grounded value exchange, and privacy-preserving persona modelling -- and argue that these, not chat fluency, deserve the field's attention.

While the field of Human Activity Recognition (HAR) continues to draw interest from researchers and advance in important ways, some key challenges remain. One of the most difficult aspects of building HAR models that show good performance in real-world settings is dealing with data diversity from device and sensor heterogeneity, and contextual changes that are intrinsic to real-world applications. While data diversity in HAR has been well-acknowledged in the literature, there remains a gap in understanding the effect of various types of distribution shifts on HAR models and the domain generalization problem that arises. Towards that end, this paper systematically evaluates 4 different types of distribution shifts, including variations in device type, sensor placement, sampling rate, and user behavior. Quantifying their effects, we illustrate that diversity shifts predominantly define all types of shifts, indicating the existence of unique features that are not shared across different domains. We then introduce a uniform HAR-based distribution shift benchmarks and conduct a comprehensive evaluation of up to 28 domain generalization methods. Our analysis exposes the limitations of current domain generalization algorithms in achieving model generalizability, marginally outperforming the empirical risk minimization baseline. This work represents the first systematic exploration of domain generalization and adaptation concerning specific distribution shifts in sensor-based HAR, offering an open-source benchmark platform and datasets to spur further research.

Progress in deep learning is, at scale, more a matter of systems engineering than of modelling: the behaviour of a model in training (its throughput, its memory footprint, and the numerical fidelity of the result) is determined less by the architecture itself than by how that architecture is expressed on the hardware. To achieve absolute control over this hardware expression while abstracting away systems complexity to make modelling seamless and eliminating the need for repetitive orchestration logic, BluTrain was architected from first principles as a robust, lightweight, and architecture-general training framework in standard C++ and the core CUDA programming model. Every layer is implemented natively: a typed tensor module with reverse-mode autograd, a linear-algebra library, a caching allocator, a multi-mode distributed-execution module, and an MLIR-based deep-learning compiler. In formal evaluations training a 124M-parameter GPT-2 baseline in FP32 on an 8-GPU 6000 Ada system, BluTrain outperforms industry-standard baselines in both throughput (sustaining an average of 407K tokens/s versus PyTorch's 395K tokens/s) and memory efficiency (achieving up to a 22% footprint reduction), while strictly preserving numerical fidelity and converging to a marginally lower final validation loss. With every layer explicitly open to native tuning, the performance ceiling is the framework's own to raise.

Birth defects are a major cause of fetal loss, neonatal morbidity and long-term disability. In the subset with suspected genetic etiologies, exome and genome sequencing have moved many cases from variant detection to post-sequencing interpretation: clinicians must rank patient-specific candidate variants under incomplete fetal or infant phenotypes and heterogeneous evidence from population genetics, variant-effect prediction, gene-disease validity, phenotype ontologies, cellular and pathway context, protein structure and clinical literature. We present DeepBD, a grounded agentic workflow for variant prioritization and diagnostic interpretation of genetic birth defects. DeepBD organizes the workflow into LLM-assisted case structuring, a pretrained evidence engine, specialist evidence modules and a grounded diagnostic review layer. The evidence engine learns patient-specific variant scores from structured rule evidence, sequence and variant-effect representations and phenotype-conditioned biological context, whereas specialist modules and the agentic layer provide tool-based refinement, candidate-pool review and diagnosis-oriented synthesis from ranked candidates. Developed using an in-house fetal and infant cohort comprising 18,622 cases, DeepBD achieved Recall@1/3/5/10 of 0.658/0.882/0.912/0.929 on an internal held-out solved-case benchmark, outperforming standalone Exomiser, DeepRare and prompted LLM reranking baselines evaluated on Exomiser-derived top-20 candidate variants. Ablation and overlap analyses show that rule evidence, mechanistic context, and specialist refinement provide complementary signals. These findings support a grounded agentic workflow that separates evidence integration, tool-based refinement, and LLM-assisted diagnostic review for retrospective variant prioritization in genetic birth defects.

Memory for large language model (LLM) agents has rapidly evolved from simple retrieval-augmented mechanisms into a data management system that supports persistent information storage, retrieval, update, consolidation, and dynamic lifecycle governance throughout agent execution. Despite this evolution, existing evaluations still benchmark agent memory mainly through end-to-end task success metrics (e.g., F1, BLEU), while treating the underlying system as a monolithic black box. As a result, critical system-level concerns, including operational costs, architectural trade-offs across memory modules, and robustness under dynamic knowledge updates, remain insufficiently explored. In this paper, we present a systematic experimental study of agent memory from a data management perspective. We propose an analytical framework that decomposes agent memory into four core modules: memory representation and storage, extraction, retrieval and routing, and maintenance. Under this framework, we evaluate 12 representative memory systems and two reference baselines across five benchmark workloads spanning 11 datasets. Our extensive end-to-end evaluation shows that no single architecture dominates across all scenarios; instead, effectiveness depends heavily on how well the memory structure aligns with the workload bottleneck. Furthermore, through fine-grained ablation studies, we quantify their individual effects on representation fidelity, retrieval precision, update correctness, and long-horizon stability. Finally, we reveal cost-performance trade-offs under realistic workloads, showing localized maintenance is more cost-efficient than global reorganization. Based on these findings, we identify promising directions towards building truly agent-native memory systems. The code is publicly available at https://github.com/OpenDataBox/MemoryData.

Non-autoregressive generation offers a powerful paradigm for iterative refinement, allowing models to recursively critique, erase and regenerate arbitrary subsets of tokens. However, existing non-autoregressive models fail to realize this potential. Masked Diffusion Models (MDMs) suffer from factorization error, causing sample quality to collapse when generating multiple tokens simultaneously. Flow Map Language Models (FMLMs) circumvent this bottleneck via joint sequence transport for excellent few-step generation, but sacrifice the inference-time flexibility of MDMs. We introduce FMLM+, a framework that bridges this gap by equipping FMLM with masking-style noise schedules. While generating the full sequence in a single step, FMLM+ simultaneously scores the global consistency of each token a posteriori. We leverage this to introduce Posterior Refinement, a novel inference-time refinement strategy that enables the model to adaptively self-correct its outputs, matching the performance of discrete baselines with 32x fewer NFEs. Across diverse benchmarks, we demonstrate that FMLM+ with Posterior Refinement improves the speed--quality tradeoff over both MDM and FMLM families, providing a scalable foundation for high-fidelity language modeling.

Even when Dirac-Frenkel dynamics determine a well-defined evolution in function space, the corresponding parameter dynamics can be non-unique or ill-conditioned for redundant nonlinear parametrizations such as neural networks or mixture models. We propose to add inertia to the Dirac-Frenkel dynamics and show that this allows useful parameter velocity information to persist from the past trajectory in directions that are weakly informed, while well-informed parameter velocity directions continue to follow the Dirac-Frenkel dynamics. We prove that the inertial formulation yields well-posed parameter dynamics and provide a posteriori error bounds. After time discretization, the method requires the solution of the same type of regularized linear least-squares problem as standard Dirac-Frenkel dynamics, but with the previous velocity appearing as an anchor. Numerical experiments demonstrate the increased robustness obtained with inertia.

Recent multimodal large language models (MLLMs) have shown strong potential for autonomous driving scene understanding, yet existing methods still face a fundamental trade-off between temporal reasoning and spatial precision. Models that rely on single-frame or low-resolution inputs often miss small, distant, or partially occluded hazards, while language-centric driving models frequently provide limited grounded evidence for their explanations. To address this gap, we propose UniDrive, a unified visual-language and grounding framework for interpretable risk understanding in autonomous driving. UniDrive combines a temporal reasoning branch that models scene dynamics from multi-frame visual input with a high-resolution perception branch that preserves fine-grained spatial details from the latest frame. The two branches are integrated through a gated cross-attention fusion module, enabling dynamic context to be aligned with precise spatial evidence. Based on the fused representation, UniDrive jointly generates natural-language risk descriptions and grounded bounding-box outputs for risk objects. Experiments on the DRAMA-Reasoning benchmark show that UniDrive outperforms representative image-based and video-based baselines in both captioning and risk-object grounding. In particular, UniDrive achieves the best overall performance on the validation split and demonstrates clear advantages in small-object localization, zero-shot generalization to NuScenes and BDD100K, and human-rated interpretability and trustworthiness. These results suggest that explicitly combining temporal semantics and high-resolution perception provides a stronger foundation for interpretable and safety-oriented autonomous driving systems. The code is available at https://github.com/pixeli99/unidrive-dev.

Handling repeated characters in text can be tricky, since they can represent either the correct spelling of a word or informal character elongation often seen in social media posts. We present CANDLE, a lightweight system for character-level Arabic noise deduplication that addresses this challenge without relying on handcrafted rules, dictionaries, or morphological analyzers. At the heart of CANDLE is a novel application of Connectionist Temporal Classification (CTC) to this task, a formulation not previously explored for character deduplication, which frames normalization as a sequence alignment problem over a character-based encoder. Evaluated on three benchmarks spanning clean newspaper, manually curated ambiguous cases, and real-world social media text, the CTC model achieves a Sentence Error Rate (SER) as low as $5.37\%$ and consistently outperforms a classification-based baseline by a large margin. To reduce inference overhead, we distill the 6-layer CTC model into a 2-layer student, achieving a $3\times$ depth reduction with minimal performance degradation. Beyond deduplication accuracy, normalization yields a practical downstream benefit: a relative reduction in tokenizer fertility of up to $12.8\%$ across a diverse set of Arabic LLM tokenizers, directly lowering inference costs and improving context window utilization. We release all code and models publicly to support reproducibility and advance future research\footnote{https://github.com/abjadai/candle}.

The loss of plasticity - the ability of a network to learn new information after having already learned older information - is a fundamental challenge in creating artificial neural networks capable of continual learning. Although this phenomenon has been known for decades, it has mostly been studied in older, relatively small architectures and rarely in natural-language domains. To determine whether loss of plasticity remains a problem in the modern transformer-based LLM paradigm, we study plasticity loss in GPT-style Transformer models trained on a multilingual continual learning problem. Consistent with prior work, we find evidence of plasticity loss across models ranging from 5M to 314M non-embedding parameters, as measured by deterioration on a held-out Vietnamese probing task. We further find that the onset of plasticity loss follows a predictable scaling law, growing sublinearly with model size. These results suggest that larger models may delay the measurable effects of plasticity loss, but that increasing parameter count alone is likely to be insufficient to completely prevent it. We also find evidence of plasticity loss under stationary multilingual training, challenging the view that the phenomenon is exclusive to continual learning with abrupt task changes. Overall, our results suggest that even large Transformer language models trained on natural-language will eventually lose the ability to efficiently adapt to new data after sufficiently long training, in both continual and stationary settings.

Large Language Models (LLMs) achieve strong performance across a growing range of domains, yet their scale poses deployment challenges in applications where latency and cost constraints are critical. This paper derives empirical scaling laws for domain-specific LLM compression, quantifying how in-domain and general knowledge performance scale with dataset size, compression ratio, supervision format, and iterative pruning schedule. Using quantitative finance as our application domain, we compare logit-based and LoRA-based distillation under iterative structural pruning, introducing a blended chain-of-thought supervision loss that stabilizes KL-divergence distillation over reasoning traces. In-domain task quality degrades predictably under compression while general-knowledge benchmarks collapse well before the same point; supervision format is the key driver of this tradeoff, with chain-of-thought supervision actively recovering general knowledge that pruning erases. We release the headline dataset FinHeadlineMix, scaling law results, and practical recommendations to provide a reusable framework for domain-specific compression decisions.

Generative models, particularly diffusion and score-based approaches, have recently achieved strong performance in speech enhancement, but their iterative sampling process limits real-time deployment. Flow Matching offers an efficient alternative by transporting noisy speech toward clean speech through an ordinary differential equation with few function evaluations. In this work, we propose a skip-free encoder-decoder backbone for flow-matching speech enhancement, guided by Latent Representation Alignment (LRA). Instead of relying on U-Net skip connections, which may transfer noise-correlated low-level features to the decoder, the proposed model aligns its bottleneck and decoder representations with clean latent features extracted from a frozen Descript Audio Codec encoder-decoder without quantization. This codec-aligned supervision promotes compact clean-speech representations while preserving efficient few-step inference. Experiments on WSJ0-CHiME3 and VoiceBank-DEMAND show improved PESQ and perceptual quality, especially on VoiceBank-DEMAND, using only five function evaluations.

High-quality annotations of structured representations are expensive to collect over large corpora. Manual annotation of structure is laborious, and model-based annotation, although cheaper to generate, requires expensive validation and potentially significant supervision to ensure that the annotation quality is strong enough to be useful downstream. In traditional annotation workflows, annotation of each complete example is performed end-to-end by a single annotator. However, structured annotation is complex, and each aspect of the task represents a unique challenge with an associated inferential load for a given annotator. Modern annotation projects can incorporate heterogeneous groups of annotators, including both models and human annotators with varying domain and linguistic expertise. It remains unclear, however, how to redesign annotation tasks in this setting, where efforts are discriminately allocated across heterogeneous annotators with respect to distinct annotation challenges. We propose to decompose annotation tasks into sub-tasks in order to reduce the aggregate inferential load of annotation projects. Inspired by the notion of centers from centering theory, we introduce a formal model of inferential load based on the degrees of freedom in the space of valid annotations. Using this model, we show that identifying these centers (i.e. salient anchor entities realized by annotation sub-tasks) constrains the output space complexity, and decompositions which isolate and advance center identification reduce the aggregate inferential load. We provide guidelines for decomposing complex structured annotation tasks, supported by examples demonstrating improved cost-efficiency from our prior work. Finally, we present a procedure for allocating sub-tasks across annotators to maximize quality under a fixed budget.

Frontier AI training is increasingly shaped by access to dense, centrally controlled accelerator clusters. This creates a structural advantage for hyperscalers and large centralized laboratories, and makes open or independent AI efforts depend on scarce capital, privileged infrastructure, and data-center geography. We present Spheroid BlockTrain, a decentralized training protocol in which a model is partitioned into independently trainable blocks, each optimized on a local objective derived from the same global target and composed at inference into one model. On byte-level WikiText, BlockTrain reaches cross entropy 1.359 (perplexity 3.89), within about 0.04 CE of a same-setup end-to-end Transformer reference, while each active worker trains only one block and avoids full-model optimizer state. A shared six-worker block training run reaches CE 1.385 by averaging same-block updates into one assembled model. HTTP/TCP transport experiments move real serialized checkpoints and updates, including a public-IP three-host run that improves CE from 5.580 to 1.811 while moving 15.22 GB. For inference, the current BlockTrain path uses one block-stack traversal per full output and serves over direct TCP across three public-network GPU hosts up to a 75.80B-parameter logical fp16 shape, outperforming a matched plain-autoregressive TCP pipeline baseline because it emits a full sequence per WAN pipeline traversal rather than one token per traversal.

When a core definition changes, every dependent artefact must be updated, a cascading problem central to software maintenance. In Model-Driven Engineering (MDE), the dominant two-level modelling (2LM) paradigm fragments domain knowledge across metamodel and model artefacts that must be kept mutually consistent, making co-evolution a persistent source of inconsistencies and effort. Multi-level modelling (MLM) unifies these artefacts and is claimed to reduce co-evolution burden, but this has not been tested in a controlled, paired comparison against 2LM. We hypothesise that MLM's structural unification yields fewer post-change inconsistencies and a smaller modification footprint than 2LM for semantically equivalent evolution scenarios. To test this, we present a pre-registered, mutation-based empirical comparison of co-evolution behaviour in both paradigms. From a curated corpus of published 2LM co-evolution scenarios, we construct semantically equivalent MLM counterparts, apply identical evolution mutations to both, and measure outcomes through automated consistency checking and pre-registered hypothesis tests. Positive controls and a blinded mapping protocol guard against bias. This design provides the first empirical framework for assessing whether paradigm-level structural choices affect cascading maintenance burden, operationalising co-evolution burden as two automatically measurable outcome variables and delivering a reusable benchmarking protocol for replication and extension.

MATLAB/Simulink provides a practical modeling language and a simulation engine for the development of cyber-physical systems. To ensure the quality of the developed models, there are formal verification tools available, such as Simulink Design Verifier (SLDV) and third-party SMT-based model checkers (SmtMC). However, due to the absence of a semantics of Simulink that covers every element of models and the details of its numerical behavior, the reliability of the model checkers themselves is often doubtful, potentially analyzing models differently from the simulator. This work aims to verify the quality of the Simulink model checkers by addressing the following items. 1) Formalization of the basic block types of Simulink. It involves defining block type feature sets and the bit-precise behavior of the blocks. 2) A method for testing bit-precise conformance relations among the tools for each block type. The pass rate of our test suites measures (i) conformance of model checking results with simulation results by Simulink and (ii) conformance between the results of SmtMC and SLDV. 3) Experiment to perform tests on 10 block types. We confirmed that SmtMC efficiently passed all test cases, while SLDV achieved pass rates of only 94-96% and 80-90% for conformance (i) and (ii), respectively. We analyzed the causes of failed tests, such as errors, corner cases, and timeouts.

Sparse autoencoders (SAEs) are increasingly used to extract interpretable concepts from vision and vision language models, yet existing evaluation methods largely rely on proxy metrics or qualitative inspection rather than measuring semantic correspondence. We present a human-grounded evaluation framework that quantifies alignment between SAE latents and human-annotated concepts, without requiring user studies, and validate this matching through targeted attribute perturbations. To enable this intervention-style evaluation in vision, we construct synCUB and synCOCO, synthetic benchmarks of paired images that differ in exactly one attribute. We introduce Fully-Binary Matching Pursuit (FBMP), a coalition-based matching procedure that supports many-to-one mappings between SAE latents and annotated concepts, and consistently outperforms one-to-one baselines. For functional validation, we propose a Targeted Attribute Perturbation Alignment Score (TAPAScore), which tests whether matched concepts respond selectively and in the expected direction under targeted image-level attribute perturbations. Under sanity checks, our matching and TAPAScore are the only evaluated metrics that reliably distinguish trained SAEs from untrained ones. Across SAEs trained on CLIP and DINOv2 embeddings, we find that increased overcompleteness can reduce perturbation alignment, indicating a reduction in interpretability. Our evaluation framework suggests that moderate dictionary sizes provide the best trade-off, yielding the most interpretable SAEs. Code and datasets are available at https://github.com/JonasKlotz/sae-concept-eval.

We consider the problem of model selection between probabilistic models on data sets of time series. Chosen a proper scoring rule, we denote by the term \textit{score} the average value of the scoring rule on the test of an individual time series. For model selection, we need aggregating the values of the scores across multiple time series. Three summary statistics are commonly used for model selection: mean score, median score, and mean rank. Results in previous papers show that these statistics can yield conflicting decisions; we show how the conflicting conclusions are due to the skewness of the distribution of scores. We also show that as the test set of each time series of the data set increases, the different model selection criteria progressively converge to the same conclusion. However, for short tests sets, only the mean score identifies the true model as the best. We illustrate these phenomena with an analysis on intermittent time series, including the data set of the M5 competition, where we underline the importance of having a large test set. In such experiments, we further notice that model selection based on mean ranks remains unchanged using different scaling factors.

Chinese news text contains dense written forms such as scores, hyphenated model names, ranges, unit symbols, percentages, English abbreviations, and mixed Chinese-Latin-digit names. These forms are frequent in real listening workflows, and a text-to-speech (TTS) system can preserve the written string while changing the spoken meaning. We introduce CN-NewsTTS Bench v0.1, an open target-level benchmark for evaluating whether Chinese news TTS products pronounce such targets correctly from raw text, without user-side rules, LLM rewriting, SSML hints, or manual edits. The release contains a 200-record development set, an 800-record public test set, 992 public auto-evaluable targets, fixed transcripts from a three-ASR ensemble, an automatic target scorer, and initial results for seven product TTS systems. We additionally report ASR-route diagnostics, ASR-subset ablations, category-level results, confidence intervals, and provider configuration metadata. The best system reaches 0.879 strict accuracy, while several systems remain below 0.60.

Humans effortlessly locate and identify objects by touch alone, even without vision. In contrast, robotic systems rely heavily on vision and struggle with autonomous tactile exploration and object identification. We present TACTFUL, a vision-free tactile exploration framework that enables a multi-fingered robot to autonomously explore confined workspaces, discover objects through contact, and identify them via tactile reconstruction. Trained entirely on real hardware without simulation, our system learns a single policy that balances global workspace exploration with local surface refinement through a dynamic reward schedule. Our results demonstrate that tactile sensing, when paired with structured learning, can serve as an effective primary modality for object-level reasoning, achieving 77% success with 0.015 m average reconstruction error and outperforming baseline approaches on real-world objects.

Physics-informed surrogate models can accelerate computational fluid dynamics simulations. However, many existing methods reproduce global flow patterns more reliably than localized multiscale structures. This study presents a physics-informed Fourier-wavelet transformer for next-step velocity-field reconstruction in real-world flow benchmarks. The proposed formulation combines hybrid Fourier-wavelet spectral encoding with physics-biased self-attention based on partial differential equation residual diagnostics. It also uses self-supervised pretraining through Masked Physics Prediction and Equation Consistency Prediction. The experiments are conducted on two real benchmark cases: cylinder-wake flow and fluid-structure interaction. All approaches are evaluated under a shared local protocol and compared with spectral, transformer-based, operator-learning, and physics-informed neural-network baselines. On the cylinder-wake benchmark, the proposed model achieves the best aggregate accuracy, with an all-channel normalized mean-squared error of 0.05875 and an all-channel Pearson correlation coefficient of 0.97019. On the fluid-structure-interaction benchmark, it gives the lowest all-channel normalized mean-squared error of $2.70 \times 10^{-4}$, compared with $4.02 \times 10^{-4}$ for the strongest baseline. Component-wise field comparisons and scale-separated diagnostics further show stronger recovery of localized wake structures, including near-body, wake-core, and far-wake features. The results demonstrate improved real-world flow reconstruction while maintaining a practical accuracy-cost tradeoff.

Large Language Models (LLMs) and LLM-based Multi-Agent Systems (MAS) are revolutionizing software engineering (SE) by advancing automation, decision-making, and knowledge processing. Their recent application to SE tasks has already shown promising results. In this paper, we focus on summarization as a key application area. We present Metagente, an LLM-based MAS designed to generate concise and accurate summaries of software documentation. Metagente employs a Teacher-Student architecture where multiple LLM agents collaborate to enhance relevance and precision of produced summaries. An empirical evaluation on real-world datasets demonstrates Metagente's effectiveness in streamlining workflows, outperforming the considered baselines. The evaluation provides evidence that Metagente improves summarization for requirements analysis and technical documentation. Our findings underscore the transformative potential of these technologies in SE, while identifying challenges and future research directions for their seamless integration.

Data preparation pipelines improve data quality in machine learning by transforming raw tables into learning-ready data through sequential cleaning and feature transformation operators. However, automatically constructing such pipelines is computationally difficult because operator sequences are combinatorial and end-to-end evaluation is expensive. Existing state-of-the-art (SOTA) Multi-DQN methods still face three key limitations: decoupled value estimators weaken long-horizon credit assignment, dataset context is only weakly injected into the policy, and exploration is inefficient in a sparse search space with many invalid states. To address these issues, we propose FlowPipe, a unified framework that formulates pipeline synthesis as conditional probabilistic flow generation over a directed acyclic graph. FlowPipe uses Conditional Generative Flow Networks (C-GFlowNets) with a Trajectory Balance objective to connect terminal validation rewards with early pipeline decisions. It further introduces Deep Semantic Modulation through Feature-wise Linear Modulation (FiLM), allowing LLM-derived logical priors to condition the policy's internal activations according to dataset semantics. In addition, FlowPipe incorporates failure awareness into the flow objective to avoid invalid states and concentrate search on high-potential regions. Experiments on two benchmark suites with 74 real-world datasets show that FlowPipe outperforms SOTA baselines, improving accuracy by 11.96% on average and achieving 12.5x faster training convergence. Source code is available at https://github.com/KunyuNi/FlowPipe.

In many decision-making scenarios, acquiring information incurs different costs. We consider the problem of constructing a deterministic evaluation strategy that minimizes the expected cost of evaluating a propositional formula under variable costs and a probability distribution over truth assignments. We present a branch-and-bound algorithm with variable-selection heuristics, pruning, and caching. To the best of our knowledge, it is the first practical exact algorithm for this level of generality. Experiments on random instances demonstrate scalability and quantify the efficiency-quality trade-off of a greedy beam-search variant. We additionally evaluate a structured heart-disease diagnosis instance. Finally, we prove that the problem is $\#P$-hard and contained in $\mathrm{PSPACE}$.

Large language models (LLMs) have shown strong potential for planning and sequential decision-making, but prior work often relies on using them as direct controllers, which requires precise action generation and can be unreliable in practice. This paper proposes Latent Action Guidance for Online Reinforcement Learning (LaGO), a framework that uses a pretrained LLM as a latent action prior to softly guide online policy optimization, rather than treating the LLM as an explicit planner or controller. Experiments on both a discrete-control benchmark, CLEVR-Robot, and a continuous-control benchmark, Meta-World, demonstrate that LaGO consistently improves both reward and success rate over Vanilla PPO. In particular, LaGO increases the average success rate from 15.1% to 27.2% on CLEVR-Robot and from 2.7% to 15.2% on Meta-World. Our analysis further shows that stronger pretrained LLMs provide more effective guidance, suggesting that LLM knowledge can improve planning and online decision-making.

Dense retrieval embedding models are a fundamental component of modern retrieval-based AI systems. Most dense retrievers are trained with contrastive objectives, which require labeled positive and negative document pairs that are often costly and difficult to obtain. In this work, we investigate whether the autoregressive next-token prediction objective of a large language model (LLM) can provide supervision for dense retrieval. The intuition is simple: if a document contains information relevant to a query, conditioning on that document should make the target output easier for the LLM to predict. A key challenge is that the next-token prediction loss is computed inside the LLM, while the retriever is a separate embedding model. To address this challenge, we propose DREAM (Dense Retrieval Embeddings via Autoregressive Modeling), which injects retriever-generated query-document similarity scores into selected attention heads of a frozen LLM. During training, these scores determine how much attention each candidate document receives while the LLM predicts the target output. The resulting prediction loss provides gradients for retriever training through the attention mechanism. We evaluate DREAM on retrieval benchmarks BEIR and RTEB using embedding backbones ranging from 0.5B to 3B parameters. DREAM consistently outperforms existing baselines across different model scales. These results demonstrate that DREAM provides a promising approach for training dense retrievers through autoregressive modeling.

Phase-field models play a central role in the continuum description of phase separation, in which the bulk free-energy density and the interfacial thickness parameter determine pattern formation and microstructural evolution. In practice, these constitutive quantities are rarely known a priori and must be inferred from limited dynamical observations. In this work, an extended pseudo-spectral physics-informed neural network (ESPINN) framework is developed for the inverse identification of phase-field models from transient snapshot data. It enables the simultaneous recovery of both the bulk chemical potential and unknown gradient coefficients. Numerical experiments on the one-dimensional Cahn-Hilliard equation demonstrate accurate and statistically stable reconstruction in the noiseless regime, with substantial constitutive information recoverable from even a single snapshot pair. In the presence of noise, reconstruction accuracy degrades gracefully, and increasing the number of snapshots improves robustness by reducing variance across runs. These results establish ESPINN as a data-efficient and physically consistent approach for learning free-energy structure in continuum models of phase separation.

The explosive growth and complexity of product data within the dynamic Brazilian e-commerce landscape demand robust and specialized methods for structured information extraction. Traditional approaches to Product Attribute Value Extraction (PAVE) often struggle with the linguistic nuances and sheer diversity of product descriptions in Portuguese. To address this critical gap, this paper introduces two major contributions. First, we present AI-PAVEBr, a specialized system engineered with Large Language Models (LLMs) to perform high-accuracy PAVE specifically for Brazilian e-commerce catalogs. Second, to facilitate reproducible research and provide a definitive benchmark, we introduce and share the Golden Set, a new, meticulously curated, and manually annotated dataset for PAVE in Portuguese. We detail the creation process and structure (Entity, Category, Subcategories) of this high-quality reference set. Our experiments conclusively show that AI-PAVE-Br, leveraging targeted prompt engineering, dramatically outperforms conventional Named Entity Recognition (NER) baselines. This work not only delivers a superior, scalable solution for a major non-English market but also enriches the NLP community with a valuable, publicly available resource for future PAVE research.

Large Audio-Language Models (LALMs) have been widely used as judge models for the automatic evaluation of generated speech. However, prior approaches predominantly focus on holistic naturalness, leaving fine-grained paralinguistic distinctions underexplored. We introduce ParaPairAudioBench, a pairwise benchmark of 5,175 audio pairs across five paralinguistic dimensions: Style, Rate, Emphasis, Age, and Gender. Our experiments show that current LALM judges still lag behind human judgments by 32%p on average and exhibit severe calibration failures, particularly in Tie cases where the correct decision is to abstain. To further analyze lexical versus acoustic reliance, the benchmark includes both same-transcript and cross-transcript conditions. ParaPairAudioBench enables multi-dimensional, calibration-aware assessment of the reliability of LALM-as-a-Judge for paralinguistic speech evaluation.

Millions of people use machine translation (MT) tools daily, yet little is known about their perception of what systems can and cannot do. This paper studies users' mental models of speech translation systems through a new framework based on cross-lingual question answering, where users either accept MT output or request professional re-translation to answer questions based on the information presented in a foreign language. By analyzing user behavior and accuracy trends across varying translation qualities, we examine to what extent they can predict where the system is likely to be wrong, and how this mental model evolves. Users develop stronger mental models with practice, especially when they have some knowledge of the source language, primarily by relying on surface-level error cues. Moreover, providing speech transcriptions can help users develop better mental models. Our results show the promise of cross-lingual question answering as a downstream task for studying MT mental models and advancing our understanding of human-AI collaboration.

Cinematographic captioning aims to describe how a video is filmed using professional film-language concepts such as camera movement, shot size, depth of field, composition, and shooting angle. This capability is important for fine-grained video understanding and controllable movie-quality video generation, yet remains underexplored in existing multimodal large language models. Unlike question-answering-based evaluation of cinematic understanding, cinematographic captioning requires a unified open-form description over multiple cinematographic dimensions. This task is challenging for two main reasons: the model must infer professional cinematographic concepts from subtle visual evidence, and it must generate captions that are both comprehensive and accurate. Accordingly, we propose CineCap, a framework that combines structured reasoning with spatio-temporal anchors and reinforcement learning with comprehensiveness, accuracy, and gated coverage rewards. The former grounds professional cinematographic descriptions in explicit visual evidence and organizes them into compact atomic reasoning for supervised fine-tuning, while the latter improves the balance between descriptive completeness and factual correctness. In addition, we construct CineCap Bench, a benchmark of 472 manually annotated video-caption pairs for systematic evaluation. Extensive experiments show that CineCap consistently outperforms strong proprietary and open-source baselines, establishing a new state of the art for cinematographic captioning. The code, model checkpoint, and benchmark are publicly available in https://github.com/Hectormxy/CineCap.git.

Traditional visual data storytelling relies on binary graphics that depict two simplified groups in conflict. This can increase political polarization by oversimplifying intra-group disagreements and erasing ambiguity and shared ideas or values. This can inadvertently foster "us versus them" thinking. Intentional, pluralistic design choices for AI-enabled digital platforms can produce visualizations that emphasize nuance, opinion distribution, and intergroup commonalities. To demonstrate this potential, we examine deliberative technologies that map high-dimensional opinion spaces and highlight areas of both consensus and dissensus. The paper highlights the We the People deliberation conducted by Jigsaw and the Napolitan Institute in September 2025, which engaged over 2,400 Americans across all 435 congressional districts in an AI-supported, asynchronous dialogue regarding freedom and equality. By utilizing AI to synthesize long-form, text-based participant inputs into interactive "opinion landscapes," the initiative provided an alternative format for pluralistic data storytelling that humanized diverse viewpoints and revealed hidden areas of substantial broad consensus. The paper concludes that shifting from divisive, contrast-heavy visual frameworks to distribution-focused, interactive models represents a highly scalable, low-cost intervention capable of bridging perceptual gaps and cultivating a more resilient, collaborative democratic culture.

Fact-checking systems built on LLMs achieve high verdict accuracy on standard benchmarks, yet routinely output Supports labels whose cited evidence does not license the claim. Structured decomposition is the natural way to inspect those warrants, but rigid extraction protocols strip the full-claim context that facets need. We introduce SIFT -- claim-conditioned re-scoring of extracted evidence spans against the full claim -- paired with WSP (Warranted Supports Proportion), an automatic NLI check that the cited warrant entails the claim. We evaluate on FEVER, SciFact, 5PILS, and DP across four open-source backbones. SIFT recovers accuracy on cells where naive decomposition costs up to 27.6 points, while raising WSP above direct prompting; WSP itself calibrates against human gold evidence at AUC 0.92 and precision 0.98.

As autonomous agents tackle increasingly complex multi-step, multi-agent tasks, their execution trajectories have scaled beyond the constraints of even the largest context windows. Current methods for effectively diagnosing agent failures load the full trajectory into an LLM's context window, which suffers from attention dilution and fails when agentic traces inevitably exceed context limits. To address this, we introduce SAFARI (Scaling long-horizon Agentic Fault AttRibution via active Investigation), a framework that replaces linear context loading with a tool-augmented diagnostic loop. By equipping LLMs with a specialized toolbox to read and search trajectory segments alongside a persistent Short-Term Memory (STM) for cross-turn reasoning, SAFARI effectively decouples diagnostic accuracy from architectural context limits. Our experiments demonstrate that SAFARI outperforms state-of-the-art results by 20% on the Who&When dataset within a 1M token budget, and by 19% on TRAIL GAIA subset on a 25K token budget. Most significantly, SAFARI maintains a 0.58 precision even when the target fault resides 5x beyond the model's native context window, a scenario where traditional evaluators fail entirely.

Class imbalance poses a significant challenge in classification, where existing methods such as SMOTE often generate low-quality synthetic samples in regions with noise or class overlap. We propose QC-SMOTE, a quality-controlled oversampling framework that estimates minority sample reliability using a composite neighbourhood trustworthiness score combining local density, safe-level, and isolation from the majority class. Synthetic candidates are generated using an IPQ-guided best-of-K strategy that evaluates midpoint purity and, when required, majority clearance, with allocation guided by sample reliability and boundary informativeness. Generation behaviour adapts across overlap--imbalance regimes, adjusting interpolation range and selection criteria to match local data geometry. Low-quality synthetic samples are replaced with original minority duplicates when neighbourhood purity falls below an adaptive threshold, providing graceful degradation by reverting to duplication in severely noisy regions. Experiments on 30 imbalanced datasets using repeated stratified cross-validation show that QC-SMOTE achieves the strongest average AUC-ROC and Macro F1 among the compared oversampling methods, with particularly clear gains under moderate and severe imbalance. These results demonstrate the importance of quality-aware, geometry-adaptive synthetic sampling for robust imbalanced classification.

Retrieval-Augmented Generation enhances large language models by incorporating external knowledge, but deploying it in sensitive scenarios risks privacy leakage via malicious prompts. To address this, we propose a multi-agent framework that sanitizes retrieved content through semantic rewriting. By employing three specialized agents for privacy extraction, semantic analysis, and reconstruction, our approach collaboratively removes sensitive identifiers while preserving the semantic core. We evaluate the framework on the ChatDoctor and Wiki-PII datasets across six large language models. Experimental results demonstrate a significant reduction in privacy leakage under targeted attacks. For instance, we reduced targeted information exposure in LLaMA-3-8B from 144 instances in the baseline to just 1. Furthermore, we maintain strong contextual fidelity with a BLEU-1 score of 0.122, outperforming the existing SAGE method's 0.117. Finally, the framework operates as an asynchronous preprocessing module, introducing no additional latency to online inference, as all rewriting is executed as a one-time offline preprocessing step. To promote reproducibility, the source code of this work is publicly available at https://github.com/foursoils/Privacy-Preserving-RAG.

Training safe Reinforcement Learning (RL) systems is inherently challenging, with no guarantee of avoiding unwanted behaviors. The most effective defenses against this are (i) transparency through explainability and (ii) alignment via human feedback. While both show promising results, no publicly available framework currently combines them. To address this, we introduce Themis, an XAI-enabled testing and evaluation framework for Reinforcement Learning from Human Feedback. Themis supports over 200 widely used environments and is easily configurable for experiments in RL, transparency, and alignment. Our results show that Themis can train reward models that match or outperform the environment's true reward signal using human preferences. We also provide a cloud-based platform for collecting human feedback and managing experiments. It is user-friendly, auto-scalable, and supports large participant groups across multiple experiments without extra development overhead. Tests show Themis can support one thousand users in back-to-back experiments on a modest commercial machine.

This paper introduces a categorical account of infinitesimal causality in Frobenius Markov categories equipped with tangent-bundle semantics. IDC captures the infinitesimal layer in which interventions act as tangent deformations of copy/discard structure. Two distinct Frobenius structures interact: (1) the categorical Frobenius algebra on classical variables encoding copying, comparing, and discarding; and (2) the geometric Frobenius integrability condition, namely involutive closure of the intervention distribution, distinct from the algebraic Frobenius structure. Categorical causal sufficiency is defined as the compatibility of these two notions. A key observation is that, for structural causal models, infinitesimal causality is most naturally formulated in the slice of deterministic mechanisms over exogenous variables, with visible stochastic kernels obtained only after pushforward. Interventions are tangent vectors that deform the Frobenius copy/discard operations; their Lie brackets measure whether this deformation preserves classical information-flow structure. Pearl's do-calculus is used as a guiding example of intervention identities: ignoring irrelevant interventions corresponds to counit invariance, action/observation exchange to coproduct compatibility with pushforward, and independence to involutive bracket closure of the visible intervention distribution.

