# Research Papers — 2026-09 · bundle page 214 of 266 > Full-text markdown of papers #10651–10700 of 13298 published in 2026-09, newest first. AI-readable mirrors of the full 13298-paper 2026-09 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-09/page/214 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/213.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/215.md) ### A Quantitative Evaluation Framework for Temporal Explainability in Echocardiographic Video Segmentation Authors: Jiyoo Noh, Jonathan H. Chan Published: 2026-09-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.08043v1 Deep learning has achieved state-of-the-art performance in echocardiographic video segmentation, with an increasing number of models incorporating temporal information. However, quantitative evaluation of temporal explainability remains largely unexplored. We propose a quantitative framework for evaluating Grad-CAM explanations using four complementary metrics measuring temporal consistency, saliency motion, anatomical overlap, and temporal overlap. Using EchoNet-Dynamic, we compare a baseline 2D U-Net with ConvLSTM U-Net models trained across multiple temporal strides. While segmentation performance remained comparable across all models, intermediate ConvLSTM explanations exhibited substantially lower saliency consistency and greater centroid motion than final prediction explanations. Temporal Bottleneck explanations were significantly more stable than Encoder Bottleneck explanations across all strides, while final ConvLSTM Decoder3 explanations were broadly comparable to those of the 2D U-Net. Importantly, conventional frame-wise explanation metrics cannot determine whether variation in intermediate explanations reflects meaningful temporal feature evolution or explanation instability. These findings establish a preliminary quantitative framework for temporal explainability and motivate temporal-aware XAI methods that explicitly account for evolving representations in medical video models. --- ### MamMA: A Mamba-Based Pedestrian Trajectory Prediction Algorithm Considering Occupancy Map and Pedestrian Awareness States Authors: Juncen Long, Xiaofeng Jin, Gianluca Bardaro, Simone Mentasti, Matteo Matteucci Published: 2026-09-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.08041v1 Many pedestrian trajectory prediction algorithms have been proposed to improve the safety of navigation for mobile robots working in human-robot coexistence environments. Some pedestrian trajectory prediction algorithms extract information about obstacles near pedestrians from top-down view images to improve the accuracy of trajectory prediction. However, mobile robots typically create local occupancy maps using LiDAR, rather than top-down view images. Meanwhile, the vision sensors on board robots provide egocentric view images, which contain fine-grained behavioral information about the pedestrians near the robot. To better use the information collected by LiDAR and on-board vision sensors, we propose MamMA, a Mamba-based pedestrian trajectory prediction algorithm considering occupancy maps and pedestrian awareness states. MamMA divides the occupancy map by patches and extracts obstacle features from each patch to create map features. Pedestrian awareness states are divided and considered, as some studies show that awareness states affect the perception and speed of pedestrians. Furthermore, a Mamba-based model is proposed to predict the future trajectories of pedestrians based on different types of features. Experiments on the STCrowd, SiT, JRDB, ETH, and UCY datasets show that MamMA achieves better average displacement error and final displacement error than the state-of-the-art algorithms. --- ### VEX-Bench: Benchmarking LLM Agents for Assessing Exploitability of Software Supply Chain Vulnerabilities Authors: Jiahao Shi, Edward Tsien, Yifeng Di, Hongjiao Zhang, Yuan Tang, Ronit Dey, Ilona Shishov, Gal Netanel, Zvi Grinberg, Vladimir Belousov, Bat-Zion Rotman, Ilan Pinto et al. Published: 2026-09-07 Categories: cs.CR, cs.LG, cs.SE arXiv: https://arxiv.org/abs/2609.08040v1 The software supply chain has become an increasingly exposed attack surface because of its reliance on intricate yet fragile dependencies. Existing defenses such as GitHub Dependabot often raise many false alerts because their coarse-grained matching cannot determine whether a vulnerable dependency is actually exploitable. Security analysts typically spend substantial time assessing vulnerability exploitability case by case. Recent LLM agents have emerged as promising candidates for this task given their advanced capabilities in coding and cybersecurity, yet no existing benchmark evaluates them on it. Prior benchmarks target zero-day settings, where agents detect and exploit previously unknown vulnerabilities. In contrast, software supply chain security focuses on how known vulnerabilities in upstream dependencies affect downstream projects. This requires agents to reason across repositories and determine whether an upstream vulnerability is exploitable in the downstream project. To address this gap, we introduce VEX-Bench, the first benchmark for evaluating LLM agents' ability to assess the exploitability of software supply chain vulnerabilities. It contains 75 real-world cases mined from GitHub and labeled by security experts, covering Python, Java, and Go. We evaluate nine models across three agent harnesses. While GPT-5.5 and