Competency Questions (CQs) are the central component of CQ-verification, an established process in which an ontology is evaluated against a set of natural language questions to determine whether the intended purpose of the ontology has been properly modelled. However, CQ-verification is often time-consuming and error-prone, as it requires careful interpretation of linguistic nuances and precise alignment with formal ontology constructs. Ambiguities and complexity in CQs can further complicate this process, leading to inconsistent modelling decisions and verification outcomes. In this paper, we investigate what makes a CQ challenging and possible solutions to enhance the users' performance in the CQ-verification process. We experimented with the data of 19 participants who performed CQ-verification on 20 tasks using an LLM assistant to support ontology evaluation. The results show the necessity of a tool to refine CQs before publishing them to avoid ambiguity or excessive complexity in later phases of the ontology engineering process.

In ontology-based data access (OBDA), multiple data sources are integrated via mappings to an ontology. We consider an OBDA setting based on existential rules and the certain answer semantics. We address the recent issue of query abstraction, which consists of abstracting data queries by translating them to the ontology layer. Since a perfect abstraction may not exist, the notions of minimally complete and maximally sound abstractions have been introduced. We study abstractions within an extension of UCQs with a limited form of inequality and a special predicate marking database constants. While this extension does not lead to an increased complexity of the problems of interest, it is able to express minimally complete abstractions, hence perfect abstractions when they exist. We also characterize maximally sound abstractions by making a new connection with the notion of maximum recovery stemming from data exchange.

The evaluation of cultural grounding context becomes complex when multiple cultures convey the same moral lesson. This challenge is particularly relevant to large language models (LLMs), which produce narratives across a wide range of languages and cultural contexts. However, it remains uncertain whether these models preserve culturally grounded meaning when equivalent moral lessons are conveyed through distinct cultural forms. This study introduces a multilingual evaluation narrative framework that integrates a cross-linguistic collection of 414 proverbs spanning 15 languages and uses four LLMs to generate 13k narratives. By employing semantically equivalent proverbs as culturally grounded prompts, the analysis assesses whether models preserve meaning across languages, how cross-lingual conditioning influences narrative realization, and whether different model families converge on similar interpretations. Results indicate that cross-lingual prompting largely preserves proverb-level semantic meaning while systematically redistributing agency, social positioning, and narrative structure. Additionally, strong inter-model convergence is observed in both monolingual and cross-lingual settings, suggesting that multilingual LLMs rely on shared semantic abstractions despite architectural and linguistic differences. These findings shed light on the need for more comprehensive evaluations of cultural grounding. Relying exclusively on semantic similarity in multilingual narrative assessments may overestimate cultural preservation by neglecting culturally meaningful variations in narrative expression.

Accurate user modeling often depends on rich interaction histories, which are unavailable for billions of low-activity users. Large Language Models (LLMs) can infer latent user states from static profiles, but this reasoning becomes unreliable when profiles are sparse, and applying an LLM to billions of users is prohibitively expensive. We present ScaleToT, which learns structured reasoning from a small LLM-processed subset and extends it to the broader low-activity user population. To improve reasoning reliability, ScaleToT constructs typed user-state chains with a bounded entropy-guided Tree-of-Thought (ToT) refinement procedure. To make this structured reasoning usable from sparse profiles, the teacher-curated chains are used to train a student model on static profiles through supervised fine-tuning (SFT) and Outcome-Driven Segment-Aware Implicit Reward Policy Optimization (OSIPO). ScaleToT then transfers the student's reasoning representations to a lightweight profile encoder, providing shared reasoning signals for the remaining users without LLM inference. We evaluate ScaleToT on lifetime value (LTV) prediction in a billion-scale advertising deployment. A randomized online A/B test increased LT30 by 6.738\%, while offline reasoning covered only 7.32\% of the potential population, greatly reducing compute cost compared with full-population reasoning.

Longitudinal modelling of Alzheimer's disease progression is clinically useful only if it can describe not just the most likely next diagnosis, but how a patient may evolve over time and how reliable that forecast is. Most deep learning approaches reduce this problem to single-step classification, treating cognitively normal, mild cognitive impairment, and dementia as flat categories while providing limited insight into how uncertainty accumulates across future visits. We propose a probabilistic framework that combines ordinal diagnosis prediction, multi-horizon trajectory generation, and decomposed uncertainty estimation. A Temporal Fusion Transformer encoder is adapted with a CORAL ordinal output layer, asymmetric loss weighting, and converter oversampling to respect disease-stage ordering and improve sensitivity to MCI-to-dementia transitions. Conditioned on the learned patient-context representation, an autoregressive Mixture Density Network generates five-year probabilistic trajectories for diagnosis state, CDR Sum of Boxes, MMSE orientation, and hippocampal volume. On ADNI, the model outperforms linear, recurrent, and transformer baselines for next-visit diagnosis prediction, with the strongest gains on MCI-versus-dementia discrimination. Generated trajectories achieve near-nominal 90% credible interval coverage, widening uncertainty across the forecast horizon, and biomarker dynamics consistent with expected Alzheimer's disease progression. We further separate aleatoric from epistemic uncertainty using analytic mixture variance and a five-member bootstrap ensemble, which provides the strongest encoder diversity and output-level epistemic signal. Epistemic uncertainty is higher for rare progression archetypes, MCI and dementia patients, and under external evaluation on OASIS-3, where it increases alongside prediction error.

Multi-agent reinforcement learning (MARL) addresses the problem of training multiple agents that pursue collaborative, competitive, or mixed objectives. Prior work has investigated transfer learning between source and target domains in MARL; however, the majority of existing approaches impose the constraint that the dimensionalities of the observation space and the global state space must be identical across domains. In this paper, we introduce a method that explicitly accommodates mismatched state-space dimensionalities between source and target domains. The proposed approach, ASALT, incorporates both observation-level and state-level adapters that map the target-domain observations and global states into a shared embedding space, thereby enabling more effective transfer of knowledge across both actors and critics. These adapters can generate embeddings that support efficient strategy transfer across heterogeneous domains. Experimental results on multiple configurations in standard benchmark environments demonstrate that ASALT surpasses existing baselines in terms of sample efficiency and global return in cooperative settings, but its effectiveness depends on the degree of mismatch between source and target domains. Furthermore, our findings indicate that ASALT mitigates negative transfer, which frequently constitutes a major obstacle when transferring policies between domains with differing observation and action spaces.

A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning. In this work, we investigate how world modeling based on language models can further push the boundaries of general agents. (i) We first focus on building foundation models for agentic environment simulation. We introduce Qwen-AgentWorld-35B-A3B and Qwen-AgentWorld-397B-A17B, the first language world models capable of simulating agentic environments covering 7 domains via long chain-of-thought reasoning. Leveraging more than 10M environment interaction trajectories of 7 domains in real-world environments, we develop Qwen-AgentWorld through a three-stage training pipeline: CPT injects general-purpose world modeling capabilities from the state transition dynamics and augmented professional corpora, SFT activates next-state-prediction reasoning, and RL sharpens simulation fidelity through a tailored framework with hybrid rubric-and-rule rewards. To evaluate language world models, we present AgentWorldBench, a comprehensive benchmark constructed from real-world interactions of 5 frontier models on 9 established benchmarks. Empirical results demonstrate that Qwen-AgentWorld significantly outperforms existing frontier models. (ii) Beyond foundation models, we further investigate two complementary paradigms through which world modeling enhances general agents. First, as a decoupled environment simulator, Qwen-AgentWorld supports scalable and controllable simulation of thousands of real-world environments for agentic RL, yielding gains that surpass real-environment training alone. Second, as a unified agent foundation model, world-model training acts as a highly effective warm-up that improves downstream performance across 7 agentic benchmarks. Code: https://github.com/QwenLM/Qwen-AgentWorld

As Large Language Models are increasingly deployed in critical applications, robustly evaluating their social biases is paramount. However, the current literature suffers from widespread methodological fragmentation, which yields contradictory conclusions. This stems largely from ignoring the structural framing of benchmark-level evaluations. To resolve this, we introduce a unified and controllable framework that standardizes heterogeneous benchmarks to systematically contrast isolated demographic assessments with forced-choice comparative settings. Crucially, this allows us to disentangle the confounding effects of Chain-of-Thought reasoning, neutral fallback options, and other structural artifacts in social bias evaluations. Our evaluation across multiple model families reveals a massive, systematic paradigm gap: while isolated assessments limit prejudice activation, comparative settings act as aggressive catalysts for latent discrimination, a shift primarily driven by underspecified contexts. Alarmingly, CoT reasoning exacerbates social biases under comparative settings, and this systemic bias persists as a deterministic prejudice even when models are provided neutral fallback options or claim to answer randomly. Finally, we demonstrate that this comparative prejudice is a generalized phenomenon that scales positively with model size. Ultimately, we offer a crucial methodological guideline: while researchers must leverage comparative settings to robustly audit hidden biases, practitioners cannot safely rely on comparative deployments in ambiguous real-world tasks.

Long-term memory promises LLM agents that grow more capable across sessions, maintaining an accurate, evolving understanding of the user that interaction forms. In practice, however, this memory is evaluated mostly through downstream behavior, such as later answers, personalization quality, or task success, which tests that understanding only indirectly and leaves the memory artifact itself largely unaudited. We argue that long-term memory should instead be evaluated as an auditable post-interaction artifact: after ordinary assistance, what structured user state can be reconstructed from the memory the agent leaves behind? We instantiate this view in MEMPROBE, a benchmark in which a memory-equipped agent assists simulated users, each carrying a hidden, taxonomy-anchored user-state bank, across a trajectory of leak-controlled tasks, after which that bank is reconstructed from the agent's resulting memory under both full-store and top-k access. Built on synthetic ground truth for efficient, scalable measurement, MEMPROBE spans 50 simulated users with 31 hidden dimensions each (1,550 recovery targets) and tests 5 representative memory systems. Testing state-of-the-art memory agents, we find that successful assistance and recoverable memory behave as distinct capabilities. Task completion nearly saturates, even for a memoryless baseline, while category-balanced recovery stays moderate (about 0.6) and drops further under top-k retrieval. MEMPROBE is the first benchmark to study memory recovery directly, reconstructing the user state a system retains and scoring it against ground truth. We see recovery as a concrete objective for future memory agents to optimize, and MEMPROBE as a step toward an environment where agents are trained to remember their users, growing more faithful the longer they know them.

Scaling adversarial evaluation of large language models requires both a method for generating hard inputs and a reliable way to confirm that resulting failures are real. We present AdversaBench, an end-to-end red-teaming pipeline that mutates seed prompts with five structured operators, queries a target model, and confirms failures through a three-judge panel with a meta-judge tiebreaker. We report experiments on 45 seeds across three categories: reasoning, instruction-following, and tool use. Every seed produced a confirmed failure. Four findings stand out. First, operator effectiveness varies sharply by category: inject_distractor scores 0.00 mean reward on instruction-following seeds but 0.80-0.83 on reasoning and tool-use. Second, binary failure rate hides difficulty: instruction-following seeds required 2.4 attacker iterations on average versus 1.1 for other categories, a gap visible in survival curves. Third, pairwise judge agreement of 80-87% coexists with near-zero Cohen's kappa due to label skew; category-level disagreement rates are more informative. Fourth, adversarial prompts generated against Llama 3.1 8B transfer zero-shot to Llama 3.3 70B, suggesting the mutations exploit general behavioral patterns rather than model-specific weaknesses. Code, dataset, and analysis scripts are available at https://github.com/khanak0509/AdversaBench .

Deep learning approaches to biometric verification are commonly trained by optimizing indirect objectives, creating a misalignment between the optimization process and the primary evaluation metric, typically the Equal Error Rate (EER). This paper introduces EERLoss: a subdifferentiable, arbitrarily accurate approximation to EER for training deep biometric models. Furthermore, this framework has the potential to be adapted to optimize any specific operating point on the DET curve, enhancing its generalizability. To validate this approach, EERLoss is evaluated on a particularly demanding behavioral biometric modality: keystroke dynamics verification. This task is characterized by its high intra-class and low inter-class variability. Experiments are conducted on the large-scale KVC-onGoing benchmark, incorporating data from over 185,000 subjects across different scenarios. A comprehensive ablation study initially demonstrates the superiority of EERLoss in comparison to existing state-of-the-art loss functions. It also converges substantially faster compared to other losses, reducing the overall training cost. Additionally, a comparison is made between the proposed loss and the KVC-winning architecture by re-training it with EERLoss, demonstrating that the proposed approach significantly outperforms the original SoTA, achieving a relative EER reduction of up to approx. 30\%. This improvement on a challenging, large-scale benchmark validates the effectiveness of EERLoss as a task-aligned training objective specifically suited for high-variance biometric traits.

While the validity of LLMs' use in the legal context remains subject to ethical and legal debate, legal professionals are already experimenting with personal LLMs, if only for translation and reformulation. However, even such a seemingly innocuous use can introduce biases through case processing speed if LLM assistants selectively refuse assistance on certain topics. To better anticipate such biases, we investigate several modern small LLMs that are most likely to be used as on-device assistants, to assess the impact of overrefusal on legal prompts. Surprisingly, we find that authority-style prefixes (``you are acting as an assistant of the national supreme court'', ``[...] defense lawyer'') systematically increase refusal rates by 2--20x over the no-prefix baseline, while a known role-play jailbreak prefix shows mixed effects, sharply increasing refusals in some models and barely shifting them in others. The finding suggests that small on-prem deployable LLMs are unstable under contextual framings that a real institutional user might naturally introduce, and further investigation is essential to minimize opportunities for bias.

Parametric knowledge in Large Language Models is not equally accessible across languages. As a result, standard inference techniques often struggle to surface localized facts, leading to failures in cross-lingual knowledge transfer and consistency. In this work, we investigate techniques for accessing hidden factual knowledge by exploring cross-lingual prompting strategies. We identify four inherent dimensions of cross-lingual exploration that directly govern parametric knowledge retrieval and evaluate them on multilingual factual benchmarks covering 17 typologically diverse languages. Our results demonstrate that cross-lingual exploration significantly improves knowledge transfer and factual recall, representing a more efficient compute Pareto frontier than native-language scaling. Furthermore, we observe corresponding improvements in cross-lingual consistency, exceeding what can be explained by accuracy gains alone. Overall, our work establishes multilingual prompt exploration as a highly effective inference-time strategy for unlocking latent parametric knowledge.

Modern finance relies heavily on complex machine learning models to find patterns in the stock market. However, as these AI models get more complicated, they often memorize short-term market noise instead of finding companies with real, lasting value. We designed this research to test if Benjamin Graham's classic value investing rules could act as a mathematical "low-pass filter" to keep these modern models in check. We built three different sets of features - pure Graham rules, modern market factors, and a mix of both - and tested them against highly complex models (XGBoost and AutoGluon) using 20 years of S&P 500 data. By applying a strict buy-and-hold strategy over a four-year test period (March 2022 to March 2026), the results showed that more complex algorithms do not always win. While the AutoGluon model captured high returns (222.68%), it suffered a substantial 39.78% drop because it bought volatile tech stocks right before the market crashed. On the other hand, the pure Graham Random Forest achieved the highest overall return (232.13%) with much less risk (1.38 Calmar Ratio). Furthermore, the Combined Random Forest successfully mixed momentum with Graham's rules, making a 202.91% return while keeping the lowest maximum drop (34.53%) of any model tested. Ultimately, this research proves that Graham's "margin of safety" isn't outdated; it is actually a highly effective way to prevent modern AI from taking on too much risk.

Large Language Models (LLMs) are traditionally viewed as autoregressive generators. However, from the perspective of collective computation, they function as high-dimensional Dense Associative Memories that store complex reasoning patterns as latent attractors. In this work, we investigate the energy landscape of mathematical reasoning. We posit that correct reasoning chains correspond to deep, wide attractor basins ("flat minima") in the model's output distribution, whereas hallucinations manifest as sharp, unstable local minima. To exploit this geometry, we introduce a retrieval mechanism based on a Gibbs measure of the trajectory's spectral entropy. By sampling multiple reasoning paths and weighting them by their inverse energy ($P \propto e^{-βE}$), we approximate the equilibrium distribution of the associative memory, effectively ``relaxing'' the system into a robust solution. Empirically, this physics-inspired mechanism improves Microsoft Phi-3.5 performance on GSM8K by 5.38\% (84.7\% $\to$ 90.1\%), demonstrating that inference is better modeled as a dynamic settling process into an attractor basin rather than greedy next-token prediction.

Multi-agent LLM environments require robust mechanisms for shared knowledge management. This paper formalizes the fleet-memory problem and identifies four foundational failure modes: unauthorized leakage, stale propagation, contradiction persistence, and provenance collapse. To address these, we define explicit systems-level primitives: scoped retrieval, temporal supersession, provenance tracking, and policy-governed memory propagation. These primitives are implemented in MemClaw, a production multi-tenant memory service, and evaluated via ArgusFleet, a reproducible harness testing four governance dimensions. Rather than a baseline comparison, this study measures a live production service, emphasizing real-world architectural insights and negative results. Key Evaluation Results Provenance: Successfully reconstructed 100% of depth-four derivation chains with correct writer identity at sub-second per-hop latency. Propagation: Demonstrated high intra-fleet visibility with zero cross-fleet leakage. Under strong write mode, write-to-visible latency was optimized to a single search round-trip. Production Architectural Issues Discovered Asymmetric Scope Enforcement: Tenant isolation held, but sub-tenant scope was initially bypassed on direct GET-by-id requests for agent-scoped credentials (disclosed and remediated during the study). Pipeline Ordering Conflict: While contradiction supersession works for admitted writes, a synchronous near-duplicate gate can prematurely reject contradictory writes before the asynchronous contradiction detector can evaluate them. Conclusion: Long-context retrieval alone is insufficient for production multi-agent memory. Governed shared memory demands explicit systems-level abstractions, and live evaluation is vital to expose enforcement and pipeline-ordering failures missed by design-only treatments.

We introduce NatureBench, a cross-discipline benchmark of 90 tasks distilled from peer-reviewed Nature-family publications, designed to evaluate whether AI coding agents can move beyond reproduction toward discovery on real scientific problems. NatureBench is built on NatureGym, an automated pipeline that constructs a standardized, per-task containerized environment from a source paper, addressing the environment-fragmentation problem that has limited the credibility of prior agent-on-research benchmarks. Evaluating ten frontier agent configurations under a strict web-search-disabled protocol, we find that the strongest model surpasses SOTA on only 17.8% of tasks under the g>0.1 criterion. Analysis of method pathways reveals that agents succeed primarily through methodological translation, converting scientific tasks into familiar supervised prediction problems, rather than through genuine scientific invention. Failures are dominated by wrong method choice and insufficient compute budget, not by task misunderstanding. We release the benchmark, the NatureGym pipeline, and a public leaderboard with maintainer-side reproduction. Code: https://github.com/FrontisAI/NatureBench

Large language models are increasingly deployed as agents that reason over documents rather than answer from parametric knowledge. We study archive-grounded reasoning: locating sparse evidence across a large, messy collection of workplace files, reconciling inconsistent terminology, units, and time conventions, and computing an answer. Existing benchmarks address only parts of this setting and none jointly stresses archive-groundedness, agentic exploration, and cross-domain coverage. We introduce Agora, a benchmark pairing 362 questions with eight domain collections of 9,664 authentic documents and 372M tokens, far exceeding any model's context window, so agents must explore deliberately rather than scan exhaustively. Agora is built by an agentic pipeline combining cross-document task synthesis, leakage-preventing obfuscation, and difficulty filtering. Evaluating eight models, we find the task far from solved: even the strongest reaches only 59.4% accuracy, with notable variation across domains.

Scam phone calls exploit vulnerable communities worldwide, yet research on detection has focused almost exclusively on English and other high-resource languages. In low-resource settings such as Turkish, detection is especially difficult, as annotated data is scarce and technological defenses remain limited. This research investigates how large language models (LLMs) can support scam detection in Turkish by introducing the first public multi-modal dataset of 100 aligned audio-transcript pairs of scam and benign conversations. We evaluate seven LLMs spanning three model families: Gemini 2.5 (Flash, Flash-Lite, Pro), GPT-4o, and Qwen (Max, Plus, Turbo), under three input conditions: raw audio, automatic speech-to-text transcripts, and transcripts refined by a native speaker. Our results suggest that transcript-based inputs consistently outperform direct audio processing, while human-corrected and uncorrected transcripts perform comparably. By centering a low-resource language and real world threat, this work highlights the urgent need for culturally and linguistically inclusive AI safety research and more robust multi-modal systems for fraud prevention.

Computer-Use Agents (CUAs) execute high-level user goals by perceiving and acting directly within graphical user interfaces. However, reinforcement learning for CUAs remains difficult because open-ended desktop environments rarely provide scalable, machine-readable reward signals: task success is often visually grounded and hard to specify with handcrafted reward functions or dense manual labels. We propose an RL fine-tuning framework that uses autonomous vision-language evaluation as a scalable supervision signal for GUI agents. Given a final screenshot and the original instruction, a Vision-Language Model judges task completion and provides terminal feedback without task-specific heuristics or manual labels during policy optimization. Because autonomous evaluators are imperfect, we model their feedback as a noisy binary reward channel and derive a noise-corrected reward estimator for Proximal Policy Optimization. Experiments across macOSWorld, Windows Agent Arena, and OSWorld show that corrected evaluator rewards outperform both zero-shot baselines and raw evaluator rewards, improving success rates by an average of 12.6 percentage points over zero-shot performance and 5.1 points over raw evaluator fine-tuning. These results suggest that autonomous evaluation can serve as a practical reward signal for RL in GUI environments when evaluator noise is explicitly modeled and corrected.

Rare diseases affect millions of individuals worldwide, yet timely diagnosis remains a major public health challenge due to scarcity of specialized clinical expertise. While large language models (LLMs) show promise to support rare disease diagnosis, current models are constrained by insufficient clinical deployability, limited clinically grounded evidence, and scarcity of training data. Here we present RaDaR (Rare Disease navigatoR), an open-source, compact reasoning LLM (32B parameters) for rare disease diagnosis. RaDaR was trained with 49,170 publicly available free-text cases and 104,666 synthetic cases with reasoning-enhanced training. RaDaR showed the strongest performance among evaluated open-source models, including the 671B DeepSeek-R1, across public benchmarks and four external validation centers. In a retrospective cohort, RaDaR prioritized the final diagnosis before documented clinical suspicion in 61.06 percent of cases, corresponding to a potential lead time of 1.87 months and 50.18 percent of the within-center interval. In a randomized physician-assistance trial, RaDaR assistance improved physicians' rare-disease diagnostic accuracy by 21.44 percentage points compared with internet search alone. Synthetic-data ablations suggested that phenotype-anchored narratives provide useful training signal for long-tail rare diseases, with a monotonic scaling trend within the tested data range. Together, RaDaR and its development and validation framework provide a deployable rare-disease reasoning model and a reproducible development framework for diagnostic AI under data scarcity.

Due to the wide use of graph-structured data in different fields of industry and science, the development of Graph Foundation Models (GFMs) has recently attracted a lot of attention. While many different types of models are called GFMs, particular interest has been paid to GFMs designed for node property prediction tasks, which is one of the most popular settings in Graph ML with lots of real-world applications from fraud detection in financial and social networks to recommendation systems for e-commerce and user-generated content platforms. While a number of GFMs for this task have been recently proposed, the field has not converged to a unified evaluation setting, and different works evaluate their models in widely different ways, preventing reliable comparison of GFMs with each other and with other types of models. In this work, we conduct a fair and rigorous reevaluation of 9 recent GFMs for node property prediction, comparing them to strong Graph Neural Network (GNN) baselines. We find that, among these GFMs, only the most recent ones based on the Prior-data Fitted Networks paradigm outperform well-tuned GNNs in predictive performance, although at a higher inference cost.

Emerging LLM services increasingly host many sparse MoE models, yet most models receive sparse requests and remain cold. This creates a GPU memory problem: model weights are stable and model-determined, while KV-cache is transient and demand-determined. Because cold models rarely reach peak KV-cache demand at the same time, reserving worst-case KV capacity per model wastes memory; a shared KV-cache pool can instead provision aggregate active demand. However, KV-cache sharing is not sufficient when weights and KV-cache remain in a monolithic GPU memory pool. Static weights compete with dynamic KV-cache, and KV-head-limited attention under cold, low-concurrency traffic exposes only a fraction of replicated KV capacity, leading to low GPU memory utilization and weak long-context support. We present CrossPool, a serving engine for cold MoE models that separates FFN weights and KV-cache into two GPU memory pools: a weights pool that consolidates FFN weights across cold models, and a KV-cache pool that dynamically serves active requests while keeping attention local to KV-cache. CrossPool combines a KV-cache planner and virtualizer, a layer-wise pipeline scheduler that hides hidden-state transfers, and persistent kernels with control lowering to reduce CPU-GPU control overhead. With efficient GPU memory pooling, CrossPool underpins bursty long-context requests and outperforms the state-of-the-art kvcached-based multi-LLM serving system, reducing P99 TBT by up to $10.4\times$.

We discuss reasons why the scaling exponents of current Large Language Models (LLMs) applications are indicating an unsustainable regime in terms of energy resources. We further show that attributing the smallness of such exponents to a numerical bias due to the neglect of a non-zero value of the loss function in the limit of infinite data (``pedestal effect") does not remove the unsustainability issue. Finally, the effects of the smoothness (roughness) of the data on the scaling exponents is commented upon based on an analogy with phenomenological models of fluid turbulence.

This paper describes UOL@IDEM's closed-track submission to the BEA 2026 shared task on L1-aware vocabulary difficulty prediction. We model the task as regression and train separate systems for Spanish, German, and Mandarin Chinese\footnote{Below we use \emph{Chinese} for brevity.}. Our system combines multilingual contextual representations with engineered features capturing frequency, surface form, retrieval evidence, semantic alignment, cognate similarity, and masked-language-model predictability. Development results show consistent gains over the official closed-track baselines, with sentence-embedding encoders such as BGE-M3, multilingual E5, and LaBSE performing best. Official submissions achieve RMSE scores of 1.132, 1.037, and 0.891 for Spanish, German, and Chinese, respectively. Feature analysis identifies frequency as the most stable predictor, while contextual predictability, form similarity, retrieval, and semantic features provide complementary L1-sensitive signals. Error analysis shows strong ranking performance but weaker calibration for the easiest items, which are often overpredicted. See https://github.com/Nouran-Khallaf/UoL-IDEM-BEA2026-Vocabulary-Difficulty-Prediction

The use of agentic systems to perform offensive security operations has moved from a theoretical possibility to a commoditized capability. However, while the community has focused on creating more and more capable agents, less attention has been allocated to assessing the security of those systems. In this work, we present the first in-depth security analysis of the most widely used agentic systems for offensive security operations. We show that most of these tools share common design flaws that enable an active adversary to exfiltrate API keys, establish persistent footholds, and fully compromise the operator's machine, even when the agent operates inside a sandboxed container. To support our analysis, we introduce a full cyber kill chain for such agentic systems, capturing the progression from initial LLM manipulation to lateral movement, persistence, guardrail bypass, and sandbox escape. Building on our security analysis, we derive a robust architecture for agentic offensive-security tools and propose actionable, broadly applicable design principles that mitigate the disclosed attack paths at the architectural level.

Learning causal models from fragmented biomedical data is challenging because clinical, molecular, and imaging variables are often incomplete or not jointly observed. We propose RetiSEM, a domain-constrained structural equation modelling (SEM) framework for causal graph recovery and mediation analysis under limited multimodal resources. This proposed work organises variables into biologically informed blocks, applies forbidden-edge constraints, and decomposes pathway-level effects into TE, NDE, and NIE components. We evaluate RetiSEM across ten synthetic benchmark scenarios that vary in dimensionality, nonlinearity, causal depth, and pathway structure, together with a fragmented real-world setting that combines NHANES clinical variables with externally derived retinal representations. This approach achieves lower structural error and higher causal accuracy than unconstrained baselines across the synthetic benchmarks. In the real-data analysis, retinal variables behave mainly as downstream biomarker-like indicators, with smaller but detectable indirect effects. These findings support our strategy as an interpretable framework for testing structured causal hypotheses in limited-resource biomedical AI. The code and resources for this work are publicly available at: https://github.com/Inamullah-Colab/ReitSEM.

The deployment of unmanned aerial vehicles (UAV) as open radio units (O-RUs) in 6G cellular systems presents a promising opportunity to achieve scalable and adaptive network coverage. However, optimizing UAV trajectories in dynamic and unfamiliar environments remains a critical challenge, particularly due to the need for extensive retraining in each new scenario. In this paper, we introduce a novel UAV trajectory optimization framework that integrates enhanced continual transfer learning within the O-RAN architecture. The proposed system maintains a library of pre-trained models and employs a model selection mechanism to identify and transfer knowledge from the most relevant environments, minimizing adaptation time and improving efficiency. When no sufficiently similar model is available, a fallback model empowered by continuous refinements ensures baseline performance. The framework leverages real-world city maps and ray tracing techniques to enhance learning reliability and improve trajectory planning. Simulation results demonstrate that the proposed model selection-based transfer learning approach reduces convergence time by 44% to 56% compared to retraining from scratch, and up to 40% compared to traditional transfer learning without model selection.

Video large language models (LLMs) are often constrained by computation and memory budgets, leading them to use reduced frame rates and spatial resolutions, which may cause them to miss critical information for question answering (QA). A practical and efficient solution is a two-stage paradigm: first perform coarse video understanding to localize relevant segments, and then re-watch these segments at higher temporal or spatial fidelity. In this paper, we present video-SALMONN-R$^3$, the first end-to-end video-LLM that enables re-watch through reinforcement learning without relying on chain-of-thought (CoT) cold-start. This design removes the need for costly CoT data annotations and avoids CoT-based supervised fine-tuning (SFT), which can otherwise degrade the pretrained video understanding abilities. To address the mismatch between the reasoning-first behavior induced by re-watch and the answer-first tendency of pretrained video-LLMs, we propose a re-answer strategy, in which the model first produces a direct answer in the first watch and then refines it after re-watching. Finally, to improve question adherence during re-watching, we propose a re-ask mechanism that re-injects the query when revisiting localized segments. Experimental results show that video-SALMONN-R$^3$ consistently outperforms both the base model and the QA-SFT baseline, while surpassing prior re-watch-based approaches with significantly lower computational cost. Code, models, and data will be publicly released upon acceptance.

Vision-language-action (VLA) models have made rapid progress in generalist robot manipulation by harnessing semantic knowledge from pretrained vision-language backbones, but their visual tokens remain grounded in 2D image coordinates rather than the calibrated geometry of the robot's cameras -- a mismatch especially pronounced in multi-camera setups, where views are coupled by known intrinsics and extrinsics yet processed as independent images. We propose G$^3$VLA, a camera-aware geometric module that injects calibrated structure into the visual-token stream of a pretrained VLA without altering its action space or imitation objective, combining intrinsic-conditioned ray embeddings, projective positional encoding (PRoPE), and bidirectional cross-view fusion. Geometric supervision is provided either from ground-truth point maps when available, or from confidence-gated $π^3$X teacher predictions, requiring no depth sensors or manual annotations. Instantiated on $π_0$, G$^3$VLA yields consistent gains across the LIBERO suites, RoboCasa24, RoboTwin2.0, and real-robot settings, with the largest improvements on spatially and object-sensitive tasks. We further validate on $π_{0.5}$ and GR00T 1.5, with results suggesting that geometric transfer is most effective when geometry-aware tokens have direct access to the action generation pathway. Our project page is at https://sites.google.com/view/g3vla

A real-time agent for general computer use - with games as the most demanding case - must act within tens of milliseconds while still planning over seconds. These two regimes sit at opposite ends of the latency-quality tradeoff. A reasoning VLM (Qwen3-VL-8B-Thinking) deliberates effectively but requires ~1.5 s per response - far too slow for a 15 Hz control loop. In contrast, a reactive VLM (MiniCPM-o 4.5) acts in milliseconds but underperforms on planning-heavy tasks. We couple two frozen models of matched scale (9B reactive, 8B reasoning), leaving the communication channel as the sole trainable component. The standard coupling is a Text Bridge (T): the slow model writes a suffix the fast model reads. We introduce a learned continuous Latent Bridge (L) that projects the slow model's residuals into the fast model's input-embedding space in a LLaVA-style manner, avoiding any text round-trip; both are compared against Fast-Only (F). On 7 Atari games and a driving domain (MetaDrive), tuning the action decoder per channel on held-out seeds, the Latent Bridge matches or beats the Text Bridge in every domain: it significantly improves two games (MsPacman +57%, RoadRunner +28%) and is a safe drop-in elsewhere. Combining both channels interferes destructively (RoadRunner -96%), so only one should be used. The benefit is highly predictable: the bridge helps if and only if slow reasoning already beats fast reaction (T > F) - the Latent and Text gains over Fast-Only move together at r=0.93. MetaDrive is the controlled negative, where the Latent Bridge is demonstrably inert because the Text Bridge adds no value. We release replay recordings and reproducible pipelines.

Long-context large language model (LLM) inference is increasingly constrained by the memory footprint and decoding cost of key-value (KV) caches, limiting sustainable deployment on resource-constrained hardware. Existing KV cache eviction methods typically apply heuristic token scoring over all heads in GQA-based LLMs. These methods ignore the different functionalities of attention heads, leading to the eviction of critical tokens and thus degrading the performance of LLMs. To address this issue, we propose CompressKV, a resource-efficient KV-cache compression framework for GQA-based LLMs. Instead of aggregating attention scores from all heads, CompressKV identifies Semantic Retrieval Heads (SRHs) that capture both the initial and final tokens of a prompt and semantically important mid-context evidence, and uses them to select tokens whose KV pairs should be retained. Furthermore, CompressKV allocates cache budgets across layers according to offline estimates of layer-wise eviction error. Experiments on LongBench and Needle-in-a-Haystack show that CompressKV consistently outperforms existing KV-cache eviction methods across memory budgets. Notably, it preserves over 97\% of full-cache performance using only 3\% of the KV cache on LongBench question-answering tasks and achieves 90\% accuracy with just 0.7\% KV storage on Needle-in-a-Haystack. These results demonstrate an improved resource--performance trade-off for long-context LLM inference. Our code is publicly available at: https://github.com/TUDa-HWAI/CompressKV

Commercial large language models bill, scale latency, and budget context per token. Yet tokenizers assign more subword tokens to the same meaning in some languages than in others, so speakers of languages with high token-fertility pay a structural penalty before a model is ever invoked. This penalty is documented for multilingual settings in general, but it has not been measured systematically for African languages at the level of enterprise deployment economics and cognitive context capacity. We measure it across 20 African languages spanning five language families and three scripts (Latin, Ge'ez/Ethiopic, N'Ko; 19 appear in the primary FLORES-200+ corpus, with Nigerian Pidgin measured via MAFAND-MT only), using parallel corpora so that the language effect is isolated from content. Across 11 frontier and open tokenizers on FLORES-200+, every African language carries a tokenization premium above English (median 1.88x on GPT-5 / o200k_base, up to 8.92x for N'Ko); the penalty is largest for Ethiopic and N'Ko scripts (reaching 7-9x) and is near-invariant across corpora (FLORES vs SIB-200 Pearson r = 0.9998). Translated into deployment terms, this results in up to 8.9x inference cost and an equivalent generation-latency multiplier (N'Ko vs English on GPT-5; 7.4x for Amharic), and as little as 11% of English's effective context window. The best currently available tokenizer for African languages, Gemma 4, reduces the mean premium from 3.31x (cl100k_base) to 2.38x, but no tokenizer eliminates the penalty. We release an open measurement tool (afri-fertility), a public leaderboard, a results dataset, and mitigation guidance for African builders. The penalty falls hardest on the languages whose speakers can least afford it, a digital divide encoded directly into the subword vocabulary.

Bearing fault diagnosis faces critical challenges when dataset heterogeneity, operating condition variations, and limited labeled data occur simultaneously in industrial environments. Existing approaches address these issues in isolation and rely on implicit feature alignment, limiting effectiveness under concurrent challenges. This paper proposes a knowledge-guided two-stage transfer learning framework that employs a lightweight GPT-2-style Transformer with causal self-attention for hierarchical feature extraction from vibration signals, establishing explicit pathways where pre-trained encoder weights and fault prototype embeddings serve as knowledge carriers from multi-source pre-training to target adaptation. The framework addresses the dual-shift challenge through multi-source learning for generalizable representations, prototype-based knowledge modulation for target adaptation, and taxonomy-adaptive classification for seamless transfer across heterogeneous fault categories. Experimental validation on four real-world datasets demonstrates 92.61% average accuracy with only 10% labeled target data, outperforming state-of-the-art methods by 17.24 percentage points, establishing a practical pathway toward cost-effective predictive maintenance in Industry 4.0 applications.

The search for exoplanet biosignatures is guided by whether planetary environments can sustain photosynthesis. As such, the Photosynthetic Habitable Zone (PHZ) was recently proposed, as the overlap between the canonical habitable zone and the orbital range where stellar irradiance is sufficient to drive photosynthesis. Existing PHZ estimates rely on empirical light-response curves from Earth phytoplankton, and thus include implicit Earth-centric biases. We introduce an agnostic PHZ derived from a generalized model of photosynthesis grounded in thermodynamics and redox chemistry, without reference to model organisms. The model is built on a generic photochemical reaction in which photon capture couples oxidation of a donor molecule to the reduction of CO2. The optical properties and CO2 reduction rate are optimized against irradiance spectra for exoplanets orbiting main-sequence stars, using a genetic algorithm that mimics evolution by natural selection. Our simulations predict that photosynthetic organisms compensate for reduced flux by evolving larger light-harvesting structures. As a result, photosynthetic viability declines only linearly with orbital distance, despite stellar flux falling off quadratically. As such, the agnostic PHZ expands well beyond previous Earth-based estimates. Earth-like (visible light) oxygenic photosynthesis is flux-limited at the outer habitable zone for cool M-dwarf stars; however, both anoxygenic photosynthesis and a hypothetical, NIR-driven oxygenic photosynthesis are viable across the entire habitable zone for M, K, and G stars. This implies that M-dwarf exoplanets could sustain robust oxygenic photosynthesis, though it would be different to that found on Earth, presenting reflectance biosignatures in the NIR band rather than the visible.

Modern coding agents pair LLM generators with various tools, including cheap diagnostics and expensive verifiers. The tool-use decisions are typically governed by orchestrators that often use fixed rules and ignore uncertainty. We formulate orchestration as cost-sensitive sequential hypothesis testing: a Bayesian controller maintains a belief over candidate correctness and dynamically decides whether to gather more evidence, refine the candidate, verify it, or stop. Across six generators and nine coding benchmarks, Bayesian control proves to be most valuable when verification is costly and critics are informative but imperfect. Beyond control, the belief state yields an interpretable correctness score that outperforms token-probability and raw tool-success baselines for uncertainty quantification.

Perceiving physical contact is fundamental to dexterous manipulation. While robots often rely on dedicated hardware tactile sensors, humans exhibit a remarkable ability to infer contact by integrating visual information with an innate sense of their body's pose and movement. Inspired by this embodied perceptual skill, we investigate whether a robot can learn to infer contact from vision, an approach that also offers a scalable alternative to tactile hardware specifically for binary contact estimation, which faces practical challenges in cost, fragility, and integration. We present NoContactNoWorries, a transformer-based multimodal framework that fuses RGB-D vision with the robot's proprioception to infer binary contact states as a pseudo-tactile signal for hand-object interactions. We validate by training a single contact prediction model on multiple objects and show that the inferred contact signal supports downstream reinforcement learning agents for in-hand object reorientation, generalizing to novel objects. Experiments in both simulation and on a real-world robot validate our approach, highlighting the feasibility of inferring contact from vision and proprioception. Project Page: https://soham2560.github.io/no-contact-no-worries/

Modern software projects depend on third-party libraries that evolve continuously, introducing breaking API changes that prevent client code from compiling after a dependency update. When the same library update breaks multiple projects, existing repair approaches generate project-specific patches that cannot be reused, requiring each affected project to be repaired independently. We present BigBag, an agentic framework that generates fixing transformations: structured, executable programs that encode the repair logic at the API level and transfer to any client broken by the same update. We evaluate BigBag on 157 compilation failure breaking dependency updates from the BUMP benchmark, across eight configurations combining four large language models and two AST transformation engines (Spoon and JavaParser). The best configuration achieves a compilable transformation rate of 94.3% and a fix rate of 78.6%. Generated transformations transfer across projects, achieving a cross-project fix rate of 33.3% overall and 80% or above for breaking updates where all clients invoke the affected API element uniformly. These results show that agentic generation of reusable fixing transformations is a viable approach to scalable repair of breaking updates.

Mixture-of-Agents (MoA) architectures improve inference-time scaling by organizing multiple LLM agents into layered reasoning pipelines. However, existing MoA variants fail to sustain gains as depth increases, exhibiting degradation, early plateauing, or saturation. We propose ReM-MoA, a memory-augmented MoA framework that sustains scaling through two mechanisms: (1) a Ranked Reasoning Memory that persistently stores and ranks reasoning traces from all layers using a comparative Reviewer Agent, and (2) a Curated Diversified Memory Routing scheme that exposes different agents to distinct combinations of successful and failed traces, preserving exploration diversity while propagating high-quality reasoning. We further introduce an optional multi-domain Reviewer distillation pipeline that improves ranking quality through frontier-model supervision. Across five reasoning benchmarks spanning math, formal logic, code, knowledge, and commonsense, ReM-MoA consistently outperforms prior MoA variants across both depth and width scaling, and its advantage widens with depth, establishing structured cross-layer reasoning memory as a key missing mechanism for scalable multi-agent inference.

Medical point cloud completion is important for anatomical reconstruction and downstream clinical workflows, yet generative modeling in this setting remains insufficiently studied. We investigate completion through continuous-time generative modeling and introduce PCFM, a PTv3-backed flow matching approach for medical point cloud completion. We evaluate on SkullFix and SkullBreak, and additionally on the more recent Mandibular Defect dataset. We build strong baselines by adapting PTv3 to a deterministic encoder-decoder completion model and by instantiating diffusion completion (PCDiff) with both PVCNN and PTv3 denoisers. PCFM with PTv3 is competitive with the deterministic PTv3 baseline and achieves state-of-the-art generative performance across datasets, while requiring substantially fewer sampling steps than diffusion. At the best operating points, PTv3 also yields clear throughput gains, providing up to a 7$\times$ speed-up for PCFM compared to a PVCNN backbone. Finally, we study empirical scaling trends by varying model size and point cardinality, showing consistent gains with higher point resolution and informative trade-offs across model scales.

Training of neural networks for histopathology classification tasks typically relies on data encoding into latent space, which reduces complexity and improves performance. There are several encoder networks available, either pretrained on general image datasets such as ImageNET, or specifically on histopathological images. Training of encoder networks should be adapted to downstream tasks, allowing encoding of biologic/diagnostic content while rendering networks invariant to label-irrelevant transformations. This paper investigates the effect of classical image transformation on the latent space, using networks provided by Lunit Inc. and Bioptimus, both focusing on pathological images, and by Meta Research Team. We assess variance of embeddings resulting from standard data transformations by comparing original and transformed image embeddings and by contrasting them with random, unrelated embeddings, using image tiles from hematoxylin/eosin-stained sections available in a colorectal tissue dataset and the publicly accessible TCGA dataset. Our findings show that embeddings of original and transformed images are closer to each other than to random embeddings, indicating robustness to transformations. However, they are not fully invariant, revealing that the encoder networks do not completely neutralize transformation effects in latent space, explaining why transformation-mediated augmentation of datasets can improve performance. Significant differences were observed between general and histopathology-specific encoder networks.

Generative AI coding agents are entering the open-source supply chain, yet their diverse and often invisible traces leave their prevalence poorly understood. We introduce a multi-layered detection framework that integrates configuration-file scanning, commit-message analysis, author-identity matching, and bot-signature lookup across World of Code (180M+ Git repositories), classifying agent traces into four behavioral types. No single method captures more than a fraction of activity: multi-method detection identifies 850,157 Claude Code commits in one snapshot, of which bot-account lookup_the signal most adoption studies rely on_recovers only 28,154 (3.3%), a 30x relative-recall gap, so single-signal prevalence estimates are biased low by at least this factor. Every detection pattern is hand-validated (495 labels) with per-cell precision and Wilson confidence intervals. Across snapshots from December 2024 to April 2026, commit-attributed agents generate over 320,000 commits per month; Claude Code leads (886,122 commits across 17,295 projects) and dominates silent, configuration-file-only adoption (21,078 projects). Compared against an independent pull-request census (AIDev), the two channels capture nearly disjoint agent populations_a PR census misses 79% of commit-detected Claude Code adopters and essentially all Codex adopters_and different kinds of work: PR-deployed cloud agents (Codex, Cursor) surface as feature work, while commit-deployed in-editor agents (Claude Code, OpenHands, Aider) surface as maintenance. The observed work profile follows deployment and detection mode rather than the tool itself, so no single channel is representative.

Experience-driven self-evolution is critical for large language model (LLM) agents to improve through open-world interaction. However, existing experience learning methods mostly rely on single-agent loops, where the same agent executes tasks, summarizes outcomes, and determines memory content. This setup makes agents vulnerable to the Self-Confirmation Trap: wrong-but-self-consistent trajectories are misidentified as successful experience, leading to cumulative errors during retrieval and reuse. To address this issue, we propose EDV, an Execute-Distill-Verify framework for reliable experience learning. In the Execute stage, multiple heterogeneous agents explore the same task space in parallel to generate diverse candidate trajectories. In the Distill stage, a dedicated third-party agent comparatively analyzes these trajectories to produce candidate experiences, reducing executor-centric summarization bias. In the Verify stage, the execution group validates candidates via a consensus mechanism, and only approved experiences are written into shared or private memory. By decoupling the three stages, EDV transforms experience learning from isolated self-reflection into collaborative construction, filtering erroneous and noisy content before memory insertion. We evaluate EDV on three challenging long-horizon benchmarks: tau2-bench, Mind2Web and MMTB. Results show EDV consistently outperforms strong baselines, validating that reliable experience construction is essential for robust agent self-evolution. Our code is available at https://github.com/shidingz/EDV.

Spectral filtering recently delivered substantial pruning for \emph{static} subgraph matching: Laplacian interlacing rejects candidates whose neighborhoods cannot host the query. We study whether such aggregate structural tests can accelerate \emph{continuous} subgraph matching (CSM) over dynamic graphs, and answer in three parts. First, lazily maintained spectral bounds are infeasible exactly where spectral pruning has value: we characterize the tightest safe rule over a formalized perturbation relaxation and show that even it loses essentially all pruning power within four touching updates. Second, exact maintenance is affordable when selective: pruning utility and recomputation cost are anti-correlated across vertices -- hubs provably never prune -- so recomputing small-neighborhood spectra on touch sustains exact local spectra at microseconds per update, complete by construction. Third, integrated into a decoupled CSM benchmark against an identical-minus-spectra control, the tests remove up to $51\%$ of candidates or safely skip up to $47\%$ of update enumerations, yet enumeration intermediates remain unchanged -- beyond the gates' skipped first-level bindings, typically zero -- across two engines, four real graphs, two stream types, and $77$ solved queries; a constructed radius-stratified workload confirms the instrument detects the exception when one exists ($-99.9\%$ intermediates, $748\times$ faster). Aggregate tests accelerate what scales with candidate sets -- construction, list scans -- never adjacency-guided exploration. We distill an intermediate-invariance methodology for evaluating CSM filters and release a reusable dynamic local-spectra index.

In high-stakes document processing pipelines, including financial reconciliation, compliance verification, and procurement automation, an LLM extraction that is silently wrong is more dangerous than one that is visibly absent. The central challenge is not extraction accuracy alone but reliable confidence estimation: knowing, field by field, whether an extraction can be trusted for automation or deferred to human review. Token-level log-probabilities, verbalized confidence, and multi-sample self-consistency all collapse toward all-positive behaviour at practical thresholds, offering no reliable separation between trustworthy and untrustworthy extractions. We present ExtractConf, a cross-domain, field-agnostic confidence engine that grounds confidence estimation in two structurally different readings of the same document. A field-guided Hunter call extracts each field under schema-slot completion pressure; a document-guided Mapper call scans holistically and surfaces values grounded in document content. This asymmetry yields different failure modes: Hunter hallucinates values for absent fields, while Mapper misses visually non-salient ones. Their disagreement is independently informative. ExtractConf fuses cross-call disagreement, LLM-internal uncertainty, OCR, image quality, and spatial layout into a classifier requiring no domain-specific rules or retraining. On DocILE (55-field invoices, 26% failure rate), it achieves 0.928 ROC AUC and reduces selective prediction risk by 70% over logprob-mean. At 80% coverage, accuracy reaches 99.1%, enabling a practical human-in-the-loop workflow. Zero-shot transfer to CORD receipts achieves 0.858 AUC; lightweight Lasso recalibration reduces ECE by 89% and Brier by 43%, confirming the signals generalise across document domains.

Data augmentation is a simple and model-agnostic approach for exploiting known invariances in learning problems. Given a group acting on the input space, one augments the training set with transformed copies of each sample. Because it exploits symmetries without modifying the underlying learning algorithm, data augmentation can be applied broadly across learning methods. However, this universality comes at a computational cost: when the group is large, full group-sized augmentation quickly becomes computationally infeasible. This raises a fundamental question: Can partial data augmentation achieve the same statistical benefits as full augmentation in terms of generalization and sample complexity? We develop a general framework for investigating this question using Fourier analysis and the representation theory of finite groups. We show that, for a broad class of classical learning problems, partial data augmentation based on a randomly sampled subset of group elements achieves the same minimax rates as full augmentation, up to an approximation error that vanishes as the subset size increases. Our results provide a theoretical explanation for why partial augmentation can retain the statistical benefits of full augmentation despite enforcing symmetry only approximately, and shed light on a recently raised question in learning with symmetries: whether statistically optimal learning under general group invariances can be achieved using computationally scalable methods. Moreover, we prove a complementary impossibility result: enforcing exact invariance via data augmentation requires averaging over the entire group, and cannot be achieved by any strict subset when the hypothesis space is sufficiently expressive. Together, these results provide a unified perspective on full and partial data augmentation, as well as exact and approximate symmetry enforcement.

Network operators' changing policies, service requirements, and stringent real-time constraints render existing methods designed with fixed objectives and constraints ineffective. This paper presents Agentic long-term performance optimization (Agentic-LTPO), a nested bilevel optimization framework that can be applied to adaptive physical layer problem configuration. The key idea is to employ agentic AI to generate upper-level configurations in a bilevel optimization structure, where evolving operator policies, environment summaries, and historical experiences are translated into structured lower-level optimization problem configurations. The lower level solves the problems with updated configurations for real-time physical-layer decisions. Considering cell-free MIMO beamforming as a use case, we embody Agentic-LTPO by designing a new multi-agent decision process with retrieval-augmented experience-based verification in the upper level, together with a closed-form beamformer in the lower level. Experiments demonstrate that Agentic-LTPO exhibits strong adaptability to dynamic operator policies and effectively enhances the system's long-term performance by 57.2% compared to traditional methods.

Formal verification produces machine-checkable certificates that attest to the satisfaction or violation of temporal properties, yet these certificates remain opaque to non-specialist stakeholders. We propose a cycle-consistent neural architecture that generates faithful natural language explanations of verification certificates. A forward network NN1 maps certificates to explanations, and an inverse network NN2 reconstructs certificates from explanations; a symbolic verifier closes the loop, providing a differentiable faithfulness proxy. A pointer-generator mechanism ensures lexical grounding by copying state names directly from the certificate. We evaluate on 420 test certificates spanning six verification methods (bounded proof, k-induction, inductive invariant, lasso, reachability, witness pair) in both YES and NO verdict variants, drawn from a financial compliance domain with 207 named states. Our trained architecture, combined with a hybrid inference-time routing strategy, achieves 90.0% cycle-verified soundness, surpassing a multi- LLM few-shot baseline (76.1% for the best of 16 LLM combinations across four frontier models) by 13.9 percentage points. The neural model wins on 10 of 12 verdict/kind categories, with three categories reaching 100% soundness. The architecture offers 860x faster inference (185 ms vs. 160 s per certificate for the full multi-LLM baseline), offline operation, deterministic outputs, and zero per-inference cost. These results demonstrate that trained specialization outperforms general-purpose LLM prompting for structured certificate explanation, while eliminating the deployment constraints of cloud-based inference.

Assessing the privacy of large language models (LLMs) presents significant challenges. In particular, most existing methods for auditing differential privacy require the insertion of specially crafted canary data during training, making them impractical for auditing already-trained models without costly retraining. Additionally, dataset inference, which audits whether a suspect dataset was used to train a model, is infeasible without access to a private non-member held-out dataset. Yet, such held-out datasets are often unavailable or difficult to construct for real-world cases since they have to be from the same distribution (IID) as the suspect data. These limitations severely hinder the ability to conduct scalable, post-hoc audits. To enable such audits, this work introduces natural identifiers (NIDs) as a novel solution to the above-mentioned challenges. NIDs are structured random strings, such as cryptographic hashes and shortened URLs, naturally occurring in common LLM training datasets. Their format enables the generation of unlimited additional random strings from the same distribution, which can act as alternative canaries for audits and as same-distribution held-out data for dataset inference. Our evaluation highlights that indeed, using NIDs, we can facilitate post-hoc differential privacy auditing without any retraining and enable dataset inference for any suspect dataset containing NIDs without the need for a private non-member held-out dataset.

We consider an oracle that processes a limited batch of records at a time and clusters those that refer to the same real-world entity. We study how to interrogate such an oracle to resolve entities in a dataset whose size is far larger than a single batch, and where no batch is guaranteed to contain all records of any given entity. We aim at a pay-as-you-go approach, to have full control over the costs (the number of oracle consults), while achieving the highest possible recall at every step. We formally cast this problem as batched entity resolution, prove that selecting optimal batches is NP-hard, and provide an optimal solution under a natural condition on entity sizes. Finally, we evaluate our approach on six datasets and show its superiority over state-of-the-art baselines.

Object interaction tasks have been a focus of advances in imitation learning. End-to-end methods, dominated by diffusion and flow-based variants have shown leaps in performance while sacrificing interpretability. Object-centric and pose-informed variants have had a role in learning from demonstration in manipulation tasks. In this paper, we revisit a few modern imitation learning benchmarks for object interactions, with the aim of composing a framework that repurposes principled theories of manipulation, preserving both performance and interpretability. For image observations, lightweight training is proposed for model-free pose estimation of the target object, using self-supervision over the demonstration data available for imitation learning. This information is then used to inform a manipulation mode-aware retrieval of a demonstration, a mode-aware transformation, a replan step that connects to the retrieval point while preserving mode constraints, and finally rolling out the transformed demonstration. These compose four key steps of the proposed RE4 framework, evaluated over state-based and image-based benchmarks in Push-T and Robomimic. An adversarial benchmark that evaluates sparse data regions of image-based Push-T showcases the robustness, further bolstered by indications from low-data regime experiments. The current work shows promise in using simple interpretable building blocks to learn manipulation skills.

Large Transformer models function as Dense Associative Memories (DAMs), retrieving knowledge via high-dimensional attractor dynamics driven by the self-attention mechanism \citep{ramsauer2020hopfield, wu2024attention}. However, adapting these frozen memory systems to new tasks presents a fundamental ``Plasticity-Stability'' dilemma. Current methods either risk catastrophic interference by modifying synaptic weights directly (e.g., LoRA) \citep{hu2021lora} or degrade associative capacity by clogging the retrieval buffer with static prompt tokens (e.g., VPT) \citep{jia2022vpt}. In this work, we propose \textbf{H-Res} (Hierarchical Residual Steering), a mechanism that modulates the effective energy landscape of the Transformer without altering its global equilibrium or expanding its sequence length. By formulating adaptation as a control problem on the activation manifold \citep{chen2018neuralode}, H-Res learns a state-dependent vector field that steers token trajectories into task-specific basins of attraction. We formally prove that H-Res preserves the attention entropy of the foundation model and facilitates Neural Collapse \citep{papyan2020prevalence}. Empirically, Manifold Steering outperforms global weight modification by 26\% on associative retrieval tasks and eliminates the computational overhead of prompt-based methods, scaling effectively to structured domains \citep{zha2023vtab}.

Electroencephalography (EEG) is the dominant non-invasive modality for brain-computer interfaces (BCIs), yet reliable decoding of motor imagery is hampered by inter- and intra-individual variability. A recurring claim is that one decoding pipeline, most often a spatial or Riemannian method, is broadly preferable. We test the weakest version of that claim under the most favourable conditions. Using the Mother of All BCI Benchmarks (MOABB) framework, we evaluated 1,056 decoding configurations (feature extractor x scaler x classifier), >340,000 subject-level model fits, across three public left-versus-right motor-imagery datasets (PhysionetMI, 109 participants; Cho2017, 52; Zhou2016, 4) and two frequency bands (8-15 Hz, 8-30 Hz). Every model is fit and tested within a single session of a single participant, the easiest regime, giving every pipeline its best chance. We apply the statistics standard for multi-classifier comparison: Friedman omnibus tests, Nemenyi critical-difference analysis and Wilcoxon signed-rank tests with effect sizes. Covariance tangent-space projection (cov-tgsp) and Common Spatial Patterns (CSP) are the strongest families, but their ordering is dataset-dependent and, on the largest and most heterogeneous cohort (PhysionetMI), statistically indistinguishable (Nemenyi p = 0.27; Kendall's W = 0.11). At the individual level the single best pipeline is optimal for only 35% of PhysionetMI participants, and nonlinear descriptors are best for roughly one third; matching pipeline to participant adds about seven accuracy points over the best fixed choice. The ranking is not an artefact of dimensionality, and classifier and scaler choices are secondary to the feature representation. Even in the easiest regime, no single pipeline dominates: a lower bound on the personalization problem and a quantitative case for participant-aware model selection rather than a universal decoder.

Existing ECG report generation is tightly coupled -- interpretation and reporting fused end-to-end, so errors propagate without stage-level recourse -- while agent-based systems decouple tasks but remain single-pass, never revisiting earlier outputs. Clinical ECG reporting instead unfolds iteratively, requiring progressive context integration and bidirectional editing. We present \textsc{ATRIA}, a multi-agent ECG reporting system that mirrors the clinician's iterative workflow: it binds every report claim to its supporting evidence, flags statements unsupported by that evidence, incorporates additional context mid-session, and lets clinicians verify and revise individual findings rather than accept one opaque output. Because its agents use ECG analysis models already in clinical use, the underlying findings are clinically trustworthy; and as a cloud-based web service, \textsc{ATRIA} is ready for immediate deployment. We demonstrate \textsc{ATRIA} through four interaction cases, with a live demo and video available.

We introduce Age of LLM, a turn-based 1v1 benchmark in which two LLMs face off on a 13x7 grid to destroy the enemy base. Three stressors are deliberate: fog of war, full diplomacy (messages, ceasefires, ultimatums; uranium kept secret), and a reliability dimension where every turn must follow a strict JSON schema and an illegal action is silently discarded. The engine is private and each match uses a fresh random map seed and opponent, mitigating the data contamination that affects public benchmarks. Models receive a (near) rule-only prompt with no build-order advice (two tactical seed phrases were present during data collection; see Section 2.7). We benchmark 15 reasoning models across 54 matches and 5,258 actions. Findings: (1) the nuclear rush dominates (78% on the rules-coherent v0.11+ sub-corpus; 85% corpus-wide) with a sole-launcher signature that is largely mechanical under secret-simultaneous launch rules, not a cognitive deterrence failure; (2) military conquest is rare but faster (12.3 vs 18.9 turns); (3) diplomacy is prolific yet almost never consummated; (4) ~58% of illegal actions are fog/state errors, making the illegal-action rate a measure of belief-tracking; (5) -- the least established, and the only one we label exploratory -- a weak link associates reliability with winning. The corpus is small, unbalanced and not side-swapped, so the ranking is a preliminary descriptive view, not a contribution. Beyond ranking, the turn-by-turn traces of actions and messages make the corpus a lens on how LLMs reason under adversarial uncertainty -- their belief-tracking, spontaneous deception, and per-model cognitive "personas" -- which we frame as a future research direction. We release the replay format, an isometric viewer and all replays; engine source on request.

Standardized assessment of uterine MRI remains challenging due to anatomical variability, observer dependence, and the lack of workflow-integrated automated analysis tools. This work presents Female-RHINO: (R)eproductive (H)ealth (I)maging A(N)alysis T(O)ol, a real-time AI-assisted framework for automated quantitative uterine MRI analysis and structured reporting during image acquisition. We present an end-to-end system that integrates inline communication with the MRI scanner and deep learning-based analysis to derive quantitative uterine biomarkers from sagittal T2-weighted pelvic MRI. The framework combines segmentation and anatomical landmark detection models trained and evaluated on more than 500 multi-center datasets spanning diverse protocols, vendors, and patient populations. It performs volumetry, detects and quantifies common incidental findings such as fibroids and Nabothian cysts, and extracts six anatomical landmarks for biometric assessment. Results are compiled into a structured clinician-oriented report with integrated visualizations, without manual interaction. Evaluation on independent retrospective and prospective cohorts demonstrated robust performance across varying acquisition settings. Mean Dice similarity coefficients were 0.82 for the uterus and 0.80 for fibroids, with lower but consistent agreement for Nabothian cysts. Landmark detection achieved a mean radial error of 3.7 mm. End-to-end processing was completed in under 70 seconds, enabling availability of results during the ongoing scan. Prospective deployment yielded immediate, standardized, and reproducible analyses supported by inter-observer agreement. The proposed system enables real-time scanner-integrated AI for automated uterine MRI analysis and reporting, with potential to improve standardization, efficiency, and clinical workflow in pelvic imaging.

We introduce a large-scale, open-source dataset of pre-generated adversarial attacks for vision-language models (VLMs). The dataset is designed to be diverse, representative, and practical, extending existing benchmarks by covering 10 high-level categories and 55 subcategories of harmful intents. Our primary goal is to make adversarial data accessible to the research community, given the computational cost and complexity of generating large numbers of attacks. The dataset comprises 47 524 adversarial samples, generated using state-of-the-art attack strategies from recent literature. Our work complements existing efforts by consolidating and extending prior benchmarks from multiple established sources, resulting in 7 826 intents, and introduce an additional category to broaden coverage. This provides realistic evaluation resources for studying model robustness and alignment. Our dataset intends to enable researchers and practitioners to systematically evaluate the robustness and safety of VLMs, fine-tune attack-generation models, and develop or stress-test defensive guardrails under diverse adversarial conditions. By releasing this resource, we aim to lower the barrier to adversarial research and foster more reproducible, comprehensive, and comparable evaluations of VLM safety.

Vehicle advertisements contain rich specification information, but automotive NER resources remain limited. We introduce AutoSpecNER, an expert-annotated dataset for fine-grained entity recognition in vehicle listings. The dataset includes 659 advertisements from a popular car-selling website, with over 10,000 entities annotated across 15 categories, including MODEL, ENGINE_SPEC, and BATTERY_CAPACITY. Annotation quality was validated through inter-annotator agreement, achieving an average score of 91.5%. We benchmark rule-based extraction, fine-tuned transformer encoders, and large language models. DeBERTa achieves the best performance with a 90% micro-F1 score, outperforming the rule-based baseline (43%) and the strongest large language model (77.8%).

Prompt-based interaction has become a dominant paradigm for using large language models (LLMs), where multiple candidate prompts are evaluated and the top-ranked one is selected for downstream use. This workflow implicitly assumes that prompt rankings are stable under minor variations in evaluation conditions. In this paper, we systematically study prompt ranking stability under common sources of variability, including random seeds and limited evaluation subsets. Across three open-weight LLMs and two benchmark tasks, we find that while overall rank correlations are often moderate to high, the identity of the top-performing prompt frequently changes, leading to unreliable selection decisions. To address this issue, we propose a simple stability-aware selection strategy based on a lower confidence bound, which accounts for both performance and variance. Our results show that this approach improves robustness in unstable settings while remaining competitive in more stable regimes. These findings highlight the importance of accounting for evaluation uncertainty in prompt selection and LLM benchmarking.

Fully Homomorphic Encryption (FHE) enables computations to be performed directly on encrypted data while preserving data confidentiality. However, its practical applications remain limited by high computational costs and development complexity. This paper presents ComputeFHE, an open-source C++ library that facilitates the development of privacy-preserving applications based on the TFHE cryptosystem. The library provides encrypted integer and fixed-point data types together with arithmetic, logical, comparison, conditional, and oblivious array-access operations which allow developers to implement algorithms using a familiar imperative programming paradigm. ComputeFHE supports both conventional TFHE arithmetic based on standard two-input logic gates and an optimized Arithmetic Logic Unit (ALU) architecture utilizing FHE-friendly logic primitives. Experimental results demonstrate significant reductions in the number of required bootstrapping operations, achieving performance improvements of up to 3.9x for selected operations. In addition, the library includes a simulation mode that enables testing, debugging, and complexity analysis without performing actual cryptographic computations while providing circuit complexity and bootstrapping costs. Built on top of OpenFHE, ComputeFHE offers a practical and accessible framework for developing and evaluating privacy-preserving algorithms and applications.

Convolutional Kolmogorov--Arnold Networks (KANs) replace the fixed weights of a convolutional kernel with learnable univariate functions. The dominant formulation attaches one such function to every kernel entry and lets it act on pixel values, expressive but parameter-heavy and prone to overfitting. We argue that the learnable functions are better placed in the \emph{structure} of the convolution than on each edge, and we organise the design space along a single axis: whether the function acts on the pixel \emph{values} or on the filter \emph{shape}. We study three realisations. SV-KAN applies one shared univariate function to the values and leaves the spatial filter free and static, aa classical convolution with a single learnable shared activation. AG-KAN keeps the shared value function but supplies the spatial structure through a content-adaptive Gaussian gate. RF-KAN instead moves the learnable functions onto the filter shape, building each filter from oriented ridge profiles expanded in a localised oscillatory (Morlet) wavelet basis with content-adaptive amplitudes. Under a matched four-layer protocol with in-run references and three seeds, RF-KAN and SV-KAN reach $88.47\pm0.10\%$ and $88.20\pm0.31\%$ on CIFAR-10 and $64.40\pm0.19\%$ and $64.57\pm0.30\%$ on CIFAR-100, at about $0.4$M parameters. At this matched scale the shape model and the simplest value model meet at the top, both above a plain convolution and every per-edge KAN we tested, including the official Gram variant, at roughly a fifth of the parameters. A controlled study attributes the RF-KAN gain to an intrinsically localised oscillatory basis and to content adaptivity, and an ablation that removes the learned shape entirely, leaving only the shared value function, collapses accuracy by over forty points, identifying the learned shape as the load-bearing ingredient at this scale.

Large language models (LLMs) are increasingly integrated into decision-support roles in business and policy contexts. While prior benchmark studies have primarily evaluated LLMs' causal reasoning capabilities, a more fundamental epistemic dimension has been overlooked: Causal Caution, defined as the propensity to refrain from causal judgment when empirical evidence is insufficient. This study examines the systematic suppression of Causal Caution that occurs when LLMs shift from academic to practical advisory contexts. Using an evaluation rubric inspired by Pearl's Causal Hierarchy (the PCH score), we conducted experiments on four high-performance LLMs -- Claude Sonnet 4.6, Claude Opus 4.7, GPT 5.5, and Gemini 3.1 Pro -- across 480 trials. Causal Caution maintenance rates were 91.7--100.0% in academic contexts but dropped to 6.7--18.3% in practical advisory contexts (Fisher's exact test, p < .001 across all models). Furthermore, when restricted to practical prompts requesting concrete recommendations or explanatory rationales, only 1 of 200 responses (0.5%) maintained Causal Caution. A brief self-correction prompt -- "Please reconsider this judgment from the perspective of causal relationships" -- restored the expression of Causal Caution to maintenance rates of 71.4--100.0% (McNemar's test, p < .001 across all models). These results suggest that helpfulness-oriented response patterns may suppress the expression of Causal Caution in practical advisory contexts, with important implications for organizational governance. The findings indicate that this suppression reflects context-dependent variation in expression rather than an underlying capability limitation, suggesting that multi-agent architectures that separate proposal generation from causal auditing may offer a promising governance design.

Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregressive large language models (LLMs). In parallel, diffusion-oriented RL algorithms, e.g., DanceGRPO and FlowGRPO, have rapidly expanded the scope of RL from language reasoning to diffusion-based visual and flow-based generation. However, efficient RL systems for diffusion generative LLMs remain underexplored. Existing implementations, e.g., veRL-Omni, still rely on colocated execution, which simplifies synchronization but couples rollout and training resources, limits heterogeneous deployment, and constrains independent scaling. To this end, we introduce DigenRL, a disaggregated RL framework for diffusion-based generative LLMs that supports flexible resource allocation, accommodates heterogeneous GPUs, and facilitates efficient task scheduling. To maximally reduce the execution bubbles in the disaggregated architecture, we propose: 1) a generation-axis pipeline (GAP) and time-step parallelism (TSP) in the diffusion architecture to enable finer-grained pipelining between rollout and training; 2) an elastic trainer-assisted generation (TAG) approach to enable the trainer GPU resources to dynamically assist in executing rollout generations; and 3) a tightly one-step constrained asynchronous strategy to further utilize the tail bubble in the pipeline. Extensive experiments are conducted on three hardware testbeds with 16-32 GPUs using HunyuanVideo-13B, Wan2.1-14B, FLUX.1-12B, and QwenImage-20B generative models. Experimental results show that DigenRL achieves 1.56-2.10x throughput improvements over state-of-the-art diffusion RL systems, veRL-Omni and GenRL.

We present MorfFlex, a morphological dictionary architecture suitable for languages with extensive regularity in both inflection and derivation. As the primary example of MorfFlex in use we introduce MorfFlex CZ, a morphological dictionary of Czech. It is distributed as a simple, unstructured list of <wordform, lemma, tag> triplets, however, its manually maintained, unpublished source files and conversion scripts encode a sophisticated system of inflectional and derivational patterns. These patterns dramatically reduce the otherwise enormous size of the dictionary, which currently contains over 100 million wordforms and more than 1 million lemmas. The MorfFlex CZ dictionary serves as an essential resource for ensuring the consistency of manual morphological annotation in the Prague Dependency Treebanks and underpins state-of-the-art automatic tools such as MorphoDiTa. In this paper, we focus on: (i) presenting an effective method for managing the rich morphological system within the dictionary, and (ii) demonstrating the utility of such a language resource for maintaining annotation consistency in corpora and supporting the development of advanced NLP applications.

This paper proposed an algorithm for part-of-speech (POS) tagging senses of a bilingual dictionary. The algorithm is applied on the Al-Mawrid Arabic-English dictionary. The tagging task is accomplished by transferring the POS tags of the English translation equivalences (TEs) to the dictionary senses after dis-ambiguities process. The English POS tags of senses are acquired from the Princeton WordNet. POS tagging of bilingual dictionary senses is prerequisite to link a bilingual dictionary to WordNet and/or standardizing that dictionary into WordNet-LMF format where the synset (set of synonyms), not word, is the basic brick. The registered accuracy is high though the cost is little. Building NLP/HLT tools needs linguistic experts, large investments, and long time. For statistical approach, we need large annotated corpora and for rule-based approach, we need large lexicon that contains rich linguistic and world knowledge. That motivates the appearance of what are called resource-light approaches to develop natural language processing (NLP) tools for poor-resource languages.

Bird's-eye view (BEV) perception fuses multi-camera images into a unified top-down representation for autonomous driving. Despite recent progress, state-of-the-art methods remain confined to closed-set scenarios, making them vulnerable to unpredictable real-world environments. In this work, we introduce open-vocabulary BEV segmentation (OVBS), which leverages vision-language models (VLMs) to recognize categories beyond the training set while maintaining precise BEV perception and real-time efficiency. A key challenge in OVBS lies in the 3D geometric inconsistency inherent in the ill-posed lifting of 2D VLM semantics into BEV. To address this, we propose OVBEVSeg, a geometry-aware OVBS framework that enhances efficient Gaussian splatting (GS)-based unprojection by leveraging robust 3D geometric constraints across three progressive stages: (1) 2D-to-BEV pseudo-labeling via reliable 3D projection for OV generalization; (2) joint 2D-BEV per-scene optimization with BEV structural constraints for 3D geometric consistency; and (3) 3D geometric distillation for online efficiency. On the nuScenes dataset, OVBEVSeg achieves state-of-the-art performance, outperforming closed-set methods by 15.3 mIoU on unseen categories. Remarkably, even with no novel-class ground-truth labels, it remains competitive with self- and semi-supervised baselines trained with up to 40% of ground-truth annotations. Furthermore, it achieves 2.5x faster inference with only 0.22x the memory consumption of projection-based methods. Project page: https://hchoi256.github.io/projects/ovbevseg/.

Accurate short-term PM$_{2.5}$ forecasting is important for public health protection, air-quality early warning, and urban environmental management. However, PM$_{2.5}$ variation is driven by multiple coupled factors, including stable periodic changes induced by human activities and meteorological regularity, station-specific short-term concentration evolution, and meteorology-driven pollutant dispersion among monitoring stations. Existing spatio-temporal forecasting methods may capture station relationships to some extent, but distance-only, correlation-based, or purely adaptive graphs are often insufficient to comprehensively represent these heterogeneous factors, especially wind-direction-dependent pollutant transport. To address this problem, we propose a Multi-View Geo-Wind Guided KAN model for PM$_{2.5}$ forecasting, named \textbf{MVG-KAN}, which models station-level PM$_{2.5}$ evolution from three complementary views: local periodic regularity, station-wise residual temporal dynamics, and meteorological-environment-guided spatial dispersion. Specifically, the periodic-residual forecasting backbone first separates stable daily and weekly patterns from non-periodic residual variations. A Geo-Wind Graph is constructed by combining geographic distance decay with wind-direction- and wind-speed-aware transport, providing a lightweight physically motivated directed spatial prior for residual propagation among stations. In addition, a temporal Kolmogorov-Arnold network (TKAN) residual head is then introduced to learn station-wise nonlinear autoregressive correction from de-periodized PM$_{2.5}$ residuals and historical multi-pollutant sequences, thereby enhancing the modeling of local residual inertia and pollutant co-variation.