Claude Opus 4.6 reach approximately 80% F1 on binary vulnerability-status classification, only GPT-5.5 surpasses 70% macro-F1 on fine-grained justification classification. This gap highlights the challenge of moving beyond binary exploitability assessment to identifying fine-grained exploitability reasons. Code and data: https://github.com/steven1518/vex-bench --- ### SAFER-Activities: A Dataset for Smart Assessment of Fall Events and Routine Activities Authors: Diwas Lamsal, Pramod Wickramatilake, Jednipat Moonrinta, Mongkol Ekpanyapong, Matthew N. Dailey Published: 2026-09-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.08038v2 Smart healthcare monitoring systems require precise action recognition to ensure well-being and timely intervention in critical situations such as falls, particularly for mobility-challenged individuals. Existing datasets are often clip-based, lacking the frame-level detail needed to recognize actions online, as they unfold. To address this, we introduce SAFER-Activities, a dataset for fall detection and physical activity monitoring, with a dedicated subset for wheelchair use scenarios. It comprises over 66 hours of video data captured by multiple cameras, with 85,310 action instances and frame-level annotations for 30 action classes. We benchmark action recognition on SAFER-Activities with 2D and 3D skeleton models, RGB models with frozen backbones, and multimodal fusion strategies, and evaluate on in-lab, out-of-distribution, and cross-dataset test sets. Skeleton-based models generalize best under domain shift; fusing frozen RGB features with the skeleton stream improves in-domain recognition over the baseline CNN1D, most clearly on the wheelchair subset, but degrades out of distribution. Cross-dataset and qualitative evaluations confirm that models trained on SAFER-Activities transfer well to unseen environments and external fall data. To support research on robust fall detection and activity monitoring, we release the dataset and code at https://safer-activities.github.io/. --- ### Two-Scale Localized PCA-Net: Coarse-Global and Local-Residual Representations for Artifact-Reduced PDE Operator Learning Authors: Mrigank Dhingra, Jordan Stout, Omer San Published: 2026-09-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.08034v1 Localized dimensionality reduction improves the scalability of operator learning for high-dimensional partial differential equations (PDEs), but independently decoded local patches can introduce block offsets, interface mismatches, and spurious high-wavenumber content. We introduce Two-Scale Localized PCA-Net, which decomposes the solution into a coarse-global component and local residual corrections. A compact global PCA basis captures domain-scale structure, while nonoverlapping local PCA bases represent the remaining fine-scale residual. A block-balanced latent objective couples the two representations, and optional interface-aware fine-tuning further promotes continuity through reconstruction and trace losses. On Poisson benchmarks, the two-scale representation substantially reduces reconstruction error and visible block artifacts relative to plain and overlap-based localized PCA-Net while approximately halving PCA fitting cost relative to overlap. On heterogeneous Darcy flow, it strongly reduces interface and discrete-residual errors, with more modest reconstruction gains. Ablations show that the primary improvement arises from the two-scale output representation, while interface-aware fine-tuning provides complementary continuity refinement. Overall, separating globally coherent structure from localized residual detail provides an efficient representation for artifact-reduced PDE operator learning. --- ### Flexible Motion Generation from Language and Style References Authors: Kai Weixian Lan, Bodie Criswell, Briana Fedkiw, Zhan Zhang, Joseph Teran, Daniel Holden Published: 2026-09-07 Categories: cs.CV, cs.GR, cs.LG arXiv: https://arxiv.org/abs/2609.08032v1 We introduce FlexMoGen, a novel framework for flexible human motion synthesis conditioned on both natural language descriptions and motion style references. Text prompts are effective at defining semantic content, but they are often limited in capturing fine-grained style details such as timing, limb articulation, and expressive dynamics. A style example clip supplements the text by conveying these nuanced motion characteristics directly, enabling the model to preserve high-level intent while reproducing the desired stylistic traits. Given a text prompt and a style example clip, FlexMoGen generates high-quality motions that preserve semantic content while faithfully reflecting the target style, offering users greater control over the animation generation process. Unlike prior methods that rely on discrete style labels and do not generalize to long or multi-style generation, FlexMoGen learns a variational style encoder without style supervision and supports long, time-varying, multi-style synthesis. Our framework jointly pre-trains the style encoder and a text-to-motion latent diffusion model within a unified architecture, modulating motion style through a lightweight adaptation module. It integrates an efficient relative positional encoding scheme and is trained on both stylized and non-stylized datasets, enabling strong generalization to unseen text-style combinations. Experiments show that FlexMoGen achieves the best balance between content fidelity and style