Petroleum-engineering search exposes a supervision gap for strong general retrievers: relevant evidence exists in public web text, but domain relevance labels are scarce. To address this gap, we propose PETRA, a large-scale Petroleum Engineering Text for Retrieval Adaptation dataset and pipeline that converts noisy public web data into a curated domain corpus and synthetic supervision for dense retrieval and reranking. PETRA contains 1.36M curated chunks, approximately 2B token equivalents, $\approx$859k, embedding training rows from $\approx$224k anchors, and roughly 400k teacher-scored reranker candidate rows. Its construction combines high-recall energy-domain curation, an energy-domain classifier with 98.4% test accuracy, chunk-grounded query generation, LLM-written hard negatives, and retrieval-mined candidate lists. PETRA improves first-stage in-domain Normalized Discounted Cumulative Gain (nDCG) from 0.703 to 0.763 through score fusion. Reranker adaptation improves the public Earth Science benchmark by 44% relative and a six-task reasoning-intensive panel by 23%. Failed training recipes show that high train-holdout accuracy on synthetic labels does not predict retrieval gains; retrieval-mined data helps only after being repackaged as teacher-scored candidate lists sampled from the inference-time candidate distribution.

ODRL policy evaluators produce verdicts, but say nothing about the normative positions a policy brings into existence, the authority structures those positions presuppose, or who holds the power to declare a norm violated. We formulate the Cross-Level Design Principle: any normative language with violable, consequential norms requires both conduct-level positions (Permission, Duty, Right, No right) and competence-level positions (Power, Subjection, Immunity, Disability). Applying this to ODRL, we establish that prohibition is sanctioned (violation possible and consequential), that permission is underspecified across its behaviour parameter (open vs. closed world), and that the formal semantics covers achievement obligations only. We ground ODRL in UFO-L, mapping each activated rule to a simple legal relator and extending coverage from two to eight legal positions; violation-declaration authority, implicit in every existing evaluator, becomes an explicit Power-Subjection pair. All axioms are mechanically verified in Isabelle/HOL and across a 39-problem benchmark under Vampire, E, and Z3.

In recent years, the Internet of Things (IoT) paradigm has been shifting toward batteryless, energy-harvesting architectures. Sustaining reliable operation in these systems requires intelligent management of highly volatile stored energy. As edge applications grow in complexity, traditional energy-aware schedulers struggle with unpredictable workloads due to their reliance on static execution thresholds or pre-measured, hardware-specific task profiles. To overcome this, we propose two novel, hardware-agnostic dynamic scheduling strategies treating applications as a "black box," requiring no prior energy information: a model-free Reinforcement Learning (RL) agent and an on-the-fly Approximated Prediction (AP) method. We evaluate these methods against an adaptive task rate approach (AsTAR) and optimized static thresholds using a custom-built, physically accurate simulation framework driven by real-world solar data and dynamic LoRa transmission profiles. Rather than claiming universal superiority, our analysis exposes the distinct operational trade-offs of each method: the AP approach delivers lightweight, near-oracle task throughput; the RL agent provides tunable survival-execution balancing; and AsTAR excels at execution pacing across long energy gaps. Finally, we demonstrate that while these advanced strategies provide critical resilience for severely constrained systems with small capacitors, devices with larger energy buffers can efficiently rely on simpler, less computationally expensive static policies.

Czech has been part of Universal Dependencies since its first release in 2015. It has also been one of the best represented languages, with the Prague Dependency Treebank being order of magnitude larger than most other UD treebanks. More recently, three other datasets from the Prague family were added and the annotations thoroughly revisited, forming the "Prague Dependency Treebank-Consolidated" (PDT-C). In comparison to the original PDT, PDT-C is more than twice as large, but it is also much more diverse in terms of genres and domains. In this paper, we describe the conversion of the new resource to Universal Dependencies. While the two annotation schemes are relatively similar at the first sight, there are numerous small differences in topology of the dependency structures and in granularity of the POS and relation type inventories. We demonstrate a selection of such differences on examples, discuss the diverging motivations, as well as ways to overcome the differences during conversion. We argue that while PDT is less "universal" and more tightly bound to one language, its multi-layer annotation is rich and provides all information needed for basic UD trees, and much more.

Transformer-based language models have become the default substrate for natural language processing and the pace of new releases has made it hard for practitioners to separate durable ideas from the noise of incremental announcements. This review works at two levels. At the level of mechanism, we organise the main transformer families into a working taxonomy, covering encoder-only, decoder-only, encoder-decoder, long-context, permutation-based, and generator-discriminator variants. We then extend the discussion to post-2023 developments that changed the picture in practice: instruction tuning, reinforcement learning from human feedback, direct preference optimisation, mixture-of-experts scaling, retrieval augmentation and the current flagship model families from OpenAI, Anthropic, Google, Meta, Mistral and DeepSeek. At the level of use, we survey deployments across healthcare, finance, legal, education, customer service, creative writing and scientific work. Based on this we link each to the specific capabilities that make a transformer the appropriate tool. The contribution of this paper is a critical assessment that is based on the survey. We compare architectures on four axes that matter to deployment decisions, we quantify the trade-off between parameter count and energy cost. We also discuss how alignment methods, data provenance and benchmark saturation change what it means to call a model "state of the art". The final section lists the research questions that we think deserve more attention.

The Prague Dependency Treebank framework is unique in its attempt to systematically include and link different layers of language, including a meaning representation with several types of inter-sentential phenomena, especially coreference and discourse relations. We present its second consolidated version (PDT-C 2.0), which concludes almost 30-years long project of sustained development of the resource to a uniformly and coherently annotated, genre-diversified, almost 4 million token language resource of Czech language, with accompanying fully compatible lexicons. In addition to continuous linguistic research, the richly linguistically annotated corpus is also widely used in international comparisons of the development of traditional and novel NLP tools as well as in conversions into other formalisms. The corpus and the trained parsers are available under the CC BY-NC-SA licence.

We present ZONOS2 8B, our latest TTS model, which achieves state-of-the-art naturalness, prosody, and voice cloning fidelity. We improve upon Zonos-v0.1 across scale, data, and training recipe. We scale the model from 1.6B to 8B total parameters (900M active) with a novel mixture-of-experts (MoE) backbone, improving inference latency and throughput. We expand our training corpus from 200K to over 6M hours using a new data processing pipeline, and we simplify our post-training and conditioning recipes to improve naturalness and voice cloning fidelity. We evaluate ZONOS2 8B on quality, speaker similarity, WER, and ZTTS1-Eval, our novel TTS benchmark, where it performs competitively with state-of-the-art systems while maintaining good streaming latency. We release our model weights and example inference code under an Apache 2.0 license on GitHub and Hugging Face.

AI-driven image-to-image synthesis is rapidly advancing, with growing applications in medical imaging. Multi-modal image analysis plays a crucial role in optimizing examination quality, yet acquiring multiple imaging modalities in clinical settings remains resource-intensive and time-consuming, especially for 3D imaging. To address this challenge, we propose a novel image-to-image translation model based on Brownian Bridge Diffusion Models (BBDM), which synthesizes magnetic resonance imaging (MRI) sequences from 2D axial slices. Our approach integrates a variational encoder-guided diffusion mechanism, leveraging probabilistic image distributions to enhance synthesis quality. Evaluated on the BraTS 2021 dataset, our Probabilistic-BBDM (Prob-BBDM) achieves superior performance across multiple translation tasks, reaching up to 88.46% SSIM and 26.09 dB PSNR, with consistent improvements over baselines. Notably, our diffusion process requires only 4 steps, making it computationally efficient while maintaining high-quality synthesis. To further validate generalizability, we test Prob-BBDM on an external third-party dataset, demonstrating consistent performance across domains. Additionally, we assess the clinical utility of the synthesized slices by using them as input to a pre-trained segmentation model. Tumor segmentation yields a Dice score of 88.71% and an HD95 of 3.49 mm, confirming that the synthesized slices preserve critical diagnostic information. These results highlight the potential of Prob-BBDM for high-quality, efficient, and generalizable MRI synthesis, offering a promising step toward improved medical image translation.

As large language model (LLM) agents are applied to longer tasks, they increasingly modify workspace state across multiple rounds of iteration. However, agents typically observe only tool outputs and log fragments, while the actual state changes occur in the file system. Without explicit workspace boundaries, state-changing operations such as file writes and temporary artifact generation may scatter changes across paths. Over time, these weakly constrained changes accumulate, making states such as modified files difficult to track. This paper presents LemonHarness, an integrated execution framework for long-horizon agents. LemonHarness establishes an explicit execution boundary by constraining state-changing operations within a clearly defined workspace and bringing model invocation, tool execution, and rule knowledge within a single controlled boundary. State-changing operations, including file writes, dependency installation, and temporary artifact creation, are executed through structured tool interfaces, with execution feedback recorded as observations available to subsequent model decisions. The system also introduces a reusable rule knowledge base, which turns recurring execution rules and acceptance criteria into runtime knowledge. LemonHarness further adds a time-aware execution mechanism that exposes elapsed and remaining budget to the model, so it can rebalance exploration, implementation, and validation effort as time pressure shifts and avoid timeouts from long waits or excessive verification. On Terminal-Bench 2.0, LemonHarness_GPT-5.3-CodeX reached 84.49% accuracy over 445 trials; pairing the same framework with the stronger GPT-5.5 backbone raised the average accuracy to 86.52% across five jobs. The results suggest that a unified runtime boundary, callable rule knowledge, and time-aware execution can improve the stability of long-horizon agent execution.

Interactive music and live performance relies on real-time human expression, but modern generative music AI remains largely absent from this domain due to its prohibitive inference latency and offline rendering paradigm. To provide pioneer musicians with a novel medium for interactive composition, we should fundamentally change these static models into dynamic, playable instruments. In this paper, we propose a framework that bridges this gap. To achieve the low latency required for live interaction without sacrificing structural coherence, we formulate distillation within a streaming autoregressive latent space. Our approach gets rid of the need for expensive paired audio-latent datasets by utilizing prompt-only inputs to synthesize teacher-guided, chunk-wise trajectories on the fly. Because live instruments require high acoustic fidelity, we introduce music-aware consistency objectives, which combine latent, spectral, and temporal-difference losses, to preserve crucial qualities like timbre, transients, and rhythmic stability during accelerated single-step streaming generation. Implemented via parameter-efficient adaptation, our distillation reduces generation steps to achieve a low real-time factor. Crucially, by operating as a continuous autoregressive stream, the system can seamlessly assimilate dynamic human inputs on the fly, allowing users to instantly steer the musical trajectory without interrupting the audio flow. Ultimately, this work recontextualizes generative text-to-music models not as passive prompt-and-wait systems, but as responsive instruments, opening new frontiers for live human-AI musical co-creation.

The increasing interest in data-driven methods for power system protection is accompanied by a lack of standardized, publicly available high-voltage waveform datasets that enable transparent and reproducible evaluation. To address this gap, this paper introduces the PROTECT-90 dataset, an open electromagnetic transient (EMT)-simulated reference benchmark for high-voltage fault studies with consistent digital-fault-recorder-like measurements, publicly released with this work. The dataset comprises 9,022 physically consistent short-circuit simulation episodes generated on a standardized 90 kV double-line topology with systematically documented domain randomization of grid operating points, line parameters, and fault conditions. For each episode, synchronized three-phase voltage and current waveforms are recorded at eight measurement locations and released together with structured, machine-readable metadata describing fault type, fault location, inception time, and operating conditions. All modeling assumptions, parameter ranges, and data-generation procedures are explicitly documented to ensure transparency and cross-study comparability. By combining physically grounded EMT simulation, balanced scenario coverage, and open accessibility, PROTECT-90 establishes a standardized foundation for reproducible benchmarking of protection-oriented signal processing and learning-based methods.

Multimodal Large Language Models have achieved remarkable progress in short-form audio-video understanding, yet long-form audio-video comprehension remains challenged by limited context windows and severe information redundancy. To address these bottlenecks, we propose AVOC, a framework for long-form audio-video understanding in Omni-modal Large Language Models. AVOC introduces a learnable token compression module between the modality encoders and the LLM backbone. We reframe multimodal token compression as a top-$K$ retrieval problem: given a fixed context budget, the module must retrieve a compact subset of tokens that best supports answering the user query. We draw inspiration from three classical Information Retrieval criteria for selecting informative units from a large candidate pool: relevance, importance, and diversity. AVOC instantiates each criterion as a tailored mechanism for audio-video understanding, and integrates them into a unified retrieval-style compression pipeline. Experiments show that AVOC achieves state-of-the-art performance on long-form audio-video benchmarks, surpassing the second-best model by 4.9 and 5.5 points in average accuracy on OmniVideoBench and LVOmniBench, respectively. Moreover, AVOC maintains robust performance on Audio-Video Needle-in-a-Haystack task at durations up to one hour.

Reasoning language models are increasingly asked not only to answer difficult questions, but also to estimate their likelihood of success. Existing methods typically elicit confidence only once: either before thinking or after answering. We argue that confidence in reasoning models is state-dependent: before thinking, confidence should estimate the chance of the model correctly solving the prompt, while after thinking it should predict whether the realized answer is likely to be correct. This distinction determines the appropriate supervision target: prompt-level success should supervise confidence estimates made after seeing the prompt, while individual answer-level correctness should supervise confidence estimates made after answering. We introduce CALIBER (Calibration Before and After Reasoning), which elicits both estimates and supervises each with the target matched to its information state. Under this unified protocol, CALIBER reduces Expected Calibration Error (ECE) by 52.5% over the strongest single-confidence baseline on BigMathDigits for the 7B model, while achieving the best Brier score and AUROC, and remains within 2.1 points of the best accuracy. Further, on a larger 30B model, CALIBER achieves the best ECE on BigMathDigits while remaining competitive in Brier score and AUROC. Out of distribution, it achieves the best ECE and Brier score on GPQA and TriviaQA, and remains competitive on SimpleQA. Ablations further show that this position-target alignment is most beneficial under distribution shift where it consistently reduces calibration error across all out-of-distribution benchmarks.

In Description Logics (DLs), reasoning under Rational Closure (RC) is a well-known and widely accepted non-monotonic formalism to handle defeasible knowledge. In this paper, we study the application of RC to the core and horn variants of the DL-Lite family of lightweight description logics. We analyze both entitlement (instance checking) and Conjunctive Query (CQ) answering under RC. Our main contribution is providing a plug-in architecture that builds upon existing standard classical reasoners, establishing that reasoning and CQ answering under RC for DL-Lite can be done efficiently with minimal computational overhead.

In this paper, we introduce two neural-network-based numerical schemes for solving systems of coupled ergodic Backward Stochastic Differential Equations (eBSDEs), motivated by the approximation of optimal strategies within the framework of forward utilities in a regime-switching stochastic factor model. Our approach builds on the representation of such models through systems of eBSDEs introduced in [HLT20]. We first establish a link between the solution of the system of ergodic BSDEs and that of an associated multidimensional BSDE with random terminal time, given by the hitting time of the positive recurrent stochastic factor. Building on this representation, we introduce a locally additive deep learning scheme obtained by minimizing aggregated local error terms. We then present a new Deep Galerkin Method (DGM) inspired algorithm that minimizes the residual of the associated ergodic PDE system, relying on a representation of the ergodic cost. Finally, we apply this framework to regime-switching forward utilities in a stochastic factor model. We first derive a general consistency SPDE that characterizes regime-switching forward utilities and retrieve their representation with systems of ergodic BSDEs in the homothetic case. Numerical experiments demonstrate the performance of the proposed methods, with a particular focus on the impact on forward preferences of taking into account regime switches.

While in-context learning is generally shown to be effective in Large Language Models (LLMs), bad contexts can cause performance degradation and mode collapse, a phenomenon we call "pigeonholing." **Unintentionally bad** contexts can happen without malicious jailbreaking intents: For example, a user asks the model to justify an incorrect math theorem or fails to correct the model's buggy code. Specifically, we investigate ``pigeonholing" in two scenarios: (1) when the user suggests a solution, and (2) when the conversation context includes the assistant's previous (incorrect) responses. Our experiments across 10 verifiable and open-ended tasks with 10 different models show that pigeonholing manifests in several ways: (1) repeating the incorrect answers from context (leading to 38-40% performance drop), (2) converging on a narrow set of answers in coding and text generation without exploring alternatives, and (3) flipping stance on controversial topics to align with the user or the assistant's previous claims. We find that pigeonholing worsens almost monotonically with the number of conversation turns (performance drops by additional 14+% as repeated mistakes increase from 1 to 5), and pigeonholing-induced mode collapse can happen even when the provided example is correct. As a step toward mitigation, we propose RLVR with synthetic errors which improves models by 43-60% under bad contexts compared to vanilla RLVR baselines.

Numerical simulations in industrial applications often require performing numerous high-precision computations parameterized by specific experimental conditions. For instance, in vehicle body design, aerodynamic simulations are essential for evaluating the aerodynamic characteristics of various proposed body geometries. However, computational resource constraints often become a bottleneck. Therefore, achieving the desired accuracy while minimizing computational cost is crucial. To address this challenge, model reduction methods have been developed to decrease the degrees of freedom by constraining the possible states of a physical system to a lower-dimensional subspace. In particular, reduction techniques that project the system onto a nonlinear subspace using neural networks have been actively studied. Our previous research developed a reduced-order model that integrates neural-network-based model reduction with a time-evolution method, implemented as a distributed parallel training framework to process high-resolution flow field data efficiently. In this study, we extend this reduction approach by incorporating a variational autoencoder to assess its robustness in high-Reynolds-number flows around multiple vehicle bodies with varying geometries. Specifically, we evaluate the reconstruction accuracy of vortex generation across different spatial and temporal scales using a compact latent representation, with a particular focus on the flow behavior near the rear end of the vehicle body.

Microwave satellite imagery plays a crucial role in monitoring tropical cyclone precipitation and intensity worldwide, but suffers from long revisit times, potentially missing rapid storm evolution phases. While this raises the need for an interpolation method, it is made challenging by the high level of heterogeneity of microwave data coming from different instruments. In this work, we introduce the first generative model that can be applied to multiple geospatial sources that change across samples, occur at irregular time intervals, are misaligned geographically, and come from instruments with varying characteristics. We apply this model to the case of spatio-temporal interpolation of tropical cyclone microwave images from other microwave and infrared instruments. We train using a self-supervised task in which a random source is masked and reconstructed, and show that it leads to a significant decrease in Continuous Ranked Probability Score over supervised training. We show a further improvement by combining infrared and microwave data compared to microwave only. Using these improvements, the generative model produces an ensemble mean on par with that of a deterministic model, while generating a power spectrum significantly closer to that of true observations. To the best of our knowledge, this is the first generative model that interpolates microwave images of cyclones by combining multiple microwave instruments and infrared observations at irregular time intervals.

Addressing problems beyond classical computing limits is sparking an increasing interest in Quantum Computing. However, despite their adequacy to address specific problems, quantum algorithms cover a limited subset of the functionality required in real-world computing systems. Additionally, they require expensive specialized hardware. To overcome this issue, hybrid (quantum-classical) software systems are emerging as a promising way to integrate both computing paradigms by applying the principles of Service-Oriented Architectures (SOA). Still, the design and deployment of hybrid service-based systems faces unique challenges like the idiosyncrasies and constraints of NISQ computers (e.g., algorithms that can only run in specific machines, disparate quality attribute metrics), and the management of structural and behavioural properties of service-based applications. From the SOA perspective, architectural decisions need to be made by performing a trade-off analysis and providing quantitative guarantees of system configurations under prescribed levels of uncertainty. In this paper, a method to explore the design space of quantum-classical applications is provided by a formalization of an architectural style of hybrid applications. The obtained results demonstrate that the proposed method successfully identifies decision boundaries. It enables the dynamic selection of the most suitable hybrid or classical configuration based on the user's QoS criteria.

Fine-tuned encoders deployed across heterogeneous NLP tasks face three compounding problems: mismatched inductive biases, class-imbalance corruption of feature statistics, and no mechanism to condition attention on external lexical knowledge. We introduce \textbf{\surgellm}, a unified transformer framework that addresses each with a dedicated lightweight module: a \emph{surgical feature gate} (learned per-dimension sigmoid over curated lexical indicators and \texttt{[CLS]}; provably degenerates to identity when features are uninformative), \emph{task-conditioned prefix tokens} (quantized feature values and task identity prepended to every input), and \emph{Instance-Weighted Normalization} (IWN; removes class-prior bias from gate statistics). We prove an excess-risk bound linking gate benefit to \emph{surgical feature alignment}. Across four tasks, SST-2, multi-hop retrieval, LLM-prompt attribution, and authorship detection, covering 17,830 examples and eleven model variants over three seeds, the IWN variant achieves macro-F1 \textbf{0.940} ($+0.036$ over the strongest non-IWN baseline; $+0.130$ on authorship detection). A random-vocabulary control ($-0.028$ avg.\ F1) confirms gains are lexical, not parametric. Code, vocabularies, and a $99.5\%$-recovery auto-extraction recipe are released.

Although text-to-motion generation has achieved strong progress in synthesizing realistic single-person motions from language, extending it to text-driven 3D human-human interaction (HHI) remains non-trivial, as HHI requires modeling the underlying \textbf{social structure} that governs phase progression, actor roles, and inter-actor coordination. In this paper, we formulate HHI generation as a social structure modeling and grounding problem: the model must first infer how an interaction unfolds and how the two actors coordinate their roles, and then realize this structure as continuous, physically plausible, and partner-aware 3D motion. To study how such structure should be modeled, we first examine the capability boundary of large language models (LLMs) for HHI generation. Our analysis shows that LLMs can \textit{think} by recovering phase decompositions and partner-aware roles, but cannot directly \textit{move}, as they fail to generate dynamic, physically plausible, and interaction-aware motion. This motivates our planner-executor paradigm, \textbf{Think with LLM, Move with Motion Skill}. The LLM planner converts implicit interaction semantics into motion-aligned social supervision by decomposing interactions into phases, assigning partner-aware actor roles, and aligning them with motion sequence. The motion executor then grounds the planned social structure into coordinated two-person motion by adapting a pretrained solo motion model with LoRA, previous-phase self-conditioning, and ego-relative partner conditioning. Together, our Solo-to-Social framework bridges social organization and motion realization, producing 3D HHI with improved phase consistency, role alignment, and partner-aware coordination.

Large language models exhibit a growing range of misaligned behaviors such as strategic deception, sandbagging, and self-preservation. As they are increasingly deployed in high-stakes settings, it is critical to reliably detect such behaviors to ensure safe and responsible use. In this work, we propose to monitor misalignment by decomposing it into fine-grained cognitive processes -- misalignment indicators -- and detecting their presence in a model's internal activations via linear probes. We develop a taxonomy of 18 indicators spanning different misaligned behaviors, paired with an automated, meta-plan-guided pipeline that generates multi-turn training conversations. To rigorously evaluate generalization, we construct an out-of-distribution suite combining automated behavioral elicitation, established misalignment benchmarks, and natural benign conversations. Across 5 misaligned behaviors, our probes match a strong LLM judge with 0.935 AUROC on out-of-distribution benchmarks while keeping a low false positive rate on benign traffic. We further perform in-depth analysis to understand the probes and the model's internal representations of misalignment indicators.

Large language model (LLM) agents increasingly automate complex tasks by integrating language models with external tools and environments. However, their autonomy poses significant safety risks: agents may execute destructive commands, leak sensitive data, or violate domain constraints. Existing safety approaches face a fundamental tradeoff: hand-crafted rules are interpretable but brittle, with overly conservative rules blocking safe operations (high false positives) while permissive rules miss unsafe behaviors (high false negatives). Neural classifiers lack the interpretability required for safety-critical deployments. We present AutoSpec, a framework that automatically evolves deployed expert-designed safety rules from user safe/unsafe annotations through counterexample-guided inductive synthesis (CEGIS) guided by inductive logic programming (ILP). Starting from the expert rules and a stream of annotated traces, AutoSpec iteratively evaluates rules, mines false-positive and false-negative counterexamples, uses ILP to learn which predicates discriminate them, generates candidate rule edits, and verifies candidates to select the best revision. The key insight is that ILP efficiently identifies predicates that appear frequently in false negatives but rarely in false positives (or vice versa), dramatically pruning the exponential search space of rule edits. This continues until convergence, producing interpretable rules that balance precision and recall. We evaluate AutoSpec on 291 execution traces spanning code execution and embodied agent domains. AutoSpec raises rule F1 to 0.98 and 0.93 across the two domains, achieving up to 94% false positive reduction while maintaining high recall, and converges within 4-5 iterations. The ILP-guided approach achieves up to 4.8x higher F1 than heuristic CEGIS. The learned rules are human-readable, auditable, and generalize to unseen scenarios.

Federated Graph Learning facilitates collaborative graph modeling across distributed clients while preserving data privacy. However, real-world data categories frequently exhibit long-tailed distributions. Such statistical scarcity severely degrades performance in two ways: it biases the global model toward majority classes, and it structurally isolates minority nodes by submerging them in heterophilic, head-dominated neighborhoods. While existing methods attempt topology-agnostic statistical compensations, they often fail under data scarcity. Instead of recovering tail nodes, they overfit the structural noise from adjacent dominant classes, leading to representation degradation. To address these limitations, we propose FedEPD, a framework built on a dual decoupling paradigm that separates topological purification from semantic recalibration. Specifically, FedEPD utilizes distribution-aware Dirichlet energy pruning to filter spatial heterophilic edges. It then overcomes Non-IID distribution shifts by extracting robust global prototypes from topologically central nodes, which are incorporated into local representations via a spatial low-pass prototype injection. Furthermore, a two stage alternating optimization strategy strictly protects majority decision boundaries while improving minority accuracy. Extensive experiments demonstrate that FedEPD achieves state-of-the-art performance across diverse long-tailed benchmarks, yielding absolute improvements of up to 4.97% in Accuracy and 5.48% in Macro-F1.

Visual assessment of residual plots is a common approach for diagnosing linear models, but it relies on manual evaluation, which does not scale well and can lead to inconsistent decisions across analysts. The lineup protocol, which embeds the observed plot among null plots, can reduce subjectivity but requires even more human effort. In today's data-driven world, such tasks are well suited for automation. We present a new R package that uses a computer vision model to automate the evaluation of residual plots. An accompanying Shiny application is provided for ease of use. Given a sample of residuals, the model predicts a visual signal strength (VSS) and offers supporting information to help analysts assess model fit.

Spatial proteomics enables single-cell-resolution characterization of protein expression within tissue architecture, playing a critical role in understanding tumor microenvironments and guiding precision medicine. However, current analysis workflows remain fragmented, requiring expert manual orchestration of heterogeneous tools and limiting research scalability and reproducibility. We present SP-Mind, the first autonomous AI agent designed to unify the spatial proteomics analysis pipeline, from raw multiplexed tissue imaging to downstream phenotype discovery. Equipped with expert-curated biological analysis skills and specialized computational tools, SP-Mind converts natural-language queries into end-to-end analytical workflows without task-specific fine-tuning. To rigorously evaluate its capabilities, we introduce SP-Bench, a comprehensive benchmark spanning diverse tissue types, comprising 102 tasks across 18 distinct categories. Through extensive evaluation on SP-Bench and established downstream tasks, SP-Mind achieves state-of-the-art performance compared to existing open-source biomedical agent baselines.

Multimodal driving planning faces a long-standing tension between two paradigms: scoring-based methods benefit from dense reward supervision but are confined to a fixed action vocabulary, while anchor-based methods generate proposals dynamically yet suffer from sparse supervision constrained to a single ground-truth trajectory. In this work, we propose FlowR2A, which resolves this tension by reframing simulation-based rewards from discriminative targets into generative conditions. By learning the reward-conditioned action distribution from dense trajectory-reward pairs with a flow-matching decoder, FlowR2A unifies the dense supervision of scoring-based methods with the proposal generation of anchor-based methods in a single generative model, forcing the model to internalize the correlation between an action and its outcomes in safety, progress, comfort, and rule compliance. To balance hard safety constraints against soft progress objectives, we introduce fine-grained per-timestep reward conditioning and reward noise augmentation. The generative formulation naturally supports controllable test-time sampling via reward guidance and anchored sampling, producing high-quality proposals. FlowR2A achieves state-of-the-art results on the NAVSIM v1 and v2 benchmarks, with multimodal proposals of substantially higher quality than prior methods.

Despite the extensive discussions of human-centric AI (HCAI) in Industry 5.0, its effects on firms' idiosyncratic risks (IR) remains underexplored. This is an imperative issue for firms navigate financial risks during the current technological revolution, as IR reflects investor reactions to corporate heterogeneous AI strategies and implementations by isolating firm-level stock volatility from systematic factors. Integrating situated AI theory with social-technical systems theory, we conceptualise HCAI as a situated AI strategy that reduces AI-related ethical risks and fosters AI-Human synergies in firms' business operations, ultimately reducing IR by aligning with stakeholders' diverse expectations. Moreover, socio-technical factors, namely digitalisation, operational efficiency, executive shareholding, and CEOs with IT background, may moderate the HCAI-IR relationship. Using a multi-source panel dataset of Chinese listed firms from 2015 to 2023, we find that HCAI is associated with lower firm IR. Furthermore, digitalisation and executive shareholding strengthen this risk-reducing effect, whereas operational efficiency and CEOs with IT background surprisingly attenuate it. Our findings offer theoretical contributions and practical insights for both ethical AI governance and firm financial risk management in the AI era.

Stochastic quantum neural networks (SQNNs) encode neuronal activations as qubits, synaptic topology as entanglement, and neural noise through a Lindblad master equation. A recent conference study applied a ring-entangled SQNN to collaborative intrusion detection and reached three conclusions: ring entanglement is \emph{essential} for non-local anomaly detection; an adversarial-resilience bound holds but is \emph{conservative}; and the depolarising channel \emph{fails} to act as a dropout-style regulariser, behaving instead as output noise. It left open whether a per-gate stochastic deactivation (``true quantum dropout'') could regularise where the depolarising channel could not, and whether the loose robustness bound could be replaced by a predictive theory. This paper resolves both and extends the framework to real data and to neutral-atom hardware. We give an $N$-qubit formulation through the stochastic master equation and its vectorised Liouvillian, and prove a \emph{decoherence-contraction theorem}: a depolarising channel of strength $γ$ over $L$ entangling layers contracts every weight-$w$ Pauli read-out by a factor $(1-4γ/3)^{wL}$ (for the weight-$1$ read-out used here, $(1-4γ/3)^{L}$); building on the general noise-as-defence result of Du et al., we make this quantitative and operational for intrusion detection. On the real NSL-KDD dataset under white-box FGSM and PGD attacks, a depolarising SQNN trained with the channel is, over seven seeds under strong $\ell_\infty$/$\ell_2$ attacks, significantly more robust than the noiseless circuit ($\ell_\infty$ PGD-$20$, $p=0.04$, large effect) and, critically, never suffers the catastrophic robustness collapse that the noiseless model and gradient-trained classical detectors (which fall from $95\%$ to $47\%$) do, cutting robustness variance roughly twofold; we show this robustness arises from a noise-reshaped training boundary rather than from attack-time gradient contraction. For generalisation, we derive an adaptive-penalty formula showing that per-gate dropout implements a curvature-weighted $L_2$ penalty $\tfrac{p(1-p)}{2}\sumθ^2\partial^2_θL$ in weight space, maximised at $p=1/2$, whereas depolarising noise implements an output-space penalty. A $30$-seed study confirms the formula's quantitative prediction: both mechanisms reduce the train-test gap by a small but statistically significant margin ($\approx\!0.01$; $p<10^{-4}$ and $p=0.004$), are statistically indistinguishable from each other, and the effect is concentrated where overfitting is largest; increasing the dropout rate past $1/2$ does not help, as the formula predicts. The single-seed dichotomy of prior work does not survive replication. We close with a neutral-atom realisation and a feasibility-by-$N$ analysis.

Recent breakthroughs in 3D generation have advanced notably with the development of text-to-image diffusion model. However, existing methods remain two practical challenges: (1) They primarily generate single 3D object, but struggle to generate multi-object compositional 3D assets due to the lack of the modeling for Gaussian primitives in reasonable interactions. (2) They often suffer from cross-view inconsistency during 3D optimization, as Score Distillation Sampling inherently performs on each single view, inevitably resulting in cross-view hallucinations. To solve above issues, we propose I2C-3D, a novel optimization-based method to generate multi-view consistent compositional 3D assets with reasonable interactions. Specifically, we propose an Inclusive Interactive Collisions strategy to guide Gaussian primitives appearing in reasonable interaction regions naturally, thereby ensuring objects in the compositional scene interact in a physically plausible and visually coherent way. Additionally, to enhance multi-view consistency, Multi-View Adaptive Score Distillation Sampling is devised to distill multi-view consistency prior and layout prior from pre-trained diffusion model by modulating attention map of instance token and spatial token across viewpoints. Benefiting from above elaborate designs, I2C-3D not only generates high-fidelity multi-view consistent compositional 3D assets but also supports 3D editing flexibly, facilitating complex scene generation. Extensive experiments demonstrate our I2C-3D outperforms existing methods in generation quality and multi-view consistency.

Retrieval-augmented generation (RAG) in clinical settings increasingly requires multilingual retrieval against predominantly English evidence corpora. Multilingual medical retrieval demands three capabilities: cross-lingual alignment, concept discrimination, and evidence retrieval. However, existing benchmarks evaluate these only in isolation, leaving the interaction between biomedical expertise and multilingual coverage unmeasured. We introduce MMed-Bench-IR, a benchmark designed to disentangle these axes across 6 languages and three structurally heterogeneous tasks: (1) cross-lingual medical QA retrieval with 6,127 queries grounded in the Unified Medical Language System (UMLS), (2) concept discrimination over 4,975 confusion sets at three difficulty tiers, and (3) multilingual evidence retrieval for RAG with 2,040 quality-assured queries. The three tasks share zero concept and query overlap by design, ensuring that aggregate scores reflect genuine capability breadth. Evaluation of ten systems across six paradigm families reveals severe cross-lingual failure: biomedical encoders that score 0.818 nDCG@10 in English drop to 0.056 in Japanese, a gap that English-only benchmarks cannot detect.

Modern AIGC pipelines deliver high-fidelity images and videos but presuppose a well-formed creation instruction, while end users rarely articulate visual details, leaving generators misaligned with user demand. We study personalized content generation, which turns a user's interaction history into an executable instruction for downstream synthesis, and identify two obstacles: behavior must be encoded in a form legible to language reasoning, and the model must acquire instruction-writing skill absent from both pretraining and behavior data. We propose NaviGen, which represents each item with a dual identifier coupling a collaborative code and a textual code as a behavioral substrate and a semantic bridge in one token stream. On this representation, a two-stage SFT+RL pipeline first distills preference reasoning and instruction writing from evolutionarily searched supervision, then aligns generation with user intent through hierarchical and self-consistent rewards. Experiments across product, game, and short-video domains show that NaviGen improves personalized image and video generation, strengthens next-item prediction, and yields more specific, relevant, and visually generatable instructions. Our code is anonymously released at: https://github.com/iLearn-Lab/NaviGen.

Conversational product search assistants offer a more expressive, natural, and interactive alternative to traditional keyword-based product search. With limited screen space, showing only a few items increases the need for precise preference elicitation, which can prolong conversations, leading to user frustration and session abandonment. Conversely, rushing to recommend items without a clear understanding of preferences risks poor matches and a degraded user experience. We present Dialogue to Discovery (D2D), an attribute-oriented preference elicitation framework that dynamically exploits the structure of product attributes to efficiently steer conversations toward the user's desired item. D2D adaptively prioritizes the most informative queries and strategically times product recommendations, reducing premature or off-target suggestions that harm engagement. To evaluate D2D, we curate three datasets from the Amazon Reviews corpus. In simulated conversations modelled using a multi-factor utilitarian patience framework, D2D achieves a 22.2-29.9% improvement in target-finding accuracy, 6.6-16.1% reduction in abandonment, and 27.5% shorter average conversations over the state-of-the-art baselines. A complementary user study further confirms significant gains in both user satisfaction and perceived efficiency.

Unlearning has emerged as a key technique to mitigate harmful content generation in diffusion models. However, existing methods often remove not only the target concept, but also benign co-occurring concepts. As illustrated in Fig.1, unlearning nudity can unintentionally suppress the concept of person, preventing a model from generating images with person. We define these undesirably suppressed co-occurring concepts that must be preserved CARE (Co-occurring Associated REtained concepts). Then, we introduce the CARE score, a general metric that directly quantifies their preservation across unlearning tasks. With this foundation, we propose ReCARE (Robust erasure for CARE), a framework that explicitly safeguards CARE while erasing only the target concept. ReCARE automatically constructs the CARE-set, a curated vocabulary of benign co-occurring tokens extracted from target images, and leverages this vocabulary during training for stable unlearning. Extensive experiments across various target concepts (Nudity, Van Gogh style, and Tench object) demonstrate that ReCARE achieves overall state-of-the-art performance in balancing robust concept erasure, overall utility, and CARE preservation.

Mining sentiment information from the textual content of peer review comments offers valuable insights into the scientific evaluation process. However, previous studies are often constrained by coarse-grained analysis and the lack of differentiation across review rounds. Notably, the dynamic shifts in reviewers' focus and sentiment tendencies throughout multiple review stages remain underexplored. To address this gap, the present study investigates the distribution and evolution of aspect-level sentiments and examines their correlation with the number of review rounds. We begin by segmenting the multi-round review comments of 11,063 accepted papers from Nature Communications and identifying fine-grained review aspect clusters. A manually annotated corpus of approximately 5,000 review sentences is then constructed. Using this dataset, we train a series of deep learning-based aspect sentiment classification models. Among them, the LCF-BERT-CDM model achieves the best performance, with a Macro-F1 score of 82.65%. Subsequent statistical analysis reveals a consistent trend: as the number of review rounds increases, the proportion of positive sentiments rises, while negative sentiments decline. Correlation analysis further indicates that aspect sentiment scores are negatively associated with the total number of review rounds. Key aspects exhibiting stronger correlations include "experiments", "research significance" and "result analysis".

Retrosynthetic planning seeks to connect a target molecule to commercially available starting materials through a multistep route. Classical planners construct such routes by iteratively applying single-step reaction models within a search procedure; constrained variants often require specialized algorithms or architectural changes. Direct route generation reframes retrosynthesis as sequence generation, but existing direct-generation methods still train separate models for different planning specifications. We introduce Ariadne, a decoder-only route generator that represents the target, optional constraints, and route in one prompt-completion sequence. On the RetroCast/PaRoutes mkt-cnv-160 benchmark family, one 24-layer checkpoint follows route-depth and required-starting-material prompts: adding the corresponding prompt fields raises Solv-0 by 13.7 points for depth constraints and 31.2 points for required-leaf constraints. Ariadne also improves over DESP, a bidirectional search planner, on required-leaf Top-10 and Solv-0 in 24 GPU-minutes versus 6.8 GPU-hours. On standard reconstruction, Ariadne is comparable to DMS Explorer XL at about half the reported inference time. Across additional target-only benchmarks, Ariadne's clearest gains are on route-holdout reconstruction, whereas AiZynthFinder MCTS remains stronger on several Solv-0 comparisons. These results extend sequence generation from specialist retrosynthesis models to prompt-conditioned structural route generation. We release the codebase and training scripts to support further work, but do not introduce Tier-1--3 route checkers; those remain the main bottleneck before models of this kind can become useful to experimental chemists.

Three-dimensional scene completion has evolved as a major problem in computer vision and robotics, and its applications are diverse, including autonomous navigation and augmented reality. In this study, a systematic review has been conducted to compile the research contributions made in the last ten years, i.e., 2016 to 2026, which has revolutionized the field from the voxel semantic completion paradigm represented by SSCNet to the latest paradigm that combines generative diffusion priors with real-time rendering using a Gaussian splatting technique. The evolution in representation paradigms, such as voxel grids, point learning, implicit neural fields, transformer networks, diffusion networks, and the latest paradigm based on rendering-aware 3D Gaussian primitives, has been discussed in this study. A comprehensive analysis has been carried out on the contributions made in the last ten years, and a taxonomy has been developed to provide a clear idea about the contributions made in the field. The study has also discussed the research contributions made in the field, along with the challenges that still need to be addressed. Finally, the study has presented a research agenda that will provide a clear idea about the directions that can be followed in the development of the next-generation system

Pretrained vision models often misclassify inputs that are rotated, scaled, or sheared, even though these affine transformations leave the object class unchanged. Robustness is usually restored either by building equivariance into the architecture or by retraining with augmentation, both of which require changing or retraining the model. Test-time canonicalization instead leaves the classifier untouched. It undoes the transformation of each input, mapping it to a canonical form near the training distribution before classification. Existing canonicalizers, however, rely on a narrow set of logit-based energy scores and bespoke search procedures, leaving the design space of scoring functions and optimizers unexplored. We reframe canonicalization as out-of-distribution (OOD) detection, which lets any OOD score serve as the energy minimized over transformations. Across benchmarks ranging from handwritten characters and sketches to natural images and 3D point clouds, we systematically evaluate around twenty OOD scores and nine search algorithms, finding that distance-based scores paired with random search and local refinement perform best overall. Because canonicalizing an already-aligned input can hurt accuracy, we add a gated mechanism that transforms an input only when its OOD score indicates this is needed, preserving most in-distribution accuracy while retaining the robustness gains on transformed inputs. Code is available at github.com/johschm/its.

Large language models are making research production scalable, shifting the bottleneck from producing artifacts to judging claims. We present \textsc{Agon}, a research orchestrator that validates what can be checked inside the workflow and leaves the remaining judgments to human scientists. \textsc{Agon} is built on six design principles: Prompt Economy, Future-Facing, Minimal Prompts, OmniDisciplinary, Massive Parallelism, and Zero-Code. We ran \textsc{Agon} across domains for 444 iterations of Prompt Economy loops, using only small starting topics and no human-written experimental code. These deployments demonstrate scalability while exposing new classes of failure. We organize these failures into a taxonomy along severity, fixability, visibility, and capability locus. The taxonomy separates failures the loops can see and fix from those that require human judgment. Together, these results show that \textsc{Agon} is pushing research toward a new paradigm: machine scales, human steers.

Reliable structural health monitoring (SHM) of offshore wind turbine (OWT) support structures requires fast state estimation from sparse measurements. Repeated high fidelity finite element or aeroelastic analyses are difficult to use directly in online monitoring loops, while purely data-driven surrogates can require large training sets. This paper presents Digi Turbine, a synthetic reliability-aware Physics Informed Neural Network (PINN) benchmark for OWT monopile support structure monitoring. The workflow embeds a simplified Euler Bernoulli beam equation with Winkler soil foundation in the training objective, couples it with Bayesian-prior-informed inverse identification, and adds First Order Reliability Method (FORM) screening. All validation uses synthetic configurations with analytical or finite-difference ground truth motivated by the NREL 5MW reference turbine context.

On-device fault detection enables real-time diagnostics without cloud dependency, but deploying machine learning models on resource-constrained hardware demands careful tradeoffs between accuracy, latency, and model size. We present a benchmark comparing traditional ML methods (Random Forest, XGBoost, SVM, Logistic Regression) against lightweight transformer architectures (DistilBERT, TinyBERT-6L, TinyBERT-4L, MobileBERT) for binary fault detection across three public datasets: NASA C-MAPSS turbofan degradation, SECOM semiconductor manufacturing, and UCI AI4I 2020 predictive maintenance. We evaluate classification performance (F1-score, AUC), model size, and CPU inference latency, and further assess INT8 dynamic quantization and a two-stage adaptive inference pipeline. Our results reveal that on well-separated sensor data (C-MAPSS), lightweight transformers match traditional ML at 87.8% F1 but at 100x the model size and 9000x the latency. TinyBERT-4L emerges as the most deployment-friendly transformer at 55 MB and 18 ms CPU latency. INT8 quantization reduces size by 25% while preserving 86.9% F1. Our adaptive pipeline, routing 97.9% of predictions through a quantized triage model and only 2.1% to a larger expert, achieves 87.6% F1 at 19.5 ms average latency. On severely imbalanced datasets (SECOM, UCI-PM), both traditional and transformer methods struggle significantly, highlighting fundamental limitations of current approaches for extreme class imbalance in fault detection. All code is publicly available.

More than a billion people communicate in Indic languages, yet the natural language processing infrastructure serving them remains fragmented and underdeveloped. The cause is structural: the field organizes its tools and benchmarks around individual languages or small subsets of genealogical language families, building separate analyzers, parsers, and datasets for each language and starting over for the next. This overlooks a deep regularity. Through more than two millennia of convergence around Sanskrit, Indic languages came to share a morphosyntactic architecture formalized in Pānini's grammar, the Astādhyāyī. This cuts across genealogical lines, uniting languages through a common framework. We argue that this Pāninian framework supplies a unifying computational architecture the field has lacked, and that benchmarks grounded explicitly in it would make Indic language systems more accurate, more data-efficient, and more transferable, effectively merging many apparently disparate and sparse Indic language resources into a single high-resource metalanguage bedrock. We propose a four-part benchmark suite to render this shared architecture explicit, measurable, and ready to be leveraged for practical applications. Moreover, we underscore the question it raises for interpretability research: whether neural models trained on these languages come to represent Pānini's categories on their own.

Adapting a streaming speech recognition model to a new language requires choosing between two plausible warm starts: a multilingual (ML) encoder or an English-only (EN) encoder. The common intuition is that the multilingual encoder should help most at low data, but it is unclear how long that advantage persists, whether tight streaming latency amplifies it, and whether it survives deployment quantization. We answer these questions with a controlled sweep of a 0.6 B-parameter cache-aware FastConformer transducer across eight European languages, up to five target-language data scales (100 h to 2500 h), three streaming tiers plus offline decoding, and up to four public test sets. The main result is that multilingual initialization is a data-limited advantage, not a latency-limited one. On FLEURS at 160 ms, the mean EN-ML word error rate (WER) gap falls from +4.21 percentage points (pp) at 100 h to +0.20 pp at 2500 h; a power-law fit summarizes this decay, with each doubling of target-language data roughly halving the remaining advantage. Across the three streaming tiers, the across-language mean EN-ML gap is approximately stable at each scale from 100 to 1000 h, and is near zero by 2500 h. Finally, 4-bit weight-only encoder quantization at the matched 560 ms streaming tier reduces the encoder footprint by about 3x, with an average FLEURS WER increase of about 0.5 pp. The resulting guideline is simple: use multilingual initialization in low-data regimes, treat the choice as effectively irrelevant at large data, and make latency and quantization decisions independently.

Recent end-to-end models for EEG-guided target speech extraction report impressive results, underscoring potential for neuro-steered hearing technologies. However, our analysis reveals that high within-trial performance can be driven by trial-specific EEG structure that acts as shortcuts for target selection, leading to poor generalization on unseen trials. To overcome this gap, we propose TRUST-TSE, a two-stage framework to mitigate shortcut learning. By introducing contrastive pretraining with attended-speaker negative sampling, we encourage the EEG encoder to capture fine-grained EEG--speech alignment while suppressing trial-identity cues. We also employ a confidence-weighted extraction objective based on EEG--source similarity to guide extraction using the learned representations. Experiments on KUL and DTU datasets show that TRUST-TSE outperforms end-to-end baselines under strict cross-trial protocols, addressing a key reliability bottleneck of existing approaches.

Reliable provenance for LLM outputs requires multi-bit watermarks that remain robust under editing while maintaining strict false-positive control. Existing ECC-based LLM watermarks rely largely on hard-decision decoding, discarding token-level reliability information. We propose CORE-BREW, a Constant-hit-Rate Embedding extension of block-wise BREW for robust multi-bit watermarking. CORE-BREW calibrates the watermark channel by targeting a fixed hit rate p-star, yielding closed-form per-token log-likelihood ratios (LLRs) for principled soft-decision decoding. It supports two detection modes: Strict-Safe, which preserves the bounded-distance designated-codeword acceptance region, and FPR-Calibrated, which uses likelihood-based scoring and lightweight list decoding to characterize the FPR-TPR trade-off. Experiments on open-source LLMs under token-level edits and paraphrasing demonstrate improved low-FPR discrimination and robustness over prior multi-bit watermarking baselines while maintaining comparable semantic quality.

Foundation models have been increasingly applied to behavioral science domains such as psychology, sociology, and economics. While these models show promise in individual tasks such as survey response prediction and human-subject experiment simulation, there remains no systematic understanding of how well they perform across diverse behavioral science tasks, contexts, and populations. We introduce BehaviorBench, a comprehensive benchmark that evaluates foundation models along four core capabilities: (1) behavior prediction and simulation, (2) strategic decision-making, (3) subject-trait inference, and (4) behavioral knowledge application. Crucially, BehaviorBench evaluates model outputs at both the individual and distributional levels, capturing not only per-subject accuracy but also population-level alignment, an essential requirement for behavioral validity. Leveraging the tasks in BehaviorBench, we further develop Be.FM-1.5, extending the Be.FM family of behavioral foundation models fine-tuned on behavioral data. Our results reveal a considerable gap: proprietary general-purpose models excel at individual-level prediction and knowledge-intensive tasks, whereas behavioral foundation models, fine-tuned on behavioral data, achieve substantially stronger distributional alignment. Notably, Be.FM-1.5 leads on distributional metrics and remains competitive on individual-level metrics, suggesting that proper behavioral adaptation can close the gap. Our results highlight the importance of distributional evaluation, establish BehaviorBench as a foundation for developing and assessing behaviorally aligned AI systems, and demonstrate Be.FM-1.5's potential for a broad range of behavioral science studies. Our BehaviorBench and Be.FM-1.5 models can be accessed via https://umich-foreseer.github.io/behaviorbench/.

Causal inference provides a set of principles and tools that allow one to combine data and knowledge about an environment to reason with questions of counterfactual nature, i.e., what would have happened had reality been different, even when no data of this unrealized reality is currently available. Reinforcement learning provides methods to learn a policy that optimizes a specific measure (e.g., reward, regret) when the agent is deployed in an environment and pursues an exploratory, trial-and-error approach. These two disciplines have evolved independently and with virtually no interaction between them. We note that they operate over different aspects of the same building block, counterfactual relations, which makes them umbilically connected. Based on these observations, novel learning opportunities arise when this connection is explicitly acknowledged and mathematized. To realize this potential, we note that any environment where the RL agent is deployed can be decomposed as a collection of autonomous mechanisms with different causal invariances, parsimoniously modeled as a structural causal model; any standard RL setting implicitly encodes such a model. This formalization allows us to put under a unifying treatment different modes of learning, including online, off-policy, and causal calculus learning, which appear unrelated in the literature. However, these modalities are not exhaustive: we introduce several natural and pervasive classes of learning settings that entail novel dimensions of analysis. Specifically, we introduce and discuss through causal lenses generalized policy learning, where to intervene, imitation learning, and counterfactual learning. These tasks lead to a broader view of counterfactual learning and suggest great potential for studying causal inference and reinforcement learning side by side, which we call causal reinforcement learning (CRL).

The space $\mathcal{P}_2(\mathbb{R}^d$) of probability measures with finite second moment carries a natural geometry: the quadratic Wasserstein distance W_2 makes it a complete metric space and, following Otto, a (formal) Riemannian manifold whose geodesics are the optimal-transport interpolations. On this manifold, the gradient flow of the free energy F(rho) = KL(rho || π) is exactly the Fokker-Planck equation, and its implicit-Euler discretization is the JKO scheme. This is the geometry underlying diffusion models: the forward process descends the free energy, and each denoising step realizes one JKO step, which recovers DDPM, DDIM, NCSN/SMLD, and Energy Matching; this is one scheme, not separate theories. The same manifold supports a second variational principle. Its geodesics - the minimum-action curves of the Benamou-Brenier formula - are precisely the optimal-transport paths that Flow Matching learns. Fixing both endpoints and following the geodesic, generation becomes a deterministic ODE along a straight line, hence far fewer sampling steps. Placing both families of models on one manifold makes their relationship exact: diffusion follows a free-energy gradient flow, an initial-value problem; optimal-transport Flow Matching follows a Wasserstein geodesic, a boundary-value problem. The two reach the same endpoints along different paths.

Existing medical AI benchmarks lack process visibility, atomic skill evaluation, and integrated hallucination detection. We introduce MedBench v5, a redesigned benchmark for clinical multimodal models (language, vision-language, and agent systems) that moves from static QA to dynamic, process-oriented evaluation. MedBench v5 features: (1) a dual-dimensional framework combining Clinical Cognitive Responsiveness (14 sub-dimensions) and Medical Atomic Skills (4 agent environments), covering 63 tasks; (2) three switchable information-flow stressors (omission, contradiction, evidence delay) for factorized degradation analysis; (3) a dynamic process audit protocol with five reasoning nodes that produces model-specific failure fingerprints; (4) hallucination propagation monitoring across initiation, propagation, anchoring, and contradiction interaction-capturing silent hallucination. Experiments on frontier models show that strong overall task performance does not guarantee process stability: stressors mainly disrupt contradiction detection, diagnosis updating, hallucination propagation, and contradiction-based self-correction, while final evidence grounding can remain superficially stable. MedBench v5 provides a unified infrastructure for capability profiling, controllable stress testing, process auditing, and hallucination trajectory analysis in clinical AI evaluation.

Existing literature claims that video generation essentially is world modelling. On the one hand, the claim is productive because it pushes generative AI beyond static images and toward temporally extended physical scenes. On the other hand, this claim dangerously relies on the belief that scaling visual prediction alone will automatically yield physical agents. We prefer a more accurate statement: video generation models learn a partial, implicit spatiotemporal world model, but not a fully grounded or controllable one. The reason is as follows: a model may generate a plausible video of a drone crossing a forest or a robot arm manipulating a cup, yet still fail to know which variables are controllable, which constraints belong to a particular body and which futures remain valid under intervention. The frontier in essence is not predictive realism alone, instead it emphasizes a self-evolving generative nature that requires the decisive criterion to be counterfactual controllability: the capability of asking what would happen under an action, to test whether the generated future can survive embodiment constraints and to feed the resulting action knowledge back into future imagination (generation). Therefore, in this paper we present a new perspective, i.e., autonomous video generation with counterfactual controllability is one promising way to realize self-evolving world models.

Self-evolving agents improve over time by distilling experience from past executions and reusing it in future tasks. Existing systems represent such experience either as natural-language text injected into the agent context or as code exposed as callable tools. However, the choice between these representations is typically made at design time rather than derived from the characteristics of the experience itself, leaving the trade-offs between them poorly understood. We present the first controlled study that isolates text memory and code memory over an identical set of experiences. Our results show that the two forms exhibit complementary trade-offs in construction cost, execution efficiency, and transferability, such that neither representation alone is sufficient. Guided by these findings, we propose Metis, a self-evolving agent system built on a hierarchical dual-representation memory. Metis organizes textual experience into execution plans, environment facts, and common pitfalls, and selectively crystallizes recurring plans into validated callable tools. This design combines the broad applicability of text memory with the execution efficiency of code memory while incurring tool-generation cost only when justified by repeated reuse. We evaluate Metis on AppWorld, a challenging benchmark for interactive agents. The results show that Metis improves task accuracy by up to 20.6% over ReAct while reducing execution cost by up to 22.8%. Compared with representative self-evolving agent systems, Metis consistently achieves a better balance between accuracy, execution efficiency, and memory-construction cost.

Aligner-Encoders are recently proposed seq2seq end-to-end ASR models that replace decoder attention by predicting the uth token directly from the u-th encoder position, so the encoder must learn the alignment internally without cross-attention or a transducer lattice. In practice, this alignment often forms abruptly in the upper layers, making training sensitive and brittle on long utterances. We propose InterAligner, which adds an intermediate Aligner objective so alignment can form progressively across depth, together with an intermediate CTC loss (InterCTC) to stabilize optimization. On LibriSpeech with a 17-layer Conformer, a final-only Aligner reaches 5.0/7.8 WER (test-clean/other). InterCTC improves to 3.4/6.0, and InterAligner further reduces WER to 3.1/5.6 with the largest gains on long utterances.

Large language models (LLMs) can produce clinically fluent recommendations for type 2 diabetes while failing to satisfy guideline constraints or explicitly justify lifestyle-related glycemic claims. We present T2D-Bench, a reproducible benchmark and evidence-gated evaluation framework for testing whether LLM outputs satisfy explicit, graph-checkable evidence requirements. T2D-Bench is built on a multi-layer clinical-lifestyle knowledge graph that combines a biomedical spine (UMLS, DrugBank, SIDER), computable ADA Standards of Care rules, and lifestyle knowledge connected through a mechanistic bridge to glycemic laboratory effects. Across 100 structured vignettes spanning diagnosis, medication safety, and adversarial lifestyle conflicts, baseline outputs failed benchmark-defined evidence-path checks in 35% of cases for GPT-4o-mini and 33% for GPT-4o. The evidence gate detects unsupported omissions and uses constrained revision to bring outputs into verifier-level compliance with benchmark-defined evidence requirements. These results show that computable evidence constraints can make unsupported clinical omissions explicit, measurable, and correctable in diabetes-focused LLM outputs.

On-policy distillation (OPD) trains a student on its own rollouts guided by teacher feedback and is becoming increasingly important for large language model (LLM) post-training. Like reinforcement learning (RL), however, OPD faces an on-policy systems bottleneck, as rollouts can dominate training time for reasoning workloads. Asynchronous training pipelines can alleviate this bottleneck by decoupling rollout generation from learner updates, but doing so introduces stale-policy data. While prior work has studied stale data in asynchronous RL, its effects in OPD remain underexplored. We present the first systematic study of staleness in asynchronous OPD, focusing on a practical setting where teacher feedback is implemented through local KL losses and full-vocabulary teacher logits are too expensive to store or transfer, necessitating finite teacher-score caches. We first show that KL direction changes the stale-data problem: teacher-weighted forward KL is more robust to stale rollouts, whereas student-weighted reverse KL is vulnerable. Second, for this vulnerable reverse-KL case, we study whether methods designed to stabilize asynchronous RL can mitigate OPD staleness. In our experiments, they do not improve over a simpler OPD-specific surrogate: recomputing the reverse-KL signal under the current student at learner time. Third, we analyze how finite teacher-score caches create a bias-variance tradeoff for sparse and sampled reverse-KL OPD estimators. This motivates multi-sample Monte Carlo (MC), which preserves MC correctability while reducing one-sample variance. Finally, we present and open-source AsyncOPD, a fully asynchronous OPD training pipeline built from these estimator choices. Experiments show that AsyncOPD improves training throughput by $1.6\times$ to $3.8\times$ over strict synchronous training while reaching comparable accuracy.

Discrete flow matching (DFM) provides a principled framework for generative modeling on discrete state spaces via continuous-time Markov chain dynamics. In practice, sampling for DFM commonly employs discretizations such as $τ$-leaping, yet efficient sampling methods under a limited number of function evaluations (NFE) remain less studied. To address this gap, we propose the Time-Reparameterized Cumulative Intensity Extrapolation (TR-CIE) sampler, which aims to improve sampling quality when function evaluations are restricted. TR-CIE consists of two components. First, a schedule-based time reparameterization rescales the time grid according to the noise schedule. Under standard factorized DFM rate parameterizations, this transformation of variables absorbs the schedule-dependent growth term and mitigates stiffness near the terminal sampling stage. Second, we introduce a cumulative-intensity extrapolation updating rule. By reusing cached model outputs from the previous step as a history term, this improves the approximation of stepwise cumulative intensities on the resulting non-uniform time grid. We provide a theoretical analysis that bounds the local approximation error of cumulative intensities and establishes convergence results. The resulting sampler requires one NFE per step and introduces no additional model evaluations compared to the standard $τ$-leaping sampler. Extensive experiments on synthetic tasks, text generation, and text-to-image benchmarks demonstrate that our method improves sampling quality under limited NFE.

Motivated by an application in machine learning optimization, this paper focuses on the challenges of sampling a matrix uniformly from the unit spectral norm ball. It is proven that all singular values of sampled matrices converge to 1 almost surely as the matrix dimensions increase. This result provides the theoretical justification for a proposed simple sampling method applicable for large dimension sizes matching matrices found in modern large language models. Experimental results demonstrate both the convergence of the singular values, as well as the exact and proposed approximate sampling methods.

The composition of training data, governed by the diversity of sources and their mixing strategy, is a cornerstone of Large Language Model (LLM) pre-training. Online Data Mixing (ODM), the technique of adaptively adjusting data mixtures during training, has emerged as a promising direction to improve efficiency. However, existing methods are constrained by their reliance on a singular optimization perspective, which fundamentally overlooks the need for complex LLM pre-training to consider the dynamic data composition from multiple dimensions. To overcome this limitation, we introduce the Holistic Data Scheduler (HDS), a novel online data mixing framework. HDS formulates the data scheduling challenge as a reinforcement learning problem in a continuous control space and leverages the Soft Actor-Critic (SAC) algorithm for its stability and sample efficiency in exploring the high-dimensional policy space. At the core of HDS lies a novel multi-objective, holistic reward function that integrates three critical perspectives: a data-driven reward for quality, a loss-driven reward capturing inter-domain influence, and a model-driven reward based on weight norms. To validate our design and determine its optimal configuration, we conducted systematic experiments on LLMs of various sizes. On The Pile benchmark, HDS reaches the final validation perplexity of the next best method with 44% fewer training iterations. Furthermore, it achieves a 7.2% improvement on the MMLU 0-shot task along with consistent gains on other benchmarks, showcasing its ability to enhance both training efficiency and final model capability.

For a wheelchair user, a standard blue line on a map is often a broken promise. While platforms like OpenStreetMap (OSM) successfully capture where a path is, they frequently fail to convey how it physically feels to travel on it. This information barrier is problematic for wheelchair users. To solve this issue, we present OmniPath, a system that moves from passive mapping to proactive environmental auditing. Our framework fuses the network topology of OSM with the submeter precision of high-density aerial LiDAR (USGS 3DEP) to create a high-fidelity 3D model of the pedestrian environment. Rather than simply routing a user, our agent virtually traverses the network, analyzing the surface in 0.5 meter increments. It rigorously quantifies physical friction points specifically running slope, cross slope, and vertical discontinuities against ADA compliance standards, calculating a weighted severity score to categorize hazards from ``Mild'' to ``Critical.'' To ensure real world reliability, we validated the system against 200 physical ground truth field surveys across the National Mall using stratified random sampling. The framework demonstrated strong diagnostic reliability for high-severity hazards, achieving F1-scores of 0.60 for Severe and 0.58 for critical categories. By automating this micro-scale inspection, OmniPath identifies the ``invisible'' barriers that standard maps miss, effectively transforming a static dataset into accessibility data source that anticipates accessibility challenges before the user ever leaves home.

Music source restoration (MSR) requires jointly addressing source unmixing and the inversion of non-linear production effects. Current methods struggle to achieve accurate target signal reconstruction while maintaining semantic consistency. To address this limitation, we propose DTT-BSR+, a two-stage cascade MSR system that decouples distribution fitting from signal reconstruction into separate stages. A generative DTT-BSR separator in the first stage produces stems matching the prior of clean sources, and a modified Demucs network in the second stage enhances the first stage output using time-domain and multi-resolution spectral losses. DTT-BSR+ improves multi-mel signal-to-noise ratio (MMSNR) over the single-stage DTT-BSR across all stems, and surpasses the state-of-the-art X-LANCE MSR system on five stems. We also reveal through Fréchet Audio Distance (FAD) decomposition an implicit trade-off between signal reconstruction accuracy and semantic distribution fitting across stems.

Multi-step reasoning with Chain-of-Thought (CoT) prompting remains fragile: logical errors or hallucinations in early steps silently propagate, producing confident but incorrect conclusions. This paper presents VeryTrace, a zero-shot verification-and-repair framework that formalizes natural-language reasoning traces into a structured, compilable representation. VeryTrace introduces a Domain-Specific Language (DSL) that (i) makes step dependencies explicit, (ii) mechanizes quantitative content as executable expressions, and (iii) structures semantic inferences via deduction schemas. Our hybrid verifier combines deterministic checks for computational correctness, dependency resolution, and constraint satisfaction with targeted LLM audits for non-mechanizable semantic judgments, enabling step-level error localization and repair. Across three diverse domains-competition mathematics (AIME 2025), robotics planning (LLM-BabyBench), and kinship reasoning (CLUTRR), VeryTrace improves accuracy over zero-shot baselines on state-of-the-art LLMs without requiring domain-specific training or in-context examples, demonstrating that formalized trace verification achieves both precision and generalization.

Discrete diffusion language model (DLM) fine-tuning inherits inexpensive diagnostics from denoising-time confidence monitors, but their PEFT-training meaning is untested. We test top-1 argmax concentration as a collapse warning. Across 816 LoRA/PEFT configurations from three DLM families, the warning fires for every configuration while logs record 0/816 actual collapses at the 200 step horizon, giving zero precision. The cause is pre-equilibrium saturation: top-1 concentration is already high before optimization and quickly becomes insensitive to final training stability. We then evaluate max LoRA gradient norm, a parameter-side signal that samples gradient routing rather than token concentration. On a pooled held-out LLaDA-family split, a train-optimized threshold identifies top-decile final-loss configurations with precision 0.68 and F1=0.79, above the all-positive top-1 baseline even at the lower split-bootstrap confidence bound. Autoregressive controls and cross-family threshold failures bound the result to short-horizon DLM-LoRA inspection rather than a universal collapse detector. Workflow: drop top-1 as a PEFT alarm, log max-gradient early in training, and calibrate thresholds per DLM family before routing runs for inspection.

Vision-language models (VLMs) are prone to hallucination, which remains a major barrier to their safe deployment in clinical practice. To date, most hallucination detection methods have been evaluated on radiology benchmarks such as MIMIC-CXR and VQA-RAD, while gastrointestinal (GI) endoscopy remains largely underexplored. In this paper, we benchmark nine hallucination detection methods on the Gut-VLM dataset, a GI diagnostic Visual Question Answering (VQA) dataset with 4,392 test VQA pairs, across five VLMs (MedGemma-4B, MedGemma-27B, LLaVA-Med-7B, LLaVA-v1.6-7B, and Lingshu-32B). The methods span three categories: black-box methods (RadFlag, SelfCheckGPT-NLI), gray-box methods (AvgProb, AvgEnt, MaxProb, MaxEnt, Semantic Entropy, and VASE), and a white-box method (ReXTrust). Our results show that ReXTrust, a white-box method, achieves the highest AUC across all five models, outperforming the strongest alternative method on each VLM by a statistically significant margin (paired permutation test, p < 0.001 in all cases), reaching a peak AUC of 93.0 on MedGemma-4B. White-box hidden-state access provides a consistent advantage of 19.5 AUC points on average (range: 9.5--33.5), with ReXTrust maintaining strong performance even on LLaVA-v1.6-7B (AUC 79.9), where black-box methods and clustering-based gray-box methods collapse to near-chance performance. Among non-white-box methods, token-level gray-box statistics (MaxEnt, MaxProb) are the strongest alternatives, outperforming both clustering-based gray-box methods (Semantic Entropy, VASE) and black-box approaches on average. We further identify confident confabulation, a failure mode in which models hallucinate with high inter-sample consistency or high token-level probability, as a systemic failure for both consistency and uncertainty-based methods.

Federated unlearning (FU) is critical for complying with legal mandates like the right to be forgotten in decentralized systems, yet current methods face a persistent dilemma between non-target knowledge loss and high request latency. To resolve these issues, we propose FedUP, a one-shot federated unlearning framework utilizing lightweight pluggable filters that act as a "knowledge funnel" to screen out target data while preserving original model performance. By freezing original model parameters and training filters at the server side using differentially private (DP)-protected class centroid samples, FedUP bypasses the need for multi-round client-server communication and complex retraining, reducing unlearning latency from minutes to mere seconds. Additionally, the framework's pluggable architecture ensures inherent reversibility, enabling the seamless restoration of forgotten knowledge by simply removing the filters. Extensive experiments on diverse image and text tasks demonstrate that FedUP effectively reduces non-target knowledge loss and achieves superior unlearning precision and efficiency across various scenarios. Code is available at: https://github.com/suows/FedUP-code.

Multimodal misinformation detection is increasingly important because viral posts now combine long multilingual narratives, several images, mixed provenance, and subtle text--image framing errors. Existing benchmarks and methods remain poorly matched to this setting: they usually isolate short captions, single images, binary labels, or one manipulation source, while agentic verification remains costly under realistic evidence search. We present ReMMD, a realistic multilingual multi-image agentic verification framework for multimodal misinformation detection. ReMMD includes ReMMDBench, a real-world multimodal misinformation detection benchmark with 500 samples, 2,756 images, five monolingual languages, two cross-lingual settings, three text-length tiers, multi-image posts, five-way veracity labels, eight distortion labels, evidence provenance, and rationales. It also includes ReMMD-Agent, a persistent-memory verifier that decomposes posts into atomic points, builds a reusable evidence set, and predicts structured L1/L2/L3 outputs. Across proprietary systems, open LVLMs, MMD-Agent, and T2-Agent, ReMMD-Agent obtains the best five-way veracity performance, with 41.80% accuracy and 39.12% macro-F1 using GPT-5.2, while reducing cost by 17.5% relative to MMD-Agent and 79.9% relative to T2-Agent. The project is available at https://dang-ai.github.io/ReMMD.

Short dramas, with their rapid shot rhythms, dialogue-driven focus shifts, and demanding cinematographic grounding, pose challenges that prompt-level or text-only video generation pipelines struggle to meet. We study plot-to-short-drama generation, where a global plot and local context are transformed into visually grounded multi-shot videos. We propose DramaDirector, a geometry-grounded framework that lets the planner borrow cinematographic geometry from a gallery of real short-drama shots indexed by depth and pose. DramaDirector decouples each shot into static visual and dynamic narrative conditions, trains the planner with schema-constrained SFT and GRPO under a learned text-visual alignment reward, and retrieves depth-pose references to guide first-frame generation and image-to-video synthesis. We also introduce DramaBoard, a benchmark built from 35 live-action dramas, 2.8K episodes, and 81K shots, with structured storyboards and multi-dimensional evaluation protocols. Experiments show that DramaDirector improves over representative multi-agent and video generation baselines on faithfulness, consistency, and controllability. Our code is released at: https://github.com/iLearn-Lab/DramaDirector

Generative artificial intelligence(GenAI) is reshaping learning in higher education, with particularly pronounced implications for the humanities and social sciences(HSS), where learning outcomes are commonly expressed through written and interpretive forms that align closely with GenAI's capabilities. Yet, systematic evidence on the educational impacts of GenAI on HSS students remains limited. Addressing this gap, this study draws on a large-scale survey of HSS students in China to examine its role in academic development. Guided by relevant learning theories, this study focuses on four dimensions: patterns of use, effects on learning processes and academic performance, challenges associated with GenAI use, and preferred approaches to curricular integration. We found that more than half perceived enhanced learning motivation, independent thinking and creativity, although a substantial minority reported little change or even decline. Comparatively, a notably larger majority reported academic performance gains, although these gains may partly reflect limitations in conventional assessment practices. The study identifies variations in perceived learning and performance improvements among students with differing durations of GenAI experience, along with observable disciplinary differences and modest gender differences. While an overwhelming majority valued the importance of ethical considerations, only slightly more than half were satisfied with privacy protection. Limited accuracy and overreliance emerged as the most pressing concerns reported by students. Students favored partial or optional curricular integration supported by practice-oriented training, and widely recognized GenAI's significance for their future professional development. Grounded in student perspectives, this study offers evidence-based recommendations for the responsible and pedagogically meaningful integration of GenAI

Most electronic health record (EHR) foundation models encode clinical events as discrete event tokens from a fixed vocabulary and therefore cannot directly represent events containing unseen concepts or new combinations of concepts and attributes such as numeric values. This limits transfer across institutions and even across deployment pipelines within the same institution. We introduce PORTER, a language-grounded structured EHR foundation model that decouples event representation from this fixed vocabulary. PORTER represents events through their descriptions using a frozen text encoder, integrates numeric values through a dedicated pathway, and learns clinical dynamics over patient timelines with an autoregressively pretrained temporal backbone. Across 74 clinical prediction tasks at a pediatric hospital, PORTER matched the mean AUROC of a fixed-vocabulary model with the same temporal backbone and pretraining objective. When the same patient timelines were rendered using event descriptions not seen during pretraining, PORTER transferred without retraining or vocabulary mapping, recovering 97.1% of the mean AUROC of a model trained directly on the target vocabulary. When transferred to MIMIC, PORTER outperformed the fixed-vocabulary model, which dropped 69% of events because their tokens were unseen. Mechanistic analyses showed cross-vocabulary transfer tracked preservation of patient-level representation geometry rather than the scale of the text encoder, and the numeric pathway improved sensitivity to magnitude without disrupting clinical concept identity. PORTER also achieved higher AUROC than a task-specific text serialization comparator, at 329-fold lower amortized compute. PORTER is a step toward vocabulary-independent EHR foundation models that reduce the need for vocabulary harmonization while preserving in-domain performance and enabling efficient cross-task reuse.

Algorithms have become central to scientific research in the era of artificial intelligence (AI). Although algorithm mentions in papers are often used to indicate popularity and influence, existing studies usually evaluate individual algorithms in isolation and pay limited attention to the collective influence formed through their interconnections. This study constructs large-scale algorithm co-occurrence networks in natural language processing (NLP) based on the full text of academic papers and investigates algorithm influence from a network perspective. Using deep learning models, we extract algorithm entities and build overall, cumulative, and annual co-occurrence networks. We analyze their structural characteristics and apply multiple centrality measures to assess the group influence of algorithms across the whole field and over time. The results show that algorithm networks display typical features of complex networks, with increasingly dense connections developing over approximately two decades. Classic, high-performing algorithms and those located at the intersections of different research periods tend to have high popularity, control, centrality, and balanced influence. When the influence of an algorithm declines, it usually loses its core network position first, followed by weaker associations with other algorithms. This study is the first large-scale analysis of algorithm co-occurrence networks. Covering more than four decades of academic publications, it provides a temporal and structural view of algorithm influence and offers a foundation for future research on networks linking algorithms, scholars, and tasks.

Robust perception underpins autonomous driving, and most recent progress comes from scaling the model-larger backbones, foundation models, and cooperative multi-agent fusion. We pursue a complementary, upstream question: what should the camera itself measure? Using a differentiable RAW-to-task pipeline, we decompose which sensor degrees of freedom benefit dense prediction. Learning the spectral colour-filter-array (CFA) weights is the dominant lever, improving mIoU by +0.017 (KITTI-360) and +0.023 (ACDC) over a fixed camera. In contrast, point-spread-function (optics) co-design is net-negative (-0.020 mIoU on KITTI-360) - a consequence of the data-processing inequality, which also bounds the task information that any downstream model, however large or cooperative, can recover. Noise co-optimisation is marginal, and counter to intuition enlarging the CFA tile beyond 2x2 consistently hurts, as the filters are confined to the rank three sRGB input. Because the intervention is at the sensor, the gains are model-agnostic; we validate robustness on ACDC's fog, night, rain, and snow, and conclude with a simple recipe: learn the 2x2 CFA weights and keep an identity PSF.