reflection. --- ### BanglaMemeX: Advancing Cultural Metaphoric Image Interpretation in Bangla with a Multimodal Explainable Dataset Authors: Md. Sadman Sakib, Zisan Mahmud, Md. Fahim Arefin, Md Fahim Published: 2026-09-07 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.08029v1 Vision Language Models have achieved strong performance on multimodal benchmarks, yet their ability to reason about culturally grounded and metaphor-rich content remains insufficiently studied. Internet memes present a challenging setting where meaning emerges from implicit interactions between image, overlaid text, sarcasm, and shared socio-cultural knowledge rather than literal visual recognition. This challenge is amplified in low-resource languages such as Bangla, where code-mixing, stylized scripts, and culturally specific symbolism introduce substantial distribution shift. In this work, we introduce BanglaMemeX, a culturally grounded multimodal benchmark comprising 3,000 Bangla memes annotated with multi-dimensional labels (humor, sarcasm, offensiveness, motivational intent, and overall sentiment) and human-written explanations that explicitly describe textual and visual metaphors. We systematically evaluate modern VLMs on both classification and explanation generation, revealing that current models struggle to interpret implicit cultural cues despite reasonable surface-level accuracy. Our results highlight the need for culturally-aware multimodal systems capable of grounded reasoning under linguistic and cultural distribution shift. --- ### Eliciting Self-Verification in Multimodal Reasoning Agents with Reinforcement Learning Authors: Vishwas Sathish, Viresh Ranjan, Xinliang Zhu, Arnab Dhua, Douglas Gray Published: 2026-09-07 Categories: cs.AI, cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.08025v1 Reasoning agents increasingly rely on external tools such as web search to answer complex queries. Reinforcement learning (RL) finetuning algorithms such as GRPO have improved long-form reasoning in text-only language models, particularly for coding and mathematics. Reliable tool use in multimodal agents, however, remains challenging because models must interpret text and images while integrating noisy retrieved evidence, often under sparse outcome-level supervision without explicit verification signals. We present Self-Verification via Reinforcement Learning (SVRL), an RL-only finetuning framework that trains multimodal agents to verify and filter retrieved evidence within their own reasoning traces, reducing reliance on external verifiers at inference time. SVRL also introduces a search-aware penalty that discourages unnecessary tool calls and a query-diversity reward that encourages diverse, well-formed search queries, providing fine-grained feedback on when and what to search. Finetuning Qwen-2.5-VL-7B with SVRL on only 5{,}000 visual question answering examples yields consistent gains in multi-hop VQA generalization and tool efficiency across benchmarks. Overall, SVRL narrows the gap between compact agents and much larger proprietary models while requiring substantially lower training and inference cost. --- ### What Does Multi-Agent LLM Debate Actually Change? A Layered Analysis of Disagreement and Answer Quality Authors: Chen Qian Published: 2026-09-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.08016v2 Multi-agent debate, in which several LLMs exchange arguments before producing an answer, is widely assumed to improve answer quality by surfacing genuine disagreement. That disagreement is hard to verify, and no single signal can settle it, so we organize the analysis around four questions: (A) does the debater say it disagrees; (B) does its reply text actually argue; (C) does the dissent survive once the tone instruction that produced it is removed; and (D) do the probabilities assigned to stance options change? We evaluate three-model committees on 50 open-ended GlobalOpinionQA questions under three debate tones, friendly (seek common ground), neutral, and hostile (stress-test every position). The answers differ. (A) Full agreement differs by 50.4 percentage points between the friendly and hostile endpoints, pooling replies across all three rounds. (B) The text argues too: an LLM evaluator reading the contribution and reply without the structured self-report or tone instruction confirms the pushback is real. (C) Removing the instruction produces more returns to agreement in our sample, but the primary question-level test is inconclusive. (D) In a separate open-weight study, adjusted movement toward the opposing side is detected in two of three models, relative to a filler-based reference. This measures contextual response probabilities, not lasting belief change. For final answers, debate brings no measurable quality gain: an evaluator that judges each pair in both answer orders scores the debated answer no better than the same committee's no-debate answer on all 299 pairs, a test that detects severe but misses moderate damage in our checks. Taken together, LLM debate readily changes what agents say, but we find much weaker evidence that it changes what they persistently endorse or improves the quality of the final answer. --- ### From Version Conflicts to Decision Conflicts: Selective Revalidation for Long-Running AI Agents Authors: Yongjian Lyu, Yang Ren, Ruofei Lai, Wenting Liu Published: 2026-09-07 Categories: cs.AI, cs.DB arXiv: https://arxiv.org/abs/2609.08015v1 