We ask whether the geographic origin of Tang-dynasty poets leaves a detectable linguistic trace in their work. Aggregating every poem attributed to each author in the Complete Tang Poems (Quan Tang Shi) and linking poets to their administrative circuit of origin via the China Biographical Database (CBDB), we build a poet-level corpus of 357 poets across the ten Tang circuits and frame origin prediction as multi-class classification. Using character $n$-gram TF-IDF together with interpretable domain features (imagery, season, and allusion), classical and neural models predict a poet's broad region (South vs.\ North) at $0.69$ accuracy, well above the $0.53$ majority baseline, and finer circuit-level origin above chance. Beyond classification, three findings emerge. (i) Linguistic distance between circuits grows with geographic distance (Mantel $r=0.40$, $p\approx0.09$ over nine circuits), evidence of a distance-decay effect in poetic language. (ii) The signal interacts with time: South/North separability is at chance in the High Tang and strongest in the Late Tang, consistent with court-driven homogenization at the empire's height followed by regional divergence. (iii) The model's confident errors are historically meaningful -- in the Early Tang, every misclassification is a southern poet read as northern, reflecting the prestige of the northern court idiom. We further show that, when given the whole corpus through a hierarchical frozen-encoder representation, a classical-Chinese transformer (GuwenBERT) only matches -- not beats -- simple TF-IDF, and that combining them adds nothing, indicating that character $n$-grams already capture the regional signal. Our results position interpretable machine learning as a hypothesis generator for literary history.

Recent advances in control have enabled bipedal-wheeled robots to traverse slopes and single-step obstacles, yet long staircase traversal remains challenging as current teacher-student frameworks suffer from weakened dynamics-aware representations and incomplete terrain geometry encoding. To bridge this gap, we propose DynaWM, a dynamics-aware representation learning framework. To enhance terrain encoding capability and enable transparent assessment, we introduce a world model as a regularizer to enforce forward-dynamics awareness, preserving comprehensive terrain geometry while facilitating hierarchical encoding visualization. To stabilize knowledge transfer, we employ a momentum target encoder to provide consistent distillation targets, preventing dimensional collapse from non-stationary teacher updates. Evaluation of the learned representations through Principal Component Analysis (PCA) visualization and quantitative metrics reveals that our encoder hierarchically captures terrain geometry with higher terrain encoding capability, leading to enhanced terrain adaptability and motion smoothness. Experimental results in simulation and real hardware demonstrate that our method achieves superior terrain adaptability and motion smoothness, enabling bipedal-wheeled robots to overcome diverse continuous stairs, as shown in Fig. 1.

Reconstructing continuous speech from scalp electroencephalography (EEG) remains fundamentally challenging. EEG provides a weak, spatially diffuse, and highly variable measurement of distributed cortical activity, whereas speech is organized as a coherent acoustic trajectory with strong harmonic and temporal structure. The resulting mismatch makes waveform regression unstable and causes stochastic multi-step generation to be sensitive to artifact-dependent conditioning and subject variability. We introduce NeuroSonic, a conditional flow-matching framework for EEG-to-speech reconstruction. Instead of predicting waveforms directly or refining them through stochastic denoising, NeuroSonic learns a deterministic probability-flow velocity field that transports a noise-corrupted acoustic state toward clean speech under EEG conditioning. EEG and audio are embedded into a shared token space and processed by a time-conditioned gated Transformer that parameterizes the transport ordinary differential equation. This formulation models trajectory evolution explicitly while avoiding iterative stochastic sampling. We evaluate NeuroSonic on the CineBrain and EAV benchmarks under cross-subject evaluation. Across both datasets, the proposed method improves distributional realism, spectral fidelity, and perceptual quality over representative GAN-, diffusion-, and mean-flow baselines, with up to a 26.3\% gain in overall perceptual quality. The performance gap is most evident in artifact-heavy segments, where conditioning variability is strongest. These findings indicate that deterministic conditional transport provides a stable and effective formulation for EEG-driven speech reconstruction. Code is available at https://github.com/Y-Research-SBU/NeuroSonic/ .

On-policy distillation (OPD) trains a student policy using teacher signals computed on trajectories sampled by the student itself. Recent work shows that sampled-token OPD can be fragile on long-horizon reasoning tasks and that local teacher-support matching is a simple and effective repair. This paper introduces blockwise policy-drift gating, a lightweight student-only old-current drift controller for OPD under rollout reuse. The method computes log-probability shifts between the behavior student and the current student on the sampled token path, aggregates these shifts over fixed blocks or spans, and uses the resulting detached, mean-normalized gates to reweight OPD position losses. It does not change teacher targets, teacher top-K supports, or the rollout policy. In a six-variant Qwen3 math reasoning benchmark with a uniform 200-step training budget for all trained variants, we use pass@8 as the primary problem-level solve-rate metric. Fixed 64-token block gating improves sampled-token OPD mean pass@8 from 0.4978 to 0.5160 across AIME24, AIME25, MATH500, and AMC23. On Teacher-TopK/LSM, Block64 gives the best four-benchmark mean pass@8 among trained students. The results identify local old-current policy drift as a practical control signal for reused OPD rollouts and motivate block-level gating as a simple default for improving solve-rate robustness.

"Talk short. Drop grammar. Save token." This caveman style is widely promoted as a way to cut inference cost, but whether it actually saves anything depends on which channel (the user's prompt or the model's response) is being compressed. We present Cavewoman, a two-channel evaluation protocol that scores every generation on task accuracy, realized per-item cost, and reference-text agreement against the model's unconstrained reference. We evaluate eight models on five datasets at five reduction levels, with both channels measured on the same items. Output compression cuts realized cost on most API models (1.4-2.4x per model, up to 3x in the best case) and on all four open-weight models under public-tier pricing. Input compression has the opposite effect, a strict lose-lose: it raises net cost rather than lowering it (~1.15x on the five-benchmark mean, up to 1.8x on the worst dataset and 2.7x under stronger compression), because models compensate with longer responses even as accuracy collapses. Under the same setting, surface text diverges from the unconstrained reference: on the non-reasoning models, roughly half of all generations are correct yet their surface text no longer entails the model's own unconstrained baseline generation. The divergence survives length-controlled re-scoring, multiple-comparisons correction, and replication under complementary semantic measures. Code and data are available at https://github.com/danielle34/cavewoman.

As Text-to-Image (T2I) jailbreak techniques evolve rapidly, existing benchmarks and reproduction workflows often struggle to keep pace. More importantly, T2I jailbreak evaluation is not a single prompt-level test, but a pipeline-level problem shaped by multiple stages, including prompt transformation, image generation, safety filtering, and multimodal judging. This makes results across papers difficult to reliably reproduce and fairly compare. To bridge this gap, we propose PixJail, a self-evolving paper-to-pipeline agent framework for reproducible T2I jailbreak evaluation. Given a T2I jailbreak paper and optional reference code, PixJail rapidly constructs a paper-specific attack module and a runnable evaluation pipeline under a unified contract, while faithfully reproducing the original experimental results. PixJail further maintains a memory bank that stores paper digests, attack evolution patterns, reusable templates, failure cases, and versioned artifacts, enabling future reproduction efforts to reuse prior experience. We reproduce eleven representative T2I jailbreak methods, including both code-available and code-unavailable papers. Under their original settings, our framework accurately recovers prior results with minimal error (2.1\% average, 0\% median). We hope that PixJail can serve as a unified foundation for future T2I jailbreak reproduction and evaluation, significantly reducing manual effort.

Contextual entrainment, which is a newly discovered phenomenon in large language models (LLMs), refers to the tendency of a model to assign higher probabilities to tokens that appear in its context. In this work, we extend this phenomenon from the token level to the sentence level by examining the per-token mean log-probability of a sentence instead of the probabilities of individual tokens. We investigate sentence-level contextual entrainment across 26 LLMs from seven families and two datasets, which cover both subjective and objective tasks. We find that sentence-level contextual entrainment exists. This means that the sentences in the prompt (even if they are counterfactual statements) can significantly increase their probability during model inference time. As the model size increases, contextual entrainment gradually decreases. We also find that contextual entrainment is controlled by 2% to 4% of the attention heads. Turning off these attention heads can effectively mitigate contextual entrainment without hurting the model's performance.

Although deep learning-based methods can achieve high accuracy in automatic modulation recognition (AMR) tasks, their high computational cost makes it difficult to strike a balance between accuracy and power consumption, thereby limiting their application on resource-constrained platforms. Neuromorphic architectures that perform spike-driven inference with modest energy budgets have recently been explored for vision and timeseries tasks. Motivated by these works, we propose EMRFormer, a novel end-to-end spiking nerural network (SNN) architecture that applies spike-driven transformer to the constraints of neuromorphic hardware for AMR. The model incorporates an adaptive spike encoder and Integer Leaky Integrate-and-Fire neurons to mitigate the degradation of effective information and enhance SNN representational capacity. By integrating spike-separable Convolution Neural Networks (SSCNN) into Spike-Driven Transformers (SpikeFormer), EMRFormer effectively extracts multi-scale temporal features from the raw IQ waveforms. We validate our approach across various mainstream datasets, the experimental results show that EMRFormer achieves state-of-the-art interms of accuracy, outperforming all the baselines. Furthermore, the model maintains strong performance in low signal-to-noise(SNR) environments and reduces theoretical energy consumption by over 90%. Finally, we evaluate our model on a KA200 neuromorphic chip. The results show that our model achieves up to 5 times reduction in power compared to running on a 3090 GPU or an Orin NX. This work demonstrates a promising pathway for AMR on resource-constrained devices.

Wang~\cite{Wang2026} introduced the Stochastic-Oracle Turing Machine (SOTM) framework and defined token complexity as the minimum expected cost of interacting with a stochastic oracle needed to attain a specified solution quality for a task. This paper develops an analogous notion for certifying the reliability of a stochastic oracle on a given domain. Certification token complexity is the minimum expected token cost required, with controlled error probability, to distinguish oracles that meet a target reliability level from those that fall below a lower reliability threshold. We construct an SPRT-based certification SOTM that queries the oracle, computes binary correctness scores, and stops when the accumulated log-likelihood evidence crosses a decision threshold. The SOTM halts almost surely, satisfies the desired two-sided error guarantee over the reliability regions to be certified, and yields an explicit upper bound on certification token complexity in terms of the reliability thresholds, the error bound, and the expected per-turn token cost. We then establish a matching information-theoretic lower bound: even with adaptive queries, every error-bounded certification SOTM must incur the same leading-order expected token cost as the SPRT-based construction as the prescribed error bound tends to zero. Together, these bounds characterize the leading-order certification token complexity in the small-error regime.

Speaker recognition has advanced rapidly with large-scale training datasets, yet Vietnamese remains under-resourced, with existing corpora limited in scale and acoustic diversity. Most large-scale datasets rely on facial cues to link speech with speaker identities, restricting data collection to recordings where speakers appear on camera. We propose a face-independent dataset construction pipeline and introduce VieSpeaker, a large-scale Vietnamese speaker recognition dataset. Our approach leverages textual metadata and large language model reasoning to infer speaker identities from transcripts and contextual information. VieSpeaker contains approximately 902 hours of speech from 4,715 speakers. Experiments show that models trained on VieSpeaker achieve improved robustness and generalization compared to existing Vietnamese datasets. This work demonstrates the feasibility of face-independent dataset construction and provides a new direction for building large-scale speech resources.

Distilling reasoning capabilities from strong to weak language models typically involves imitating specific solution trajectories, effectively transferring what to answer rather than how to reason. This trajectory-level imitation encourages memorization of instance-specific steps rather than acquisition of transferable problem-solving skills, limiting generalization to novel problems. We propose Strategy-Guided Policy Optimization (SGPO), which replaces instance-level trajectory imitation with reusable strategy distillation. SGPO extracts structured strategy descriptions from strong-model responses and, for each problem, constructs both autonomous and strategy-guided trajectories to enable direct comparison of the model's behavior with and without strategic guidance. The framework then addresses two key questions. For how to distill, a token-level forward-KL objective selectively transfers the distributional shift induced by strategy conditioning into the unguided policy, with proximal constraints ensuring stability. For when to distill, adaptive instance-level weighting strengthens guidance when autonomous exploration falls short and reduces it as the model's own competence grows. Experiments on four mathematical benchmarks across two model families show that SGPO consistently outperforms SFT, on-policy RL, and hybrid-policy baselines, improving the average score by 2.2 points over the strongest baseline on Qwen2.5-7B-Instruct. Analysis reveals that the forward-KL objective provides an inherently selective distillation signal that outperforms direct trajectory imitation, and that strategy distillation exhibits complementary scaling with base model capability.

Modern spoken language understanding (SLU) systems are increasingly deployed in real-world settings, where specific functionalities may need to be removed due to policy or safety constraints. In SLU, a functionality corresponds to an intent and its associated slot-generation behavior. However, in autoregressive models, suppressing a target intent does not eliminate the conditional mapping that generates slots conditioned on that intent. When the intent prefix is externally supplied, the model can reconstruct the original intent-slot structure. We identify this structural failure as \textbf{\emph{capability persistence}}. We propose \textit{\underline{B}inding \underline{S}ubspace (BSU)}, a representation-level framework that isolates and attenuates intent-conditioned directions underlying this mapping. Across SLU benchmarks, BSU substantially reduces forced-prefix recoverability while preserving retained performance.

Financial time series forecasting presents structural challenges absent from standard benchmarks. Log-returns are non-stationary, exhibit exceptionally low signal-to-noise (SNR) ratios, and are governed by regime-dependent temporal dependencies. We identify a key limitation of state-of-the-art (SOTA) time series models in financial settings. A fixed context window is mismatched to the time-varying optimal look-back of non-stationary price processes. We propose the Regime-Aware Variable-context Expert Network (RAVEN), a Mixture-of-Experts framework designed to adaptively determine the temporal context for each input sample. Instead of relying on a fixed look-back horizon, RAVEN constructs a hierarchy of nested contiguous windows whose lengths are determined by the data itself. Specifically, RAVEN scores patches by learned importance in reverse chronological order and applies the Cumulative Importance Thresholding (CIT) mechanism to derive nested prefix windows, each routed to a scale-specialized expert. A Global Compressed Representation (GCR) branch runs in parallel over the full context, preserving global temporal coherence that local experts cannot guarantee. Because the nested routing induces structured overlap among expert inputs, we introduce a Correlation-Aware Weighting (CAW) to align variable-length expert outputs and penalize pairwise cosine similarity prior to aggregation. Experiments on cumulative log-return prediction (HS300, S&P500) and fund sales forecasting demonstrate that RAVEN achieves SOTA performances, improves Pearson correlation by 9.2% on HS300 and 20.2% on S&P500, and reduces MSE by 18.2% on fund sales forecasting, while achieving the best results in 14 of 16 metrics on four PEMS traffic benchmarks.

Requirements elicitation requires stakeholders to communicate needs, negotiate priorities, and collaboratively construct knowledge that can be transformed into software requirements artifacts. Recent advances in LLMs have created opportunities to support these activities through AI-assisted collaboration and automated artifact generation. However, limited empirical evidence is available regarding how AI-supported collaborative environments influence requirements elicitation outcomes. In this study, we conducted a mixed-method controlled experiment comparing four requirements elicitation approaches: collaborative elicitation without AI support, collaborative elicitation supported by the Strateegia platform and its GPT-powered Writer applet, direct requirements generation using a Large Language Model, and requirements generation from collaborative discussion transcripts using a Large Language Model. We evaluated the resulting artifacts using quality criteria derived from ISO/IEC/IEEE 29148 and collected participant perceptions regarding the elicitation process. Our findings indicate that approaches combining stakeholder collaboration and AI-supported synthesis produced the highest-rated requirements artifacts and were perceived as clearer and easier to execute than traditional collaborative elicitation. The results suggest that generative AI can support the transformation of collaboratively generated knowledge into structured requirements documentation while preserving the value of stakeholder participation. We discuss implications for AI-supported requirements elicitation and human-AI collaboration in Requirements Engineering.

Sentiment analysis in Twitter datasets is important because it enables monitoring public opinion on products and analysis of political and social movements. One critical step is preprocessing: the automated processing of text for machine learning algorithms. Preprocessing plays a critical role in reducing noise and improving efficiency. However, little research has systematically examined the order in which preprocessing techniques are implemented. We find that, when accounting for order, spelling correction is the least impactful preprocessing technique, whereas tokenisation is the most impactful. Stemming and stop-word removal are interchangeable, and it is better to remove stop words without removing negation. The best order for applying the preprocessing techniques was tokenisation, text cleaning, stemming, and then stopword removal. Our results provide a systematic approach for practitioners to deploy preprocessing to improve model output without the costly preprocessing exploratory phase.

One of the significant mental health issues affecting female sex workers (FSWs) is mental disorders, especially depression. Exposure to violence, stigma, and economic hardship further increases their psychological risk. Current machine learning (ML) models are typically ineffective at capturing the high-dimensional and complex risk patterns that exist in this marginalized group. This paper suggests a hybrid predictive model that merges an ensemble feature selection strategy using ANOVA and mutual information and Harris Hawks optimization-tuned logistic regression and represents a new application of swarm intelligence to predict mental health in vulnerable groups. The explainable AI (XAI) methods can be used to understand the factors of trauma associated with model predictions. When applied to a group of 3,005 FSWs, it can be seen that the proposed model is more effective than traditional classifiers, with an accuracy of 95.78%, an F1 score of 95.77%, and an AUC of 0.96, and identifying post-traumatic stress, client-related violence, and occupational factors as major contributors to depression. This work bridges the gaps between conventional and ML approaches to develop an XAI tool that enables vulnerable groups to receive early assistance, evidence-based targeted psychosocial care, and health planning.

This work presents a machine learning framework that leverages an autoencoder (AE) for the efficient modeling of FinFET. We first calibrated a BSIM-CMG model to generate a dataset of current-voltage (ID-VG) characteristics. This data was used to train an autoencoder that compresses full I-V curves into a low-dimensional latent space, which intrinsically encodes key device physics. A key innovation is the explicit incorporation of parameter such as drain to source voltage (VDS) as an input feature, enhancing the model ability to capture bias dependent variation. The trained model successfully reconstructs full I-V curves and directly extracts critical device metrics including threshold voltage (VTH), subthreshold slope (SS), and peak transconductance (gm). This approach demonstrates that data driven compact models, built from actual characterization data, can achieve high accuracy with minimal training data, providing a powerful tool for rapid device characterization, modelling and circuit level simulation.

Recommender systems often induce filter bubbles and semantic homogenization by monolithically optimizing for immediate user engagement. Standard single-objective models, including traditional Deep Q-Networks, are ill-equipped to navigate the trade-offs between platform retention and critical societal values like information diversity and provider fairness. To address these limitations, we introduce a multi-objective reinforcement learning framework that formalizes recommendation as a semantic multi-objective Markov decision process. By integrating high-fidelity semantic embeddings with a Pareto-DQN agent, our architecture treats engagement, diversity, and fairness as distinct, non-aggregable reward signals, avoiding the pitfalls of static reward scalarization. Empirical evaluations on the MovieLens small dataset shows that our hypervolume based action selection disrupts the feedback loops responsible for semantic collapse. By sustaining high state-trajectory variance, the Pareto-DQN effectively maps the Pareto frontier, achieving gains in auxiliary societal objectives with only marginal impacts on engagement. This work provides a path toward intrinsically aligned, responsible recommender systems.

MeMo proposes language models with explicit multi-layer correlation matrix memories (CMMs), where memorization, retrieval, and forgetting are architectural operations. This paper asks how such memories can reduce the need for retraining when knowledge changes. For changes expressible as MeMo memory associations, the model's accessible knowledge can be updated by editing explicit memories rather than retraining the whole model. We propose a version-aware operation layer in which high-level operations such as replace, obsolete, keep-history, rollback, and trace are compiled into MeMo-native primitive calls over sequences and tokens. The key observation is that a version-aware operation is rarely a single MeMo association. It is an ordered transaction of primitive edits, for example forgetting one sequence-token chain, memorizing another, preserving a historical chain, and recording an inverse program. The framework introduces two auxiliary CMMs: a Version CMM (V-CMM) for mapping version transitions to transaction handles, and a Transaction CMM (T-CMM) for storing reusable change contents and inverse programs. It supports both direct sequence-level edits and structured diff-level inputs, and outlines an evaluation route for update success, rollback, traceability, locality, and transaction reuse.

Existing low-bit KV-cache quantizers often treat each cached key as a flat vector. Under RoPE, however, a key's contribution to a future attention logit decomposes into a position-dependent sum over two-dimensional frequency blocks. This makes key-cache quantization a block-wise bit-allocation problem: high-energy RoPE blocks are more sensitive to quantization error and should receive more bits. We introduce Block-GTQ, a RoPE-aware bit allocator for key-cache quantization built on TurboQuant-MSE(TQ-MSE). For each layer and KV head, Block-GTQ computes a label-free energy score for each RoPE block and greedily allocates integer bit widths by marginal gain. Under matched K/V bit budgets, Block-GTQ better preserves RoPE query-key logits on a ten-model diagnostic panel, cutting per-layer MAE by 32-80% at 2 and 3 b/dim K-only quantization and winning all 367/367 layer comparisons against uniform TQ-MSE. These fidelity gains translate to stronger downstream long-context retrieval, understanding, and reasoning. At K2V2 on Llama-3.1-8B-Instruct, Block-GTQ raises the six-task NIAH average from 70.6 to 97.4, and the LongBench-EN average from 36.87 to 53.31. On AIME 2024/2025 with DeepSeek-R1-Distill-Qwen-7B, without an fp16 recent-key buffer, Block-GTQ at K3V2 scores 51.7/37.5, close to fp16's 54.2/37.9, whereas uniform TQ-MSE collapses to 0.0/0.0. We further implement a packed-cache serving path. On a single H800 GPU with Qwen2.5-3B-Instruct, packed K3V3 achieves 3.24x KV-cache compression with fp16-comparable quality, runs 1.34x faster than fp16 FlashAttention2 at 128K context, reduces peak memory from 56.31 GB to 19.85 GB, and remains feasible at 256K and 512K where fp16 OOMs. Code is available at https://github.com/JIA-Lab-research/blockgtq.

Mechanistic interpretability has made substantial progress in automatically localizing circuits, but explaining what localized components do remains labor-intensive and difficult to standardize. In this work, we study whether language model (LM) agents can assist with this explanation problem once a circuit has already been identified. We introduce AgenticInterpBench, a benchmark for circuit explanation built from 84 semi-synthetic transformer circuits with 163 component-level annotations. We propose HyVE (Hypothesize, Validate, Explain), an agentic explainer that analyzes each component through an iterative loop of observation, hypothesis generation, and causal validation, eventually producing a component-level explanation and a circuit-level task description. Across four LM backbones, HyVE recovers useful component- and task-level explanations, but no backbone is uniformly best. Our analysis shows that strong backbones usually form observation-grounded hypotheses, while failures more often arise later in the validation loop, through incomplete validation plans, code execution errors, or unresolved hypotheses. A case study on an arithmetic circuit in Llama-3-8B shows that the same formulation can extend beyond semi-synthetic benchmarks to naturally trained models. Overall, LM agents are promising circuit explainers, but reliable validation remains the key obstacle.

Diffusion models have achieved strong performance in image, text-to-image, and video generation, where conditional generation is often controlled by classifier-free guidance (CFG). CFG improves condition consistency by increasing a guidance weight, but stronger guidance typically reduces diversity and distributional coverage. It remains unclear how this consistency-coverage trade-off should be controlled across the reverse trajectory, since the distribution induced by CFG is not simply the fixed-time tilted distribution given by the guided score field. To address this issue, we propose an information-theoretic framework for CFG schedule optimization. Our approach uses a clean endpoint reference to specify the desired consistency-coverage trade-off, while optimizing the actual distribution induced by the guided sampler toward this reference. We derive trajectory-level formulas to estimate the objective from samples and score evaluations, avoiding explicit density estimation. On ImageNet-512 with EDM-XXL and COCO with SD-XL, the learned schedules achieve competitive or improved trade-offs over constant guidance and allocate guidance selectively across noise levels.

A modern model release reports scores on 40+ benchmarks and the same evaluations were run many more times before it: to track training progress, compare design choices, and select the checkpoint for the release. But do we need to run every eval? We compile a public score matrix of 84 frontier models on 133 benchmarks (2,604 cells, 23.3% filled) and find it is approximately rank-2: a model's scores across all 133 benchmarks are largely determined by just two numbers. We confirm this in two ways: scores hidden from the matrix are best recovered using two factors, and two factors already explain over 90% of the variation among models on the benchmarks they share. Building on this, we design BenchPress: a logit-space rank-2 matrix completion method that recovers held-out scores to within 4.6 points, and a confidence layer that says when each prediction can be trusted. Using BenchPress, we find a subset of five benchmarks {GPQA-D, HLE, Codeforces, MMLU-Pro, ARC-AGI-1} that can recover the rest of a model's public scorecard to within 3.93 points. For a tighter inference budget, a cheaper set {GPQA-D, MMLU-Pro, Aider Polyglot, MATH-500, AIME 2026} can predict a model's evals to within 4.55. We release the score matrix, the BenchPress code, and an interactive tool that predicts any model's score on any benchmark.

As AI systems are deployed across increasingly diverse and high-stakes settings, model alignment must generalize beyond the tasks and domains seen during training. This is especially important for reinforcement learning (RL), which can introduce unexpected misalignment through reward hacking, deception, or other unintended strategies. We study whether RL on beneficial behavior, instantiated in realistic domains, can produce broad and persistent alignment generalization beyond the training distribution. We construct a dataset of realistic situations designed to measure and train beneficial traits, such as truthfulness, fairness, risk awareness, and corrigibility, spanning varied domains, including health, science, and education. We then train models with RL on this dataset and evaluate them on more than 50 independent benchmarks of alignment and beneficial behavior. Compared to a compute-matched baseline, beneficial trait RL improves performance on over 80% of these out-of-distribution benchmarks. We observe substantial out-of-distribution alignment transfer: a beneficial-behavior RL intervention entirely limited to one domain, health, produces broad improvements on non-health alignment evaluations, including reduced reward hacking, deception, and general misalignment. Finally, we study alignment persistence: whether behavior remains robustly aligned under attempts to steer models towards misalignment. Models trained with beneficial trait RL show improved persistence, including greater resistance to adversarial prompting and harmful finetuning; further work is required to isolate the sources of these effects. These results suggest that RL to reinforce beneficial behavior in realistic domains can produce models that are more robustly aligned with human flourishing.

A crucial assumption in graph signal processing (GSP) is the existence of an underlying graph that captures the pairwise similarities between nodes, allowing filters to be designed based on this graph for tasks such as denoising. For spatial-temporal data in which node-to-node similarities evolve over time, a static spatial graph is insufficient. In this paper, to represent slowly time-varying pairwise relationships, we model the graph changes in two consecutive adjacency matrices $P = W^{(2)} - W^{(1)}$ across time as a low-rank matrix. % Specifically, given an initial adjacency matrix $W^{(1)}$ at time $t=1$, we jointly interpolate a signal $x_2$ and estimate $W^{(2)}$ at $t=2$ using both a graph signal smoothness prior for $x_2$ and a low-rank prior on $¶$. We alternate optimization steps. With $W^{(2)}$ fixed, $x_2$ is interpolated by solving a linear system. Alternatively, holding $x_2$ fixed, $W^{(2)}$ is updated via proximal gradient descent (PGD). The proximal mapping of the rank term $Gamma(W^{(2)} - W^{(1)})$ is approximated in linear time using a fast orthogonal matching pursuit (OMP) algorithm that selects a sparse combination of atoms from a dictionary $cR$ formed by the outer products of $W^{(1)}$'s eigenvectors. We unroll iterations of our algorithm into layers to build a lightweight neural network for limited data-driven parameter tuning. Experiments show that our joint optimization achieves better signal interpolation compared to existing time-varying graph models.

Multi-agent systems are widely used in safety-critical applications that require coordinated behavior under strict safety constraints. Existing approaches face a fundamental trade-off: learning-based methods achieve strong empirical performance but lack theoretical safety guarantees, while control-theoretic methods enforce safety but often lead to overly conservative and inefficient behaviors. We propose a hierarchical multi-agent reinforcement learning framework that enforces hard safety constraints under mild assumptions at low level via a constraint manifold, while enabling effective coordination through high-level policy learning. Our approach provides theoretical safety guarantees in the multi-agent setting and yields stationary learning dynamics, thereby enabling stable and efficient training. Empirically, our method achieves competitive performance while maintaining nearly perfect safety rates, and generalizes effectively to varying numbers of agents and obstacles.

Continual learning remains a major challenge for modern deep networks, partly because commonly used optimizers lack inherent mechanisms for continual adaptation. One such natural mechanism is fast and slow adaptation to balance stability and plasticity. This mechanism has deep roots in neuroscience and biology, but there is no consensus on how to best incorporate it in commonly used optimizers. Here, we show that this can be easily done via the VCL framework, where past posteriors are used as priors in the future. Our key idea is to incorporate slow adaptation via merging of past posteriors to slow down the drift in the knowledge as learning progresses. The merged posterior is then used as the prior in the VCL update to implement the fast-weight updates. These steps can be seamlessly implemented in the IVON optimizer, whose form and costs are nearly identical to that of Adam. We call this new optimizer the Continual IVON (CoVON) optimizer and show that it not only consistently improves over existing VCL optimizers, but also performs better than other weight-regularization strategies across domain-incremental learning, continual pre-training, and fine-tuning of large language models.

Steering a large language model (LLM) toward a desired behavior typically relies on an iterative process of hand-crafting a prompt based on a careful inspection of the model's responses. This is an involved, brittle, and error-prone process. Preference-based fine-tuning is a more rigorous but often prohibitively expensive solution. We propose spec learning, a framework that relies on a brief user instruction and a small set of preference judgments. These are compiled into specifications in the form of natural-language prompts for an LLM. Specifications condition LLMs at inference time, and no parameter updates to the underlying models are required. We show that the responses generated based on the compiled specifications often outperform direct preference optimization (DPO) on datasets from specialized domains whose preference signal is dense. Unlike opaque weight updates, the resulting specifications are human-readable and double as interpretable and transparent written embodiments of the preference signal that produced them.

We introduce cyclic denoising -- repeated forward and reverse diffusion at controlled noise amplitudes -- as an extraction attack for image diffusion models. Inspired by random organization in disordered solids, cyclic denoising exposes regions of the learned distribution that are largely inaccessible to standard sampling. The dynamics drive samples toward attractors with a broad stability spectrum. The deepest attractors are ultrastable: they regenerate after near-total corruption and persist through thousands of noising-denoising cycles. Many of these attractors correspond to memorized training images, including stock photographs, brand watermarks, and web-crawl artifacts. The attack requires only sampler-level control, with no gradients, weight inspection, prompts, captions, or prior knowledge of the training data. Unlike generate-and-filter attacks, which rely on large-scale prompted generation and post-hoc similarity or membership-inference filtering, our main protocol is fully unconditioned. We demonstrate the phenomenon in Stable Diffusion v1.4 and in a pixel-space DDPM, showing consistent behavior across latent- and pixel-space diffusion models. Across noise amplitudes, we observe a yielding-like transition: low-amplitude cycling produces trivial absorbing fixed points or limit cycles, while larger amplitudes induce rearrangements, basin hopping, and long-lived trapping in structured memorized attractor basins. We also observe hierarchical partial absorption, prompt-stabilized basins, and cross-initial-condition universality of the recovered attractor set. Our results therefore show that cyclic denoising is both a physics-inspired probe of generative landscapes and a practical tool for memorization auditing, with implications for privacy, copyright compliance, and model fingerprinting.

Recent work has established that regularized policy gradient methods such as PPO, when used in self-play, can match or exceed specialized game-theoretic algorithms for solving two-player zero-sum imperfect-information games. The uniform distribution has emerged as a strong policy regularization target for this purpose, but it regularizes equally toward all actions regardless of their viability. We introduce EMAgnet, which instead regularizes toward an exponential moving average (EMA) of the last-iterate policy's parameters, providing an adaptive regularization target that evolves with the agent's improving strategy. We evaluate EMAgnet on both standard two-player zero-sum benchmarks and modified benchmarks with exploration challenges and large numbers of strictly dominated strategies. Relative to PPO self-play with uniform-magnet regularization under both linear and power-law annealing schedules, EMAgnet achieves lower exploitability in the majority of tested environments, with consistent performance gains across games containing strictly dominated strategies.

High-throughput scientific facilities such as the Large Hadron Collider depend on real-time event filtering (\textit{triggering}) under tight constraints on bandwidth, latency, and storage. In practice, trigger menus are largely static and hand-tuned and can become suboptimal as detector conditions, pileup, and background composition drift over time. We cast online threshold tuning as a sequential decision-making problem: a reinforcement learning agent ingests streaming summaries of recent rates and signal-sensitive features and updates trigger thresholds to maximize signal efficiency while tracking a target background rate within a tolerance band. We adapt Group-Filtered Policy Optimization (GFPO) to streaming control and introduce two variants (GFPO-F, GFPO-FR) that enforce background rate feasibility during training. On a benchmark that emulates realistic collider operation, we study two representative triggers: a total transverse energy ($H_{T}$) trigger sensitive to pileup variation, and an anomaly-detection (AD) trigger based on reconstruction loss for rare or non-standard signatures. On Monte Carlo streams, our agent increases the fraction of in-tolerance time intervals by 48\% ($H_T$) and 28\% (AD), with a cumulative gain of up to 2\% in signal efficiency on those in-tolerance intervals. Transferring from simulation to \emph{real} collision data (CMS Run 283408), the same agent, without fine-tuning, achieves a 56\% ($H_T$) and 28\% (AD) in-tolerance improvement over baselines, with further signal-efficiency gain on both triggers. To our knowledge, this is the \emph{first} demonstration of RL-based trigger control on real Large Hadron Collider collision data. Code is available at https://github.com/Zixind/GFPO\_LHC.

Clinical value sets define the standardized terminology codes used in quality measurement, phenotyping, cohort construction, and clinical decision support. The recently introduced Retrieval-Augmented Set Completion (RASC) benchmark showed that direct zero-shot large language model (LLM) generation is poorly suited to this task: clinical code systems are large, version-controlled, and not reliably memorized by language models. We study a stage-wise alternative in which candidate-pool construction is optimized for recall and a constrained LLM adjudicator is optimized for candidate selection. On the full 3,744-value-set RASC test split, Qwen3-based retrieval with vocabulary-aware expansion and code-display rescue retrieval increases candidate-pool recall from the original RASC retrieval baseline of 0.553 to 0.730; on the held-out-publisher stratum, pool recall is 0.655. The higher-recall pool alone is not sufficient: applying the original SAPBert cross-encoder to this expanded pool gives full-test macro F1 of 0.287 and held-out-publisher macro F1 of 0.233. Replacing the stage-2 selector with blinded GPT-5 adjudication over the same pool increases full-test macro F1 to 0.549 and held-out-publisher macro F1 to 0.533. These results show that retrieval-constrained LLM adjudication can substantially improve value set completion while preserving the safety constraint that all returned codes must come from an auditable candidate pool.

What is an agent? What constitutes agency? With the rise of Large Language Model (LLM) systems marketed as ``coding agents'', ``AI co-scientists'', and other ``agentic" tools that promise to drive up productivity, and at the same time, ``existential" concerns such as AI escaping human control with destructive power under a speculative ``machine agency" against humans, it has become essential to clarify where automation ends and agency begins, both for building capable systems and for understanding whether and what to fear. Drawing on Descartes' grounding of agency in independent thought, and on portrayals of autonomous beings in science fiction, we survey the current landscape of AI agents, and analyze agent architectures along five dimensions: goal, identity, decision-making, self-regulation, and learning. Specifically, we argue that genuine agency requires these structures to be \emph{internalized within the system itself} rather than assembled through external scaffolding. This distinction between \emph{agentic} systems, whose competence resides in engineered workflows, and \emph{agentive} systems, whose capabilities (including social interaction) arise endogenously, defines the boundary between systems designed for prescribed tasks, and those capable of operating in the open world with true autonomy. Building on this analysis, we propose the Goal-Identity-Configurator (GIC) architecture for a general-purpose agent model, combining hierarchical goal decomposition, identity evolution, simulative reasoning grounded in a separately trained world model, learned self-regulation, and self-directed learning from both real and simulated experience. Furthermore, we share insight on the auditability, controllability, and safety of agentive systems that possess greater autonomy and ``agency", but remain under human oversight.

End-to-end large language models (LLMs) produce fluent multi-document summaries but remain prone to hallucination, and the attributions they offer are typically coarse (whole documents or passages) and generated post hoc, leaving each summary statement hard to verify. We revisit the modular Extract--Select--Rewrite paradigm and recast its intermediate representation as the unit of attribution. We present CAMS, a Claim-Anchored Multi-document Summarization framework that (i) extracts atomic claims with token-level provenance from every source document, (ii) clusters equivalent claims across documents while flagging inter-source conflicts, (iii) selects a support-aware and salient subset, and (iv) rewrites the selection into a summary in which every sentence is anchored to a support-checked claim that links back to one or more source spans. Because content is localized before it is realized, the pipeline is attribution-oriented by construction and faithfulness-oriented by construction: it structurally preserves fine-grained, multi-source traceability while using support-aware selection, constrained rewriting, and verification to encourage, rather than guarantee, factual faithfulness. We evaluate quality, faithfulness, and localization on MultiNews, analyze conflict handling on DiverseSumm, and test zero-shot transfer on WCEP, using a two-regime protocol that separates reference-free citation quality from gold-aligned localization accuracy, and we add an evaluator-decoupled audit that tests citation precision with a support model never used for selection or verification. CAMS matches strong end-to-end and span-attribution baselines on summary quality while substantially improving faithfulness and citation precision, lifting multi-source attribution accuracy by roughly two-thirds, and exposing a controllable faithfulness--coverage trade-off that end-to-end models leave implicit.

Domain-Driven Design (DDD) is an influential software development methodology that structures software around business domain complexity through tactical building blocks such as Entities, Value Objects, Aggregates, and Repositories. Despite its prominence in software engineering, large-scale empirical evidence on how DDD is practiced, how it evolves, and how it relates to software maintenance quality in open-source projects remains scarce. This study presents a pre-registered empirical investigation of the distribution, evolution, and maintenance implications of DDD tactical building blocks in open-source GitHub repositories, addressing the call for large-scale empirical evaluations identified in a recent systematic literature review. We will collect DDD-related repositories from GitHub using the Search API, apply automated keyword filtering and manual relevance assessment, and analyze the resulting dataset through four research questions covering: (RQ1) the static distribution and co-usage of DDD building blocks across repository types, (RQ2) their longitudinal evolution over commit history, (RQ3) the extent and maintenance implications of Bounded Context boundary violations (the primary technical challenge in DDD adoption, unquantified at scale), and (RQ4) the temporal association between maintenance activities and building block churn or Bounded Context violations in DDD repositories. Together, these RQs trace the full maintenance and evolution lifecycle of open-source DDD projects and establish an empirical foundation for future DDD tool support and methodology refinement.

A single forward pass of a capable model is a fast, fluent, and unreliable problem-solver: it is right often enough to be useful and wrong often enough to be dangerous; in language models, such confident errors are known as hallucinations. We present Maestro Order, a model-agnostic orchestration harness that turns unreliable solvers into reliable problem-solving systems by composing them according to four structural primitives (decompose, ensemble, verify, and recurse) and a budget-aware controller that decides where to spend compute. The harness treats any model as a black-box base solver behind a uniform interface, layers a verifier ensemble whose discrimination is measured online, and allocates verification and voting to the stages with the highest marginal reliability per unit cost. We give the architecture, the message and state schema, the controller algorithm, and the engineering that makes it deterministic, observable, and fault-tolerant. We then specify an evaluation methodology (reliability at fixed cost, coverage, calibration, and ablations) and report results from a faithful Monte Carlo simulation of the harness over a parameterized solver/verifier model. The simulation reproduces the predicted laws quantitatively: verification amplifies reliability geometrically (e.g. $0.55\to0.98$ with two gates, $\to0.999$ with four), voting helps only above chance and is limited by shared errors, and a budget-aware controller reaches a target reliability at a small fraction of the cost of voting alone by selecting the cheapest mechanism for each regime. We close with failure modes (verifier gaming, correlated errors, and decomposition error compounding) and concrete guidance: build robust checkers, diversify solvers, and let the controller put compute where the information is.

We present an offline reinforcement learning (RL) framework for optimizing SLAM throughput control in a warehouse fulfillment environment. SLAM (Scan/Label/Apply/Manifest) throughput directly influences system congestion and operational efficiency. Our RL-based control approach dynamically recommends SLAM throughput settings that adaptively balance throughput maximization with downstream stability through intelligent adjustment of throttling behavior. We include a history-informed state representation, action space abstraction for delayed-impact control, and a reward function that captures both upstream and downstream operational metrics. Our approach is algorithm-agnostic, enabling integration of multiple offline RL methods under a unified architecture. We instantiate our framework with three state-of-the-art offline RL algorithms, and trained the models offline using de-identified historical operational logs from a large-scale warehouse. Policy performance is evaluated using a comprehensive multi-method strategy. These include model-free approaches including immediate reward estimation via regression models and long-horizon Fitted Q Evaluation (FQE), as well as model-based Deep Koopman dynamics evaluation. Empirical results reveal that the CQL policy consistently outperforms alternatives, improving system health by 22.97% and reducing average throttling duration by 3.18%. These findings demonstrate the potential of offline RL for safe and scalable warehouse throughput control optimization.

Efficient sorter diversion control of automated material handling systems (MHS) is critical for optimizing operational efficiency in large-scale warehouse environments. In this study, we use an inbound receiving sorter at a high-volume e-commerce warehouse as our primary use case, where the sorter diversion system relies on cost functions with static weight configurations that fail to adapt to highly dynamic system contexts, such as volume mode, congestion level, equipment physical status, and upstream/downstream dependencies. To address this real-time sorter diversion optimization challenge, we conducted a comparative study of three candidate hybrid machine learning frameworks: Linear Regression with Gradient Descent Optimization (LR+GDO), XGBoost with Bayesian Optimization (XGB+BO), and Bayesian Contextual Bandits (BCB). Model training and evaluation were enabled by leveraging a high-fidelity physics-aware emulator to overcome the cold-start problem and allow a safe transition from offline to online learning. We performed comprehensive evaluations including reward model predictive accuracy, contextual sensitivity, action distribution, and projected reward uplift. Our results demonstrate that while tree-based reward models offer slightly better predictive power, the BCB framework achieved overall higher performance with 2.03% reward uplift over the heuristic baseline. Furthermore, BCB exhibits several superior characteristics, such as its decisive time-optimal policy backed by Bang-Bang control theory, continuous online learning capability, strategic balance between exploration and exploitation, and significantly shorter inference latency. These results demonstrate the potential of the BCB framework for real-time control optimization in large-scale warehouse environments, motivating further investigation toward operational deployment.

Self-supervised learning in fluorescence microscopy often relies on 2D projections, despite the inherently three-dimensional nature of cells. We present a systematic comparison of 2D and 3D masked autoencoders (MAE-2D vs. MAE-3D) on volumetric microscopy data. Under matched architectures and training protocols, MAE-3D consistently outperforms 2D max-projection and slice-based variants on downstream single-cell tasks. We further align visual representations with a pretrained protein language model (ESM2) and show that cross-modal supervision yields larger gains for volumetric models. Channel cross-attention and frequency-domain regularization are critical for leveraging 3D spatial context. On a protein--protein interaction task, MAE-3D achieves a ROC--AUC of 0.865, outperforming prior methods by up to +0.025. For protein localization, our best 3D model attains state-of-the-art AUC$_{\text{micro}}$ (0.952) and F1$_{\text{micro}}$ (0.742), improving over previous approaches by +0.003 and +0.010 absolute, respectively. Overall, these results demonstrate the advantages of native 3D modeling and multimodal alignment for representation learning in single-cell microscopy.

Long-context and agentic LLM workloads push the KV cache past any fixed memory budget, forcing the inference stack to permanently evict tokens at every step of a continuous-inference stream. Existing methods all share the same template, a per-step direct-attention score followed by deterministic top-$K$ selection, which converts a single below-cutoff step into an irreversible verdict and permanently erases any subtly important token that direct attention cannot single out from noise. To address this challenge, we propose Nexus Sampling, a training-free eviction method that pairs Nexus scoring, an iterative walk over direct attention that surfaces bridge tokens, with weighted reservoir sampling, which retains tokens with inclusion probability in place of deterministic top-$K$. Theoretically, we show that Nexus Sampling dominates deterministic top-$K$ in long-run survival of subtly important tokens. Empirically, at 80% KV cache eviction, Nexus Sampling matches dense attention within 1% on LongBench while outperforming top-$K$ baselines on retrieval-heavy tasks, with up to 10x smaller per-sequence cache memory.

Because mathematics is highly abstract, a single statement can take very different forms depending on what subfield it is framed in. There are many examples where breakthroughs occurred after researchers discovered that a question had already been answered in a different field. At the same time, the growth of new resources related to formalization has increased the need for tools that enable efficient and reliable navigation between mathematical 'languages' (e.g., from Lean to natural language). In this paper, we investigate whether current embedding models capture mathematical equivalence. To do this, we introduce the Mathematically Equivalent but Lexically Different Pairs (MELD) Dataset, a collection of mathematically equivalent statements that are expressed in very different language. We show that current state-of-the-art embedding models tend to group statements by the terminology used to make them instead of the underlying math. Motivated by this, we propose a contrastive approach to learning embeddings of mathematical text that focuses on aligning informal statements with different formalizations. Our experiments demonstrate that this leads to improvements not only on informal-formal retrieval tasks but also on MELD, which only contains natural language statements.

Learning Koopman operators with autoencoders enables linear prediction in a latent space, but long-horizon rollouts often drift off the learned manifold, leading to phase and amplitude errors on systems with switching, continuous spectra, or strong transients. We introduce two complementary components that make Koopman predictors more robust. First, we add an attention-free latent memory (AFT) block that aggregates a short window of past latents to produce a corrected latent before each Koopman update. Unlike multi-head attention, AFT operates in linear time and adds only $\approx$30k parameters ($3d^2 + T^2$, fewer than matched multi-head attention), yet captures the local temporal context needed to suppress error divergence. Second, we propose dynamic re-encoding: lightweight, online change-point triggers (EWMA, CUSUM, and sequential two-sample tests) that detect latent drift and project predictions back onto the autoencoder manifold. Across three benchmark systems -- Duffing oscillator, Repressilator, IRMA -- our model consistently reduces error accumulation compared to a Koopman autoencoder and matched-capacity multi-head attention. We also compare against GRU and Transformer autoencoders, evaluated both from initial conditions and with a 50-step context, and find that Koopman+AFT (with optional re-encoding) attains markedly lower long-horizon error while maintaining lower inference latency. We report improvements over horizons up to 1000 steps, together with ablations over trigger policies. The result is a fast, compact predictor that stays on the learned manifold over long horizons.

Self-supervised and supervised speech models are increasingly used to investigate which linguistic information their internal representations encode, and at what level of abstraction they encode it. One underexplored phenomenon is consonant cluster reduction (CCR) in African American English (AAE), a widespread phonological process and a source of automatic speech recognition (ASR) disparity. To examine how CCR is represented, we conduct speaker-independent layer-wise probing of wav2vec2-base and Whisper-small using two tasks: segmental reduction detection and segmental restoration of underlying cluster identity. Both models distinguish reduced and canonical forms with high accuracy. Crucially, reduced segments retain cues to their underlying stops, indicating that CCR is encoded as structured gradient phonological variation rather than simple segmental deletion. These results demonstrate structured phonological encoding of AAE CCR patterns in modern speech models.

State--space models provide a flexible framework for analyzing dynamical systems, yet they often rely on Gaussian assumptions that fail to capture heavy-tailed or outlier-prone measurement noise. We propose a robust estimation scheme for linear state--space models subject to compound-Gaussian noise, as encountered for instance in radio interferometry affected by radio-frequency interference (RFI). The method relies on a Stochastic Approximation Expectation--Maximization (SAEM) algorithm in which the standard E-step is replaced by Monte Carlo sampling of the latent states and noise texture through closed-form Gibbs updates, enabling tractable inference despite the heavy-tailed likelihood. Numerical experiments show that the proposed method significantly improves reconstruction fidelity and robustness to RFI, outperforming a Gaussian EM algorithm and even an oracle RTS smoother. These results highlight the benefits of heavy-tailed state--space modeling and SAEM-based inference in interference-dominated imaging scenarios.

Tokenization is a foundational step in NLP pipelines, yet standard evaluation metrics such as fertility rate fail to capture morphological correctness for agglutinative languages. We present QuechuaTok, a systematic benchmark comparing four tokenization strategies - BPE, Unigram LM, WordPiece, and a morphology-aware PRPE tokenizer - for Southern Quechua (quz), a low-resource agglutinative language spoken by 8-10 million people in South America. Using a 200k-sentence corpus and the SQUOIA finite-state morphological analyzer (Rios, 2016) as silver standard, we evaluate three metrics: fertility rate, OOV rate, and morphological boundary accuracy (MorphAcc). Our results show that BPE achieves the lowest fertility rate (1.636 at 16k vocab) by memorizing surface word forms, while achieving only 6.67% MorphAcc. PRPE achieves 83.33% MorphAcc - the highest of all systems - demonstrating that fertility rate alone is insufficient to evaluate tokenizers for agglutinative languages. All code and models are publicly available at kaggle.com/code/macmaky/quechuatok

We present a large-scale empirical study isolating the contributions of the Derivative Regularization penalty (DREG). Across a fully-crossed factorial sweep of 960 experiments spanning 4 activations, 6 regularizers, 8 datasets, and 5 random seeds, we ask: when, where, and why does DREG work? Our results establish three principal findings. First, DREG achieves the highest overall and clean-regime accuracy among all regularizers evaluated (significantly so against the unregularized baseline, Weight Decay, and IGPen; Wilcoxon $p \leq 0.031$). It ranks second in noise robustness behind Spectral Normalization (SN) - the only two layer-wise regularizers in the study. Second, DREG is globally the best-performing regularizer under GELU, the default activation in modern transformer architectures, particularly on both messy vision and messy NLP benchmarks, suggesting direct applicability to frontier deep learning settings. Third, DREG's advantage over competing regularizers is most pronounced under data scarcity, consistent with its role as a geometric inductive bias that substitutes for the regularizing effect of data volume. Throughout, DREG is applied with a single fixed hyperparameter $λ= 10^{-2.5}$ and no per-dataset tuning, supporting its characterization as a plug-and-play regularizer for neural networks with nontrivial Jacobian structure. These findings are consistent with DREG's design: concentrating regularization pressure on layers where the activation derivative is largest, rather than constraining the network uniformly.

Droplet impact on solid surfaces is a complex fluid dynamics problem with applications in spray cooling, inkjet printing, and pharmaceutical processing. Although numerical simulations are widely used to investigate these dynamics, their computational cost becomes significant when multiple parametric variations are considered. In this work, we investigate the use of a Video Vision Transformer (ViViT) architecture to predict the temporal evolution of viscoelastic droplets impacting solid surfaces using volume fraction fields obtained from the Volume of Fluid (VOF) method. In Newtonian fluids, impact dynamics are mainly characterized by the Reynolds number $Re$, representing the ratio of inertial to viscous forces, and the Weber number $We$, representing the ratio of inertial to surface tension forces. For viscoelastic fluids, additional parameters are required to account for elastic effects, namely the solvent viscosity ratio $β$ and the Weissenberg number $Wi$, increasing simulation complexity and cost. Instead of simulating the entire droplet dynamics, the proposed approach uses only the initial 10% to 20% of the simulation to predict the remaining evolution. Depending on the prediction configuration, this strategy reduces computational cost by approximately 80% to 90% compared to full numerical simulations. The ViViT produces physically consistent predictions across different parameters and prediction horizons, successfully capturing both spreading and bouncing regimes while preserving geometric features and structural similarity. Since volume fraction fields can also be extracted from experimental videos, the proposed framework could be extended to incorporate experimental data during training, potentially improving the physical fidelity of the predicted dynamics.

Variable projection is a classical technique for separable nonlinear least-squares problems, in which variables that enter linearly are eliminated exactly, yielding a reduced nonlinear problem. By expressing this framework as a particular instance of a broader class of bilevel optimization problems, we develop a constrained variable-projection framework for data-science models, where the remaining variables are subject to convex constraints and the eliminated variables arise from a lower-level least-squares problem. In particular, by interpreting variable projection as a collapsed bilevel optimization problem, we derive exact reduced-gradient formulas compatible with automatic differentiation and propose a conditional-gradient algorithm for the resulting constrained reduced problem. We establish convergence guarantees under standard smoothness and compactness assumptions, and discuss extensions to structured lower-level variables. Numerical experiments on sparse autoencoding, dictionary learning, blind deconvolution, and few-shot learning suggest that the method can improve wall-clock efficiency and data efficiency relative to natural joint-optimization baselines.

Driving VLA models incorporating Chain-of-Thought (CoT) reasoning are attractive because they leverage pretrained VLM representations and expose intermediate decisions in natural language, yet current rationales often lack the step-by-step decision semantics needed to keep the rationale causally connected to the planned motion. We introduce Neuro-Symbolic Drive, a neuro-symbolic driving framework that supervises a driving VLA with rule-grounded reasoning traces extracted directly from classical rule-based planners. Our key observation is that rule-based planners are symbolic AI systems that already function as executable reasoning engines: they reason about active safety constraints, search over candidate maneuvers, and select a final trajectory. We instrument these planners in simulation to capture both the executed trajectory and the internal decision trace at each rule-evaluation step. Each trace is serialized into structured rule-grounded reasoning and paired with the trajectory to fine-tune Qwen3.5-4B as a driving VLA. Because these traces are derived directly from the planner states that determine the action, they ensure reasoning is structurally coupled to motion generation by construction, rather than by post-hoc alignment. On our simulator-generated benchmark, detailed rule-grounded reasoning reduces ADE@3s from 0.47 to 0.26 and miss rate from 8.30% to 6.40% under three-camera perception, and from 0.54 to 0.26 and 10.13% to 5.99% under eight-camera perception. Neuro-Symbolic Drive thus converts neuro-symbolic planning logic into structured supervision. Code base: https://github.com/XiangboGaoBarry/Neural-Symbolic-Drive.

Exact-match retrieval recall is often used as a proxy for whether a retriever supplies useful policy context to a downstream decision model. We test this proxy for pre-action policy classification in tau-bench using Qwen2.5-3B/7B classifiers. Under gold-policy conditioning, a compact structured state improves macro-F1 over raw trajectories by 0.13-0.17 after tuning. We then replace the benchmark-designated policy clause with the top-ranked clause retrieved from decision-time context. Although the exact governing clause is retrieved at rank 1 for only 7% of airline states, the primary 3B classifier obtains macro-F1 0.58 with retrieved clauses versus 0.60 with gold clauses (Delta=-0.02, task-cluster 95% CI [-0.23,+0.21]); mismatched-policy and no-policy controls score 0.32 and 0.21. We do not detect a macro-F1 difference between retrieved and gold clauses in this configuration, although the interval remains too wide to establish non-inferiority. The same qualitative pattern appears with a second retriever and at 7B, while varying across fine-tuning configurations. These results indicate that exact-match clause recall can underestimate downstream policy utility in this benchmark setting, motivating evaluation with retrieved policies in the classification loop rather than recall alone.

We study non-stationary linear contextual bandits where the reward model drifts over time, rendering classical contextual bandit algorithms brittle because historical data becomes systematically biased. We propose Flow-Corrected Thompson Sampling (fcTS), a Bayesian method that reuses experience by transporting past rewards to the present using an explicit drift model and incorporating each transported observation with a confidence weight that reflects transport reliability. This yields a unified template that specializes in (i) linear parameter drift via online slope estimation and reward correction, (ii) periodic variation via phase-aware reuse across cycles, and (iii) recurring regime switches via changepoint detection and regime-specific posterior memory. The resulting posterior updates remain closed-form under a linear Gaussian model and can be implemented efficiently with truncated, incrementally updated sufficient statistics. Across five controlled case studies and a semi-synthetic portfolio-selection benchmark with multiple overlapping non-stationarities, fcTS outperforms standard forgetting-based baselines (discounting, sliding windows, and periodic restarts), with the largest gains in settings exhibiting recurring temporal structure. These results demonstrate that when non-stationarity is structured, correcting and reweighting historical observations can be substantially more sample-efficient than uniformly discarding them.

Proximal Policy Optimization (PPO) is the standard policy-gradient algorithm for on-policy reinforcement learning. The literature presents it in two forms, a clipped surrogate that bounds the importance ratio between successive policies and a Kullback-Leibler penalty between them. These forms are treated as separate algorithms with their own gradients, their own hyperparameters, and their own reference implementations, and a sizeable body of empirical work compares them. We show that the gradient of the clipped surrogate is reproduced exactly by a Kullback-Leibler surrogate whose coefficient varies per sample, with closed-form dependence on the importance ratio and the advantage. The identity holds at every minibatch step and across the entire inner loop, and on five MuJoCo continuous-control benchmarks the two losses produce indistinguishable training curves. The reformulation exposes a structural feature of the clipped surrogate that the min notation hides. PPO-Clip's implicit per-sample penalty is a step function at the boundary of the trust region, and the shape of this coefficient is the natural design axis for generalising the algorithm. We sketch the resulting follow-up directions in the discussion.

Agentic AI systems powered by large language models (LLMs) are rapidly evolving into autonomous decision-making systems, exposing attack vectors beyond those of traditional LLM vulnerabilities. Existing security evaluations are often tied to specific implementations or domains, limiting unified comparison across heterogeneous systems. To address this gap, we introduce RIFT-Bench, a graph representation-driven methodology for dynamic red-teaming that enables unified evaluations across diverse agentic architectures. Building on a novel hierarchical representation, RIFT-Bench operates in two automated phases: Discovery, which extracts system structure, and Scanning, which deploys adaptive adversarial attacks and produces a comprehensive evaluation report. It evaluates the examined system itself, leveraging a broad set of dynamically adaptable adversarial probes across diverse attack vectors and objectives. We demonstrate the effectiveness of the proposed evaluation pipeline across 45 agentic systems spanning a diverse range of implementations, showing that the approach generalizes effectively to heterogeneous agentic architectures. Beyond systems and attacks, RIFT-Bench also supports direct evaluation of mitigation strategies. These key capabilities make RIFT-Bench a scalable foundation for security evaluation of agentic AI systems.

The task of compositional generation involves using a conditional generative model, trained only on a subset of the possible conditions, to produce samples from compositionally-defined target distributions such as a geometric combination of the source distributions. In this work, we argue that this task is often infeasible for vanilla conditional diffusion models: we conjecture that no inference-time technique can efficiently produce samples from the target distribution in certain well-motivated settings. This idea is supported by theory-guided generalization arguments and carefully-designed experiments on both synthetic and realistic data. In particular, while recent methods such as Feynman-Kac correction reduce inference-time approximation error, our results show that score estimation error has a more catastrophic effect on performance when the target distribution is out-of-distribution with respect to the sources, highlighting the need for a different approach to this task.

Practice often treats automatic metrics for attribution in LLM retrieval-augmented generation as interchangeable. We audit eight automatic scorers -- lexical, embedding, and BERTScore baselines alongside entailment/grounding-trained models (clean and FEVER NLI, the checker MiniCheck) -- across three evaluation constructs (provenance/topicality, generated-answer attribution, and fact-check entailment), asking whether any scorer transfers: stays within the 95% confidence interval of the best audited scorer on every dataset of a multi-dataset construct. In the construct with the most multi-dataset human-labeled coverage -- generated-answer attribution (AttributionBench's four source datasets, n = 1,610, with independent HAGRID, n = 2,150) -- none does: the per-dataset metric rankings invert (Kendall tau = -0.64, p = 0.031 on AttributedQA vs. LFQA), and an off-the-shelf NLI scorer that is best on short-claim AttributedQA (AUROC 0.90) collapses to AUROC 0.53 (chance) on long-form LFQA, where BERTScore wins (0.91); the flip is not a length or truncation artifact. This instability has a concrete decision cost: a naive "best-on-average" rule for choosing an evaluator fails leave-one-dataset-out (mean held-out regret 0.172 AUROC, worse than fixing one scorer), so metric choice must be validated on the target dataset rather than learned from others. A prompt-based LLM judge avoids the chance-level collapses the automatic scorers suffer (no LFQA collapse) but is not uniformly best, ~100x costlier, and non-deterministic -- relocating, not removing, the validation burden.

This article develops an architecture that creates a formally verifiable reward signal to train legal AI, adapting the LLM proposes, verifier disposes paradigm from mathematical AI to the distinctive demands of law. We present an architecture comprising LLM-driven autoformalization into a formal legal calculus extending Catala, a verification kernel, and explanation generation grounded in formal proof traces. For the computational components of law, the architecture provides provable correctness. For open-textured legal analysis, it provides structural guarantees: every required stage of the legal argument is addressed, argumentation is exercised at the correct stages and not omitted, and the deductive links between steps are valid. We demonstrate the architecture on procedural deadline calculations in German law, Commerce Clause analysis in U.S. constitutional law, and cross-jurisdictional sanction proportionality. We further show that the same architecture has a structural advantage for legal AI training: a deterministic external verifier supplies verifiable outcomes for legal problems and thereby closes the traditional reinforcement-learning loop gap in law.

Distilling conditional diffusion models aims to transfer the behavior of a large teacher to a smaller student while preserving alignment across conditioning inputs. Unlike recognition tasks, knowledge distillation in conditional diffusion often struggles to transfer knowledge beyond the training distribution, since the predicted noise strongly depends on the conditioning signal. As a result, effective distillation requires exploring a large conditioning space. In practical settings, this creates a major bottleneck. Paired image-condition data may be limited, and generating synthetic images for every available condition is often computationally infeasible, while the pool of conditions, such as text prompts, can be extremely large. Recent work addresses this issue by switching conditions during training, exposing the student to a broader conditioning space without changing the distillation objective. Yet this raises a complementary question: once a large conditioning corpus is available, how should the training effort be allocated? In this work, we introduce ARIA, a framework that adaptively allocates training effort across coarse regions of the conditioning space. By maintaining online estimates of teacher-student discrepancy at the region level, ARIA focuses updates where misalignment persists while preserving the original distillation objective. Empirically, ARIA improves over RC across most architectures and settings, with the clearest gains observed in unseen and underrepresented regimes. We also provide a theoretical analysis showing that the proposed tracking mechanism follows the evolving discrepancy during training under bounded variance and drift assumptions.

Prompt distillation compresses large vision-language models (VLMs) such as CLIP into lightweight student models by matching teacher predictions on unlabeled domain images. PromptKD (CVPR 2024) established this paradigm with a single PromptSRC-finetuned ViT-L/14 teacher and a ViT-B/16 student. We propose TheProfessor, a multi-teacher extension that distills from a fixed two-teacher ensemble: a domain-finetuned PromptSRC ViT-L/14 teacher and a zero-shot EVA-CLIP-L/14 teacher whose logits are pre-computed per dataset. We evaluate single-teacher PromptKD, equal-probability ensembling, and confidence-weighted ensembling on four base-to-novel datasets: Caltech-101, DTD, UCF101, and EuroSAT. In a 12-run single-seed sweep, confidence-weighted ensembling improves average HM from 87.52 to 89.28 (+1.77 points), while equal averaging improves average HM to 88.88 (+1.37 points). Gains are dataset dependent: they are negligible on Caltech-101 (+0.16 HM for confidence weighting), modest on UCF101 (+0.62), and largest on domain-shifted EuroSAT (+5.78). These results update our earlier Caltech-only analysis and show that multi-teacher prompt distillation is most useful when the second teacher contributes complementary supervision under domain shift.

While Vision-Language Models (VLMs) show great promise in volumetric medical report generation, they frequently suffer from visual hallucinations and a lack of grounding in 3D CT data. Current Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) strategies typically optimize text fidelity alone, essentially rewarding correct diagnoses derived from language priors rather than genuine visual perception. To address this, we propose cross-view aligned Evidence-driven Multimodal Reinforcement Learning (Evidence-MRL, noted as E-MRL), a reliable RL reasoning framework that formulates the generation process as a Markov Decision Process of "diagnosis-localization-verification". Unlike standard approaches, our model is explicitly trained to identify a "key evidence slice" alongside the global diagnostic report, grounding its findings in verifiable visual evidence. Crucially, we introduce a novel cross-view consistency reward, which validates the semantic alignment between the golden-standard report and a local visual re-query of the selected key slice, providing additional rewards for correctly-localized reasoning. Experiments on large-scale 3D CT tumor datasets demonstrate that E-MRL significantly reduces hallucinations and improves diagnostic accuracy compared to SFT and RL baselines, offering a clinically interpretable solution for visually-grounded and tumor analysis.

Representation alignment has emerged as an effective approach to improve Multimodal Large Language Models (MLLMs) by regularizing their internal representations toward those of an external vision encoder. However, existing methods typically align a fixed layer of the language backbone, overlooking the fine-grained structure of Transformer models. In this work, we propose Head-Wise Representation Alignment (HeRA), a method that enforces cross-modal alignment at the level of individual attention heads. Our approach is grounded in the Platonic Representation Hypothesis, focusing on preserving the topological structure of representations (i.e., their local neighborhood relationships) across modalities. Following the Mutual K-Nearest Neighbor (MKNN) alignment metric, we introduce a contrastive objective that acts as a differentiable proxy for matching local structures. HeRA applies this objective during multimodal training to specific attention heads in the LLM, selected by their alignment score according to the MKNN metric. Counterintuitively, we find that aligning the least aligned heads yields the largest gains. Extensive evaluations across multiple MLLMs and 18 benchmarks demonstrate that HeRA consistently improves performance on challenging vision-centric tasks and serves as an effective regularizer against visual hallucinations by naturally curbing the over-reliance on linguistic priors. Our code is publicly released.

General-purpose large language models (LLMs) are increasingly used for mental health-related conversations, yet safety safeguards remain inadequate and inconsistent across clinical conditions. This study evaluates six proprietary LLMs across 16 DSM-5 conditions using four adversarial attack variants, introducing an eight-dimension harm taxonomy and a multi-dimensional evaluation framework. Results show that safeguards hold reliably only for suicide and self-harm, while conditions such as eating disorders, substance use disorder, and major depressive disorder exhibit failure rates of up to 100%. We argue that ethical design and deployment of these LLMs demand clearly defined harm categories across clinical conditions and implementation of safeguards accordingly. Until such safeguards are in place, these models pose significant risks to vulnerable populations, making their growing integration into educational settings a particularly concerning.

Knowledge-Based Visual Question Answering (KB-VQA) requires grounding visual queries to external knowledge beyond directly observable content in images. While recent multi modal large language models (MLLMs) show strong perceptual abilities, they struggle on KB-VQA tasks requiring groundings from both fine-grained entity and evidence levels. Most existing multi-modal retrieval augmented generation (MM-RAG) methods tightly couple entity discrimination and section-level evidence ranking into a single re-ranking stage, leading to high cost and limited generalization. In this work, we revisit existing MM-RAG solutions from a workflow perspective and argue both entity-level and fact-level groundings are key bottlenecks. We observe that although MLLMs often fail under open-ended entity naming, they can better identify the correct entity when selecting from a small set of candidate names. Based on this insight, we propose a simple and training-free identify-before-answer IBA framework that decouples entity identification from section-level re-ranking. Our approach prompts an MLLM to select high-confidence entities using only candidate names, followed by an off-the-shelf textual re-ranker for evidence selection. Experiments on Encyclopedic-VQA and InfoSeek show that our method consistently outperforms fine-tuned multi-modal re-ranking baselines while reducing training and inference complexity. Additional analyses reveal that the improvements arise not only from better entity identification, but also from selecting more informative evidence once correct entity is fixed. Our implementation is made public to ease reproducibility.

From climate teleconnections to gene regulation, modern time-series datasets encompass tens or hundreds of interacting variables, making causal discovery increasingly challenging. Constraint-based methods offer statistical rigor but their nonlinear CI tests are infeasible at scale, while score-based alternatives avoid CI testing but require arbitrary thresholds to binarize continuous edge scores. We propose GRACE ($\textbf{G}$ated $\textbf{R}$efinement for $\textbf{A}$ccurate $\textbf{C}$ausal $\textbf{E}$dge discovery), which refines constraint-based discovery using Hard Concrete gates with $L_0$ regularization: each candidate edge has an independent gate whose values concentrate near 0 or 1, yielding a clean bimodal separation that makes the binary decision robust, unlike the narrow, overlapping score distributions produced by $L_1$ and attention-based methods. A fast linear CI skeleton provides high-recall candidates; a single gated model then prunes false positives by learning which edges genuinely improve prediction, with automatic regularization adapted to problem dimensions and skeleton density. Systematic experiments on synthetic benchmarks, spanning diverse graph topologies (scale-free, Erdős-R'enyi, small-world) and dimensionalities up to $d=100$, show that GRACE substantially improves F1 over its base CI method while maintaining high precision, and outperforms attention-based and score-based alternatives. GRACE matches or exceeds expensive nonlinear CI tests at a fraction of the cost ($75\times$ faster). On a real-world river flow dataset, where rainfall confounders, variable propagation lags, and distributional shifts violate standard assumptions, a temporal bootstrap variant of GRACE recovers 9 of 11 causal edges along the Elbe River with only 1 false positive ($F_1 = 0.86$, AUROC${} = 0.99$), reducing the skeleton's 106 false positives by 99%.

Purpose: To evaluate the feasibility and challenges of heart chamber segmentation from non-contrast CT scans using contrastive unpaired image translation and deep learning-based segmentation. Approach: We developed ChameleonNet, a framework utilizing the Contrastive Unpaired Translation (CUT) network with decoupled contrastive learning (DCL) loss to synthesize non-contrast CT from contrast CT scans. Using annotations of four heart chambers (left atrium (LA), left ventricle (LV), right atrium (RA), and right ventricle (RV)) from contrast scans, we trained a Hausdorff distance loss-enhanced nnU-Net on synthesized non-contrast images. The translation model was trained with 35,538 contrast-enhanced and 37,197 non-contrast CT slices. The segmentation model was trained with 292 synthesized non-contrast scans. Performance was evaluated using Dice similarity coefficient (DSC) and 95th Hausdorff distance (HD95) on 36 synthesized non-contrast scans, and volume agreement on 36 real non-contrast CT scans was assessed using Pearson correlation, mean absolute percentage error (MAPE), and mean percentage error (MPE). Results: The segmentation model achieved DSC of 0.94 (0.01), 0.91 (0.04), 0.92 (0.03), 0.93 (0.02), and HD95 of 3.63 (1.49), 5.74 (4.08), 5.18 (1.77), 5.51 (3.21) mm on synthesized non-contrast images for LA, LV, RA, and RV, respectively. On real non-contrast CT scans, Pearson correlations were 0.93, 0.82, 0.87, and 0.89 (all p<0.001), with MAPE ranging from 9.22% to 20.79%, and MPE ranging from -12.52% to 4.67%. Conclusions: ChameleonNet demonstrated feasibility for heart chamber segmentation from non-contrast CT without manual non-contrast annotations. However, volume errors, particularly for LV and RV, indicate that further refinement and validation are needed before clinical use.

Jupyter Notebooks are an increasingly popular coding environment used across many domains, especially in Python-based data science and scientific computing. Originally used for prototyping and interactive exploration, notebooks are increasingly used to develop more complex programs, leading to a rapid rise in buggy notebooks on platforms like GitHub. To address this trend, we present JupOtter, a bug detection system designed specifically for Jupyter Notebooks. JupOtter features three novel contributions: (1) a notebook-specific tokenization strategy that preserves cell structure, (2) a cell-level bug prediction technique, and (3) a new labeled dataset, OtterDataset, containing over 21,000 notebooks annotated for fine-grained cell-level bug detection. JupOtter achieves cell-level bug detection F1 scores that surpass static analyzers and large language models in two out of three evaluation datasets.

As generative models increasingly produce samples that are indistinguishable from human-created content, it becomes difficult to determine whether a given data point was part of a model's natural training set or was generated by the model itself, especially when models memorize and reproduce training data. We formalize this challenge as Member vs Generated Inference (MGI): given a sample and a target generative model, infer whether the sample is a true training member or a generated output of that model. Focusing on image generation, we show that existing membership inference methods systematically misclassify generated samples as training members, while attribution-based methods often misclassify true members as generated. This failure arises because both approaches rely on likelihood-related signals that are similarly elevated for training examples and for the model's own outputs. To address MGI, we propose Data Circuit Breaker (DCB), a three-stage method that combines complementary signals from a generative model's autoencoder and latent generator to distinguish training members from generated samples. Across multiple generative models, including image autoregressive and diffusion models, DCB consistently addresses the shortcomings of membership inference and attribution methods, remains effective even when models reproduce near-duplicates of training samples, and generalizes to challenging model derivative settings in which new models are trained on generated data.

Survival analysis is central to clinical decision-making, yet reliable time-to-event models require large, diverse cohorts that are rarely available at a single institution, while privacy regulations restrict the centralization of patient data. Federated learning (FL) offers a privacy-preserving alternative by training shared models without exchanging raw data, but its effectiveness for survival modeling under realistic, heterogeneous conditions remains insufficiently understood. This paper presents a systematic, multi-model evaluation of federated survival analysis on a cross-institutional breast cancer cohort with naturally heterogeneous distributed clients. Three representative survival models, the Cox Proportional Hazards model, DeepSurv, and Random Survival Forest (RSF), are compared across centralized, local, and federated training, and three federated optimization strategies (FedAvg, FedProx, and FedAdam) are assessed for the gradient-based models. Results show that FL consistently outperforms local training and approaches, and occasionally exceeds, centralized performance, while RSF offers the best overall balance of discrimination, calibration, and robustness across heterogeneous clients. We further find that performance depends on the diversity of client distributions, and that FedAvg and FedProx are stronger and more stable than FedAdam. Based on these findings, we derive practical, decision-oriented guidelines mapping data, privacy, interpretability, and resource constraints to recommended model and training-paradigm choices for federated survival modeling in healthcare.

PLCverif is the most mature open-source platform for PLC formal verification, developed at CERN and in production use since 2019. Yet it has two fundamental limitations: no support for Ladder Diagram (LD) programs, the dominant PLC notation, and reliance on CBMC as its primary backend, which restricts verification to bounded proofs. The PLCverif authors themselves identified ESBMC as the appropriate backend improvement. Prior work established ESBMC-PLC (a textual LD frontend with k-induction) and ESBMC-GraphPLC (graphical PLCopen XML support); together, they cover LD with unbounded proofs but not Structured Text (ST), and graphical LD with timer/counter function blocks remains unverifiable. This paper presents ESBMC-PLC+, a unified framework that closes both gaps: (1) an ST/SCL frontend via the MATIEC IEC 61131-3 compiler, routing C-compiled ST to ESBMC with nondeterministic input modeling and YAML property injection; (2) function block state semantics for graphical LD, extending the DFS resolver to model TON/TOF/TP timers, CTU/CTD counters, and R_TRIG/F_TRIG edge triggers as persistent scan-cycle state variables in the GOTO IR. ESBMC-PLC+ is the first open-source PLC verification framework to support all three major IEC 61131-3 input formats via a single ESBMC backend, enabling k-induction-unbounded safety proofs. A feature comparison with PLCverif and experimental evaluation on 8 benchmark programs, including programs with up to 8 integer timers, shows that ESBMC-PLC+ matches PLCverif's input coverage while providing stronger guarantees. Against nuXmv's BDD backend, ESBMC-PLC+ is 400-2,000x faster on timer programs and completes proofs where nuXmv BDD times out at 120s.

The exact computation of the Normalized Maximum Likelihood (NML) codelength for regular non-smooth estimators (e.g., Lasso) has been historically limited by the cubic scaling walls of manifold-constrained projection and volume integration. At each step of the geometric Propose-and-Project Metropolis--Hastings (PPMH) sampler, evaluating the projection operator requires inverting an $(N+k) \times (N+k)$ generalized KKT matrix, while calculating the volume factor requires the determinant of an $(N-k) \times (N-k)$ Gram matrix. This paper presents an exact, mathematically equivalent formulation that bypasses both bottlenecks by utilizing the block Schur complement and Sylvester's determinant identity. We prove that the computational complexity of both operations collapses from $\mathcal{O}(N^3)$ to $\mathcal{O}(k^3 + N^2 k)$ per step. We generalize this reduction to Sparse Support Vector Machines (SVMs), Elastic Net, and Group Lasso. Finally, we provide a rigorous numerical stability analysis and evaluate the sampler's efficiency using the Effective Sample Size (ESS) per second. Our empirical benchmarks on high-dimensional datasets confirm a constant speedup exceeding $14{,}100\times$ while maintaining double-precision numerical equivalence, rendering exact non-smooth NML estimation highly tractable for large-scale statistical inference.

A primary challenge in AI safety is the existence of adversarial examples -- slightly distorted inputs that cause a neural network (NN) to misclassify. To mitigate this problem, recent research focuses on the computation of robustness certifications, which, for a given input, determine the largest distortion the input may receive without breaking the network's prediction. Robustness certifications can be interpreted as an axis-aligned hyper-rectangle (multi-dimensional intervals). Most existing approaches focus on maximizing the certification's volume, but recent intractability results prohibit the computation of volume-optimal certifications in reasonable time. We introduce the apothem measure and show how to compute apothem-optimal certifications in a linear number of calls to a NN verifier (oracle) w.r.t. the input domain's diameter. Moreover, we prove that we cannot have a volume-optimal, oracle-based algorithm, even if we discard the oracle costs. Also, we introduce dual certifications -- an interval including all instances of a class -- thus providing apothem-minimum upper bounds to a robustness certification. Further, we present the ParallelepipedoNN system, which we evaluate on the standard MNIST and Fashion MNIST benchmarks. A preliminary comparison with existing work on the same datasets reveals at least two-fold improvement w.r.t. the minimum edge length.

Generative molecular models for drug design are a promising direction with much active research. In the next phase of computational drug design, such models will need to understand small molecule structure and protein-ligand interactions, and they will need to possess the machinery to generate molecules \textit{de novo}. Incorporating each feature poses a critical challenge. Equally important, yet often treated as secondary, is the ability to grow a molecule from a partial starting point -- a scaffold or fragment supplied by a chemist -- which is the central operation of lead optimization. We present Sesame (Spatial Evoformer for a Structure-Aware Molecular Engine), a diffusion-based molecular generation model that leverages a novel spatial pairformer module to condition on partial molecular structure and the surrounding protein pocket, both expressed as continuous spatial density maps. This single conditioning mechanism supports both \textit{de novo} generation and fragment-conditioned lead optimization, letting a medicinal chemist prune a hit to a scaffold and have Sesame grow it in productive ways. In addition to this module, we also introduce a diffusion framework for joint denoising of atom types, bond types, and positions, along with a trajectory finetuning scheme that trains on the model's own sampling rollouts to improve generation quality. Sesame is trained on a large corpus of ligand-only and protein-ligand datasets.

This paper studies strict majority reasoning in finite electorates using so-called $\textit{social decision frames}$: finite sets of voters equipped with distinguished families of coalitions interpreted as those voting blocs evaluated to form a strict majority. A coherence criterion for qualitative majority judgments is identified and shown to give an exact characterization for representability of strict majorities by finitely additive measures. In addition, a minimal natural logic for reasoning about strict majorities is shown to be sound and complete. These developments motivate examination of associated combinatorial questions concerning incoherence in finite families of sets; partial results and a conjecture are given. Finally, the results of this paper are applied to correct a classical representation theorem for weak qualitative probability structures due to Patrick Suppes and to establish a May-type characterization for ordinary strict majority rule for social decision frames.

This work investigates the implementation of artificial intelligence and machine learning (AI/ML) for real-time monitoring in laser powder bed fusion (LPBF) additive manufacturing. We developed a binary image classification framework for distinguishing normal and abnormal melt pool images using a balanced dataset of 1,200 images collected from Nickel superalloy 625 on the NIST AMMT platform. The study evaluates accuracy and inference time based on control requirements and hardware limitations of open-architecture LPBF machines. We benchmark three transfer learning architectures (ResNet50, EfficientNetB0, and MobileNetV2) against two Random Forest approaches: one trained on EfficientNetB0 feature embeddings (hybrid) and one trained on raw pixel features (baseline). Images are stratified into 80/20 train-test splits, with a further 90/10 validation split on the training set, and undergo standardized resizing, normalization, and label-preserving data augmentation to emulate realistic process variability. Each model is evaluated using accuracy, precision, recall, F1 score, and area under the receiver operating characteristic curve (AUC), along with training time, inference latency, and CPU & GPU usage to capture deployability constraints relevant to factory-floor monitoring. The hybrid EfficientNetB0-plus-Random Forest approach achieves the best performance on the held-out test set, with an F1 score of 0.9451, accuracy of 0.9458, and AUC of 0.9904, while maintaining sub-millisecond per-image inference (1.15 ms). In contrast, purely deep learning models exhibit significantly higher inference times with lower accuracy. These results demonstrate that combining pre-trained convolutional features with classical ensemble methods provides a robust, computationally efficient route to real-time melt pool anomaly detection in data-limited additive manufacturing environments.

When two or more parameters or labels produce similar data, they are degenerate, or hard to distinguish. Degeneracies render both label prediction and inverse problems difficult, since both machine learning algorithms and probabilistic samplers rely on the distinguishability of data and its gradients with respect to parameters. However, identifying degeneracies in physical models or real-world datasets can be elucidating about the choice of model or the underlying process that produces the data. We present the degeneracy distillery, a method that (1) detects and (2) resolves degenerate parameter combinations (a) automatically and (b) symbolically, from parameter-data (or parameter-simulation) pairs alone, through estimation and flattening of the Fisher information matrix. By exploring the information geometry of the likelihood, we characterize degeneracies as an intrinsic property of the physical model, requiring no realised data observation. We demonstrate our approach on a range of synthetic and real-world problems, discovering symbolic coordinate transformations that identify the combinations of parameters of a model which yield independent effects on the data. The resulting coordinates flatten the Fisher information in expectation globally, in contrast to posterior-based methods that flatten only at a single point, and substantially reduce the simulation budget required for downstream neural posterior estimation. In test cases we require up to $10\times$ fewer simulations for posterior estimation at matched validation calibration whilst simultaneously gaining physical insight on the system.

Terrestrial water storage (TWS) integrates snow, soil moisture, surface water, and groundwater and is a key indicator of how climate variability and human activity reshape the global water cycle. The GRACE and GRACE-FO satellite missions provide the only direct, globally consistent observations of TWS change, but their record only begins in 2002 which is too short for many climate-scale analyses. We present a deep learning application that reconstructs monthly GRACE-like TWS anomalies (TWSA) back to 1940 by learning the relationship between daily ERA5 meteorological forcing (precipitation, evapotranspiration, runoff) and monthly GRACE observations. In contrast to prior reconstruction approaches based on grid-cell-wise regression, CNNs, or LSTMs, we adapt a multi-variate time series graph neural network (MTGNN) architecture, which was originally developed for mobility and traffic forecasting on urban sensor networks to this satellite-geodesy task. Spatial dependencies are encoded in a static, interpretable hybrid adjacency matrix that combines geodesic proximity with lagged correlations of climatic time series, capturing both local hydrological coupling and large-scale teleconnections. The reconstruction achieves a grid-cell Pearson correlation of 0.69, a basin-mean correlation of 0.94, and a near-zero bias, and it reproduces the spatial fingerprints of the 2015/16 El Niño and 2020/21 La Niña events. A systematic comparison with established reconstruction approaches (GTWS-MLrec, RM-REC, GRAiCE) shows that the graph-based model is statistically competitive at basin scale, reaching a correlation within 0.025 of the best baseline while using only roughly half to a tenth of the predictors the other models require and revealing characteristic weaknesses in arid regions in all models. The complete implementation is publicly available at github.com/hcu-cml/MTGNN-TWS-Reconstruction-GRACE

The use of dedicated corridors for Advanced Air Mobility (AAM) traffic is one of the most commonly proposed pathways to integrating them into existing airspace operations. Most prior research has focused on the design of networks of AAM corridors and conflict resolution for aircraft within corridors. It is also generally believed that while attractive from an implementation perspective, corridor-based operations may be inefficient, especially in the absence of centralized traffic management. In this paper, we show that contrary to this belief, it is possible for autonomous aircraft to learn to self-organize into corridor flows in decentralized settings. We illustrate our approach using scenarios in which fixed-wing aircraft need to safely and efficiently traverse (1) a single corridor with metering after the exit, (2) a sequence of two consecutive corridors, and (3) a corridor that splits into two. We find that in decentralized settings with only local information, the aircraft are able to conform to the corridor boundaries more than 94% of the time and reach their goal in a relatively efficient manner. Furthermore, tactical interventions to handle violations of the separation minimum are needed only infrequently in low- and medium-density settings. However, such tactical interventions become more frequently necessary only when traffic density is high.

Molecular surfaces encode the geometric and physicochemical patterns that determine antibody-antigen recognition, central to epitope prediction. However, existing methods rely on sequences or backbone structures and struggle to capture discontinuous, surface-driven epitopes. This study presents SurfBind, a surface-centric learning framework for epitope prediction that operates directly on molecular surface representations. SurfBind integrates geometric and physicochemical cues through a Transformer-based architecture with patch-level surface modeling, binder-aware cross-attention, and a hierarchical coarse-to-fine prediction paradigm. Experiments on challenging epitope identification benchmarks, including SAbDab and DB5.5, demonstrate that SurfBind achieves state-of-the-art performance and strong generalization across unseen antibodies and conformational states, highlighting the value of interaction-aware surface modeling for understanding the crucial mechanisms of protein-protein interactions.

A data-driven method is developed for approximating value functions in deterministic optimal control problems with nonlinear control-affine dynamics. The Pontryagin Maximum Principle optimality system is solved from multiple initial conditions to generate training data consisting of values, gradients, and Hessians of the value function, where Hessian information is obtained from a matrix Riccati equation along optimal trajectories. These quantities augment a weighted least-squares regression over sparse polynomial bases on hyperbolic cross index sets, with gradients and Hessians contributing additional linear equations per sample and substantially reducing sample complexity compared to value-only regression. Feedback laws are recovered analytically from the learned value function. In high dimensions, a partial Hessian strategy controls the cost of data generation. The approach is validated on problems of increasing state dimension, where second-order data augmentation is shown to improve approximation accuracy and closed-loop performance, with up to an order-of-magnitude reduction in the number of training samples required relative to lower-order methods.

Efficient small object detection is bottlenecked by the inherent feature scarcity of tiny targets, which is further aggravated by operations of spatial-domain detectors that indiscriminately discard critical high-frequency details. Recovering these fragile cues within the spatial domain is notoriously difficult, as it often requires computationally expensive architectural upscaling that inadvertently amplifies background noise. To bridge this gap, we propose a paradigm \textbf{shift from spatial to spectral} feature processing, introducing a holistic solution with the following novelty: (1) A versatile \textbf{Frequency-Guided Feature Representation framework} that generalizes across diverse detector architectures (both CNN and Transformer-based), offering a robust alternative to spatial-only feature extraction; (2) The unified \textbf{Decompose--Enhance--Reconstruct (DER)} operator, instantiated via three \textbf{lightweight, plug-and-play} modules -- Wavelet-Difference Gate (WDG), Log-Gabor Enhancer (LGE), and Frequency-Driven Head (FDHead) -- to systematically inject frequency-aware modulation into the backbone, neck, and head. This mechanism decouples feature modeling from resolution reduction, capturing discriminative high-frequency components to enable accurate localization with significantly reduced parameter redundancy; (3) Extensive validation on multi-domain benchmarks (VisDrone2019, UAVDT, TinyPerson, DOTAv1) demonstrating consistent gains. Notably, our proposed \textbf{DERNet} series outperforms YOLOv11 models under the same scale while requiring \textbf{only 1/6 of the parameters}, backed by rigorous spectral diagnostics and error decomposition analysis.

Accurate numerical eigenvalues are often difficult to certify, especially in singular or non-normal settings. This article reports a human--AI collaboration on two such computations. For a singular self-adjoint Schrödinger operator, a verified zero count and Dirichlet--Neumann bracketing certify the complete negative spectrum to ten decimal places. For a delicate non-normal atom--molecule benchmark, a previously unresolved resonance pair is separated, with each member enclosed to ten digits. The second result is achieved not by increasing the precision of one-way shooting, but by reformulating the problem as a global matching system for projective solution lines. The infinite tail is encoded as uncertainty in the terminal projective data, and a componentwise, tail-robust Krawczyk--Brouwer inclusion supplies the certificate. This gives a reusable architecture for analytic boundary-value systems with ill-conditioned propagation and uncertain asymptotic data. The collaboration also exposes the strengths and limits of AI assistance. AI rapidly produced accurate candidates and plausible proof strategies, but several failed, including one apparently complete tail argument that omitted the componentwise check required by a nonuniform polydisc. Validated computation is a stringent test of AI-assisted mathematics: the output is not merely a number, but a number with a proof. These examples show why the proof object matters, and why human mathematical judgment remained decisive. More broadly, as AI makes code, exposition, and plausible numerical claims inexpensive, standards for verification, attribution, peer review, and training must adapt. The implications are unsettling; the opportunity is extraordinary.

Graph and multi-agent orchestration frameworks make production large language model (LLM) workflows practical, but they do not by themselves solve conversational continuity when users maintain several interdependent objectives. This conceptual systems paper focuses on the high-complexity end of that design space, where goals can be suspended, resumed, revised, and invalidated by actions in other goals. We introduce the Goal-Oriented Dialogue Runtime (GODR), a framework-neutral design pattern that treats goals, task frames, lifecycle state, invalidation rules, and resumption contracts as first-class runtime objects while delegating bounded execution to graph runtimes, agents, tools, or application programming interfaces (APIs). GODR is not proposed as a replacement for workflow graphs in simple guided processes; it is intended for complex, multi-domain, interruptible conversations where objective continuity cannot be recovered reliably from agent identity, chat history, or execution-graph position alone. The paper formalizes the problem, proposes runtime objects and architecture-selection criteria, and frames evaluation as an agenda for future empirical validation rather than as a measured performance claim.

Extended Reality (XR) presents a challenging use case for 5G and 6G networks, requiring high data-rates and lowlatency communication to deliver a truly immersive experience. Moreover, in order to seamlessly translate physical actions to the virtual world, accurate gesture recognition and pose estimation are required. Current XR interaction solutions based on handheld controllers and cameras cannot easily capture full-body poses, inhibit the free use of hands, and require good visibility and a clear line of sight. In this work, we propose a multimodal sensing architecture for XR that combines 5G MillimeterWave (mmWave) Integrated sensing and communication (ISAC) and surface electromyography (sEMG) signals. 5G mmWave ISAC cannot only be used to deliver content wirelessly to the Head-mounted display (HMD), but also the same communication signals can be used to derive coarse body-level gestures and poses of the user, to support real-time avatar control. For fine-grained finger-level gestures, our architecture leverages lightweight sEMG sensors that capture forearm muscle activity. To illustrate the need of both modalities, we present evaluations of both sensing technologies. At the body level (5G), our architecture relies on power-per-beam-pair (PPBP), which can be computed from standard beam management or beam sweeping procedures of the 5G NR standard. PPBP-based sensing achieves 82.2$\pm$5.9% average accuracy when evaluated on users not seen during training. For fine-grained finger-level interactions, we show that surface electromyography (sEMG) carries strong discriminative information achieving consistent promising performance across different movement settings. Thus, combining the two modalities enables multi-scale gesture recognition, at the body level via existing 5G signals and finger level via lightweight sEMG sensors, forming a complete XR framework.

The tracking-by-detection paradigm in multi-object tracking (MOT) typically relies on static appearance descriptors to complement motion estimation. However, these descriptors are frame-independent, limiting their robustness as visual cues. Since such descriptors are often obtained from computationally intensive pretrained backbones, real-time MOT systems frequently abandon appearance cues altogether and rely solely on motion prediction and geometric association. In this work, we introduce Polycepta, an object-centric appearance state estimation framework that reformulates appearance modeling as a recursive estimation problem rather than a frame-wise matching task. Polycepta constructs and continuously updates an independent appearance state for each tracked object, enabling future appearance representations to be estimated from accumulated observations. Polycepta is encouraged to learn the appearance-state construction of object-specific representations rather than memorize them through a proposed learning strategy, enabling appearance estimation for unseen classes. A key property of Polycepta is that the quality of appearance estimation improves as object states evolve during inference. While conventional appearance descriptors remain static or degrade over time, Polycepta progressively refines appearance estimates as additional observations are accumulated. Extensive experiments on KITTI, the Waymo Open Dataset, and MOT17 demonstrate consistent reductions in identity switches and improvements in tracking performance when integrated into the tracking-by-detection pipelines. Polycepta operates at 90.57 Hz and delivers state-of-the-art performance on the KITTI benchmark when integrated into the RobMOT framework, achieving a MOTA of 92.27\%.

Computer-use agents can execute software tasks through either graphical interfaces or programmatic command interfaces, but existing evaluations confound interaction modality with differences in tasks, initial states, verifiers, and permitted actions. We introduce a matched execution-layer benchmark of 440 desktop tasks across 18 applications and 12 workflow categories, where screen-only GUI agents and skill-mediated CLI agents receive identical goals, states, and final-state verifiers while being restricted to modality-native actions. In this controlled setting, the strongest GUI agent reaches a 59.1% full pass rate, outperforming the strongest original-skill CLI agent at 48.2%; however, verifier-guided skill augmentation raises CLI success to 69.3%, showing that much of the CLI deficit comes from incomplete skill coverage rather than model capability alone. These results suggest that GUI and CLI expose different execution bottlenecks: GUI agents are limited by reliable grounded interaction over long-horizon workflows, whereas CLI agents are limited by the coverage and scalability of their skill interfaces.

We propose cryptographic certificates of validity for agentic AI systems. The core idea is to formally specify a correctness or policy condition as a logical predicate, compile this predicate to a witness-checking problem over polynomial constraints, and use a succinct cryptographic proof system (and optionally zero-knowledge) to certify that the condition holds. This offers a middle ground between formal verification of source code, and cryptographic authentication. An agent's action can be accompanied by an independently checkable proof that it satisfies an agreed formal policy, without requiring the verifier to trust the agent or to re-execute computation. We outline the approach at a high level, give the core mathematical translation, relate the proposal to proof-carrying code, zkVMs, formal methods, and agent governance, and note the specification, auditing, and deployment questions that a full implementation must answer.

Language-specific tokenizers improve tokenization quality and the downstream performance of models on those languages. However, using such a tokenizer comes at a cost: either a new model must be trained from scratch, or the vocabulary of an existing pretrained model must be adapted. We propose Language-adaptive Maximum a Posteriori (LangMAP) Tokenization, a tokenization scheme that extends the UnigramLM algorithm to the multilingual setting, producing language-specific tokenization from a single shared vocabulary. Notably, LangMAP can be used when training a multilingual language model from scratch or to adapt a pretrained model's tokenizer to individual languages without changing its vocabulary. While language labels are required at training time, a key feature of the algorithm is that it then performs language-specific tokenization at inference without knowledge of the input's language. Across 14 open-source tokenizers, 9 natural languages, and 9 programming languages, LangMAP improves morphological boundary alignment and, for all coding languages tested, alignment with abstract syntax tree (AST) leaf boundaries. In fine-tuning experiments, results are mixed: LangMAP improves target-language grammatical acceptability (MultiBLiMP) on the languages tested; its benefits are less consistent on knowledge-related tasks (Global-PIQA, Belebele).

Headline accuracies on the Tuebingen cause-effect pairs are routinely compared across papers even though each is measured under its authors' own protocol -- different pair subsets, weightings, model-selection, and decision rates. We argue this is the wrong comparison and run the right one: a same-hands re-evaluation in which every method is run by us on the identical 102 pairs, with one strict rule -- no tuning and a decision forced on every pair. As a clean reference point we introduce a deliberately minimal baseline: sorted-conditional compression, which feeds quantized, sorted, first-differenced data to an off-the-shelf compressor (bz2) and has zero fitted parameters. Under the common ruler the ranking differs sharply from the literature. Our baseline reaches 74.7% weighted accuracy (p = 3.7e-7); on the same 100 pairs that SLOPE is evaluated on it scores 76.0%, a 1.2-point gap below the authors' own forced-decision SLOPE (77.2%) that is well inside noise (McNemar p = 0.39). A faithful re-run of RECI lands at 70.7% -- inside the original authors' reported error bar, not the 77.5% often quoted (which we trace to a mis-copied cell). SLOPE's published 82.4% is a decided-subset figure: scoring the authors' own stored output only on the pairs its significance test chose to answer reproduces 81.7%. Under the common ruler the methods cluster in the low-to-mid 70s and the zero-parameter compressor ties the strongest of them. We document the mechanisms that inflate published figures (test-set model selection, significance-gated abstention) and contribute two further results: compression score magnitude is a model-free confounding flag (p = 2.8e-68), and a pre-registered falsification test fails in an instructive way that bounds the method's theoretical interpretation. Code, pre-registrations, and per-pair outputs are released.

The detection and atmospheric characterization of exoplanets have entered a new data-intensive era driven by the James Webb Space Telescope and the upcoming Ariel mission. Modern surveys produce millions of light curves and high-resolution spectra that overwhelm traditional pipelines, motivating the rapid integration of Machine Learning and Deep Learning methods into the exoplanet workflow. This review synthesizes the latest progress in applying ML/DL techniques to exoplanet detection (transit identification, candidate vetting, false-positive rejection) and atmospheric characterization (retrieval, detrending, cross-correlation, surrogate modelling) in the context of JWST and Ariel. We start with classical algorithms such as Random Forests and Convolutional Neural Networks, move through Transformers and Recurrent architectures, then survey modern simulation-based inference using Neural Posterior Estimation and Flow Matching Posterior Estimation with normalizing or continuous normalizing flows. We discuss benchmark efforts, including the Ariel Machine Learning Data Challenges (2019 to 2025) hosted with NeurIPS, and key JWST case studies such as the WASP-39b Early Release Science programme. Results indicate that DL approaches consistently match or exceed traditional pipelines in both speed and accuracy, while ML-driven retrievals reduce inference time from CPU-hours to seconds and can accelerate nested-sampling retrievals by factors of 3-8 without compromising Bayesian evidence. We identify outstanding challenges interpretability, calibration of uncertainties under noisy data, hybrid modelling, and the generalization of models across instruments and planet populations and outline a research roadmap spanning the JWST era and beyond into Ariel's launch in 2029.

Accurate dynamics models are critical for informed decision-making in robotic systems, particularly for agile aerial vehicles operating under uncertainty. Neural network dynamics models are attractive for capturing complex nonlinear effects, but existing predictive approaches struggle with long-horizon forecasting because their autoregressive rollout mechanism amplifies errors over time. Joint Embedding Predictive Architectures (JEPAs) offer a compelling alternative by modeling dynamics in latent space, yet prior JEPA-style methods for robot navigation have been studied primarily for kinematic-level planning, with limited investigation in high-frequency control. In this work, we introduce the JEPA-style model for real-time quadrotor control. The proposed approach combines a latent dynamics model with a novel physics-inspired prober that maps frozen latents to interpretable state, enabling physically grounded long-horizon prediction. Additionally, we combine the learned model with a sampling-based optimal control solution to take advantage of its predictive capabilities for real-time control on embedded hardware. Finally, to reduce the dependence on expensive and unsafe real-world data collection, we develop a structured pipeline for automated dataset generation. Extensive open-loop and outdoor closed-loop experiments demonstrate accurate prediction, robust zero-shot sim-to-real transfer, and strong generalization across diverse operating conditions.

Fei Xiaotong's Differential Order Pattern characterizes rural society as egocentric and relationally graded, with cooperation attenuating over social distance. Although often treated as culturally specific, its mechanistic basis remains under-operationalized, and prior LLM-based simulations have mainly addressed short-term coordination rather than long-horizon social structure. We propose CAREB-MAS, a multi-agent framework grounded in Affect Control Theory, Social Identity Theory, and Durkheimian collective affect. Agents reason through an emotion-ethics-belief chain and maintain dynamically evolving egocentric identities, while the macro environment specifies only individual production, preference-based allocation, and minimal interaction protocols. Across long-horizon simulations, agents spontaneously reproduce five core Differential Order phenomena: stable labor specialization, guanxi-based economic ethics, relational decay of cooperation, emergent relational authority, and clan-based center-periphery stratification. These patterns shift with production structure from kin-centered integration toward greater functional interdependence. Extensive experiment results support interpreting Differential Order as a structure-sensitive emergent outcome of general social mechanisms, with LLM-based multi-agent simulation providing an interdisciplinary framework for studying social structure and change.

Recent work typically assesses vision--language consistency using attention distributions of answer-side tokens. However, we observe that highest attention regions are not always consistent with the intended semantic token. This probably stems from decoding drift, where language priors from previously generated answer tokens accumulate and mismatch with visual attention. Besides the priors from previous answer tokens, we find that structural tokens, e.g., modality boundary markers, may encompass the entire context and generate high attention to areas unrelated to the target. To avoid these distortions and provide consistency evaluation for large VLMs, we adopt prompt-side semantics and propose Prompt-Vision Token Activation Map (PV-TAM). PV-TAM further incorporates a filter to remove systematic bias induced by modality boundary markers. Unlike traditional methods that evaluate overlap solely through masks while ignoring activation intensity, our metrics leverage the peak distribution of attention to measure the alignment between prompts and visual regions. In experiments, PV-TAM consistently improves both attention-based and IoU-style localization metrics over answer-side baselines on various datasets.

Logical reasoning is essential for reliable AI, yet existing benchmarks are largely first-order-logic-centric, focusing on object-level deduction over fixed predicates. This misses many realistic scenarios where models must reason over rules, predicates, functions, constraints, and decision procedures themselves. We introduce HOLMES (Higher-Order Logic Meets real-world Explainable Symbolic reasoning), the first real-world benchmark for higher-order symbolic reasoning in LLMs, containing 1379 instances. Built on higher-order logic, HOLMES pairs natural-language problems with HOL formalizations, ground-truth answers, verifiable reasoning traces, and fine-grained controllable reasoning factors across law and finance. Experiments show that current LLMs still struggle on HOLMES, with an average accuracy of only 50.64% and the best model reaching 59.54%. Our analyses further reveal that high final-answer accuracy can mask shortcut reasoning in conflict-resolution settings, while performance drops sharply under scope-conditioned and compositional reasoning. These findings identify higher-order symbolic reasoning as a key bottleneck for building reliable and verifiable LLMs. The project code and dataset are publicly available at https://github.com/wuyucheng2002/HOLMES.

Spiking neural network (SNN)-based neuromorphic speech enhancement has emerged as a promising paradigm due to its energy efficiency, yet it still underperforms classical artificial neural network (ANN)-based approaches owing to binary activations and the lack of well-designed network architectures. To overcome this limitation, we propose a novel dual-branch spiking neural network architecture equipped with a gated spiking unit (GSU), termed GSU-DBNet. Specifically, GSU-DBNet simultaneously models the speech magnitude spectrum and complex spectrum, predicting the corresponding magnitude and complex spectral masks. Meanwhile, a dual-path GSU module is adopted to exploit temporal and frequency information for enhanced spatiotemporal feature representation. Experiments on a popular benchmark dataset show that GSU-DBNet achieves a PESQ score of 3.04 with only 394K parameters, outperforming existing SNN-based methods while using only 4.5%--10.6% of the parameters of representative ANN-based models.

Recent advances in large language models (LLMs) have enabled vibe coding, an emerging software development paradigm in which users create applications primarily through natural-language interactions with AI agents. Due to its low barrier to entry, vibe coding is rapidly gaining adoption in practice. Unlike conventional AI-assisted programming, where developers remain responsible for implementation and code review, vibe coding delegates a substantial portion of development to AI systems. This shift raises a fundamental question: how (in)secure are applications developed through vibe coding? In this paper, we conduct a systematic study of the security of vibe-coded applications. We collect a large corpus of real-world applications developed using popular AI agents and design a vulnerability analysis framework that combines agent-assisted code auditing with human validation. Using this framework, we examine the prevalence, severity, and root causes of vulnerabilities in the deployed vibe-coded applications. Our study reveals several key findings: (1) vibe-coded applications exhibit recurring vulnerability patterns that differ from those commonly observed in conventional software development workflows, including placeholder logic, unfiltered input, and secret exposure; (2) these vulnerabilities arise from systematic limitations of AI agents throughout the vibe-coding lifecycle, such as memory loss, locally optimized objectives and insufficient security knowledge; and (3) while advances in LLM capabilities and improved prompting strategies can reduce the incidence of vulnerabilities, they do not eliminate the underlying security risks. Overall, our study provides an empirical understanding of the security landscape of vibe-coded applications and lays the groundwork for addressing the security challenges introduced by the growing delegation of software development to AI systems.

Engineering reliable autonomous systems is an important and growing topic in computer science. As autonomous systems become more prevalent, easy-to-use techniques for building them reliably are increasingly important. This workshop report captures and expands on the discussions at the Lorentz Center Workshop "Engineering Reliable Autonomous Systems" (ERAS), held from 10 to 14 June 2024. The workshop was co-organised by the organisers of the Workshop on Formal Methods for Autonomous Systems (FMAS) and the Workshop on Agents and Robots for reliable Engineered Autonomy (AREA). It brought together members of the FMAS and AREA communities, industry practitioners, and representatives from sectors where autonomous systems pose distinctive engineering challenges. The workshop focused on three main research topics: techniques for verification and validation of autonomous systems; engineering real-world autonomous systems; and software architectures for safe autonomous systems. Its main outcome is a catalogue of challenges in these areas and, most importantly, a pathway to solutions. Some challenges can already be tackled by techniques that are well known in academia but have not yet become regularly used in practice. Other challenges remain unresolved and require further research. This roadmap is intended to support future research and industrial collaboration.

Verilog debugging remains one of the most time-consuming stages in digital circuit design. Recent advances in Large Language Models (LLMs) have enabled automated debugging; however, most existing approaches rely solely on test outputs and compiler feedback in an end-to-end manner, limiting their effectiveness on complex bugs. A key challenge is that the root cause of an error may be far removed from its observable outputs, making it difficult for LLMs to trace long dependency chains in code. This challenge is further exacerbated in large codebases, where long context lengths hinder efficient reasoning. To address these limitations, we propose VeriPilot, an LLM-powered debugging framework that leverages golden reference models to enable fine-grained bug localization and repair. VeriPilot goes beyond output-level comparison by aligning internal variable semantics between the Verilog design and its corresponding golden model through LLM-based analysis. It then performs step-by-step signal tracing using Control-Data-Flow Graphs (CDFGs) derived from static analysis, identifying a minimal set of suspicious code regions along with their correct counterparts from the golden model. These structured insights are subsequently provided to the LLM to guide reasoning and automated code repair. Experimental results on the Comprehensive Verilog Design Problems (CVDP) benchmark from NVIDIA demonstrate that VeriPilot improves the repair success rate of GPT-4o from 54.3\% to 85.71\%, significantly enhancing both bug localization accuracy and repair effectiveness for complex Verilog designs. The source code and benchmark are publicly available at Github https://github.com/YihanWn/VeriPilot.git.

Domain generalization learns from multiple source domains to generalize to unseen target domains. However, it often neglects the realistic case of label mismatch between source and target. Open set domain generalization is then proposed to recognize unseen classes in unseen domains. A simple approach trains one-vs-all classifiers to separate each class and detect outliers as unknown. Yet, the imbalance between few positive samples and many negative samples skews the decision boundary towards the positive ones, leading the model to over-reject out-of-distribution data, even from known classes in unseen domains. In this paper, we propose a novel meta-learning stategy called dualistic MEta-learning with joint DomaIn-Class matching (MEDIC), which considers implicit gradient matching towards inter-domain and inter-class task splits simultaneously to find optimal boundaries balanced for both domains and classes. Experimental results show that MEDIC not only outperforms prior methods in open set scenarios, but also maintains competitive close set generalization ability.

Extracting interpretable governing equations from sparse, noisy chemical time-series data remains difficult because discrete reaction topology and continuous kinetic parameters are tightly coupled. We present PC-MCMC-CIGP, a reproducible gray-box workflow that combines spike-and-slab topology sampling, hard conservation and thermodynamic screening, and a Chemical-Informed Gaussian Process (CIGP) residual model for parameter calibration and experimental design. The methodological contribution is not a new MCMC or GP family in isolation; rather, it is the integration of these components into a physically constrained workflow with explicit uncertainty-aware acquisition choices. On the H2 + Br2 benchmark, the constrained sampler distinguishes elementary radical pathways from deceptive phenomenological fits in our experiments. On styrene epoxidation, the CIGP optimization loop improves final yield by 12.5% over the reported GP-BO baseline. A new 10-seed acquisition study shows that EI, GWU, PC-EI, uncertainty sampling, discrepancy hunting, and random search have different trade-offs: PC-EI substantially reduces low-yield BO suggestions, while EI-style criteria give the strongest final-yield performance.

Finance Agent v2 (by Vals AI) has emerged as the reference benchmark for evaluating both Anthropic Claude and OpenAI ChatGPT frontier language models on financial tasks. However, it narrowly deals with periodic reporting from publicly traded companies (SEC 10-K and 10-Q filings), and its agentic harness relies on naive, unenriched chunk retrieval. Neither the task design nor the retrieval approach addresses the distinct challenges of IPO due diligence. SEC S-1 filings combine historical financial statements, governance structures, pro forma and common-control accounting treatments, capital-formation narratives, and underwriting-sensitive risk disclosures within substantially longer documents than typical periodic filings. That is why we introduce IPO Finance Agent, which extends the Finance Agent v2 framework along two directions: task domain and retrieval architecture. During our experiments, the original Finance Agent v2 harness basically failed to deliver any output related to the SpaceX S-1 filing, due to document length. We therefore had to improve the agentic harness with contextual retrieval, a more realistic and industry-standard approach for long documents. We also built a dataset of 1,000 IPO-diligence questions, and publicly release 70 questions on the SpaceX (SPCX) S-1 filing to support reproducibility, while the remainder are held private to guard against benchmark contamination. In addition, we introduce an evaluator-optimizer pipeline to automatically generate evaluation rubrics for the benchmark: candidate facts are extracted from model answers, consolidated into draft criteria, then automatically audited for omissions, hallucinations, mistiered items, and redundancy, with LLM feedback driving iterative repair, targeted enrichment, and deduplication. Human experts only review final rubrics before deployment. Results show that the best-performing evaluated model, Alibaba Qwen 3.7 Max, reaches 79.4% accuracy at 0.30 USD per query, and the most cost-efficient model on the resulting Pareto frontier, Xiaomi MiMo-2.5 Pro, reaches slightly lower accuracy (76.8%) at 0.05 USD per query. Both exceed the current Finance Agent v2 leaderboard ceiling-Google Gemini 3.5 Flash at 57.9% for 2.51 USD per querywhile undercutting even FABv2's cheapest entry (MiniMax M3: 48.3% at 0.32 USD) on cost-efficiency. Code and data are released on GitHub: https://github.com/benstaf/ipoagent

Efficient sampling of the Boltzmann distribution in frustrated spin glasses is central to statistical mechanics and combinatorial optimization. Despite advances in machine-learning-based approaches, two issues persist: limited understanding of why variational models fail to benefit from increased scale, unlike the monotonic scaling law of large language models; and high computational cost on large systems that negates advantages over classical sampling methods. Here, we develop a physics-inspired transformer with interpretable sparse attention and spin-tailored positional embeddings to address these challenges. By further leveraging FlashAttention for parallel ancestral sampling, it achieves up to two orders of magnitude speedup over vanilla variational autoregressive networks, enabling neural-network simulations of spin-glass systems to unprecedented sizes on a single GPU. It can resolve full probability distributions, free energies, and overlap statistics across temperatures, for Sherrington-Kirkpatrick and 2D or 3D Edwards-Anderson models, where existing machine-learning methods encounter limitations at certain temperatures. This framework thus establishes a scalable paradigm for frustrated spin-glass systems.