Long-running AI agents may read state, reason, wait for tools or human approval, and perform an external action much later. The state that justified the action can change in the meantime. For example, after an agent proposes an 80 GBP refund under a limit of 100, a customer-name change affects only presentation metadata, a new limit of 90 still permits the refund, a limit of 50 invalidates it, and a refund issued by another worker must prevent a duplicate. Standard optimistic concurrency control and version checks can detect that previously read state has changed, but by themselves do not determine whether that change invalidates the pending action's justification. We call any detected version change a version conflict; when that change invalidates the action's justification, it is also a decision conflict. ATR records the explicit, executable conditions that justify a pending action and rechecks only the conditions affected by a change before releasing the external operation. It can retain the action, refresh non-decisive metadata, require replanning, or block execution; a target-side transaction or compare-and-set binds checked state to commit. Across 210,000 controlled executions over 15 mutation cases, ATR matched every developer-specified outcome with no false allows or blocks. In ten durable SQLite checkpoint/resume cells, it evaluated 0.6 conditions per change versus 6.0 for FullScan. At 4,093 recorded reads, ATR took 9.3 microseconds versus 2595.9 microseconds for FullScan. These deterministic results establish controlled feasibility, not production generality or automatic extraction of the required conditions. --- ### A Black-Box Adversarial Attack on Human Pose Estimation and Keypoint-Based Action Recognition Models Authors: Kacper Mroczek, Michal Kepski Published: 2026-09-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.08013v1 Human pose estimation and keypoint-based action recognition models are increasingly deployed as components of video understanding pipelines, yet their vulnerability to adversarial attacks remains insufficiently studied. Temporally coherent black-box attacks have been previously studied in visual object tracking, where the attack feedback can be defined using bounding-box overlap measures such as Intersection over Union (IoU). However, human pose estimation produces keypoint configurations rather than enclosing boxes, making box-level similarity poorly suited for measuring pose degradation. We propose OKS Attack, a decision-based black-box attack that uses Object Keypoint Similarity (OKS) as the attack feedback signal, directly targeting the spatial structure of human poses rather than their enclosing boxes. Experiments on the Penn Action dataset show that OKS Attack consistently reduces pose quality across evaluated pose estimators, with mean OKS decreases ranging from 0.0802 to 0.1494. In a downstream cross-dataset action-recognition evaluation, the attack reduces accuracy by 6.18 to 13.86 percentage points and outperforms query-matched random-noise perturbations. The attack is effective across both top-down and single-stage pose estimation models. The source code will be made publicly available at https://github.com/KacperM33/OKS_attack --- ### TaskGuard: Task-Conditioned Restoration Utility for Risk-Aware Object Detection Authors: Vung Pham Published: 2026-09-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.08011v1 Image restoration is commonly applied before object detection under adverse conditions, yet a visually improved image need not improve the downstream task. We study this mismatch as restoration utility prediction: given a degraded image and its candidate restoration, should the restoration be used or should the original observation be preserved? We introduce TaskGuard, a post-hoc controller for frozen restoration and detection pipelines. TaskGuard characterizes the realized restoration residual through its interaction with detector sensitivity and predicts whether the intervention is task-beneficial. Exact regional counterfactuals reveal substantial within-image utility heterogeneity, while a deployable pseudo-gradient preserves statistically reliable directional information. Feature-group ablation further shows that task-conditioned evidence contributes information beyond detector-response and residual statistics. The TaskGuard utility predictor is trained only on Gaussian degradation and frozen before final evaluation, then transferred to unseen motion blur, rain, and defocus. Across these unseen families, TaskGuard reduces lossnegative interventions by 54.2% (family macro) and practical per-image detection deteriorations by 37.0% (pooled), while preserving 98.8% of the Always-Restore COCO AP. On natural-rain DAWN, it reduces loss-negative interventions by 97.9% while retaining 77.8% of the AP improvement obtained by deraining. These results support restoration utility as a task-conditioned property of the specific intervention rather than image appearance alone. --- ### Sparks of In Silico Cognitive Science: Theories from Simulated Data Can Generalize to Humans Authors: Akshay K. Jagadish, Younes Strittmatter, Nori Jacoby, Eric Schulz, Nathaniel Daw, Thomas L. Griffiths, Suyog H. Chandramouli Published: 2026-09-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.08003v1 Behavioral foundation models have been proposed as stand-ins for human participants across settings, but it is unclear whether theories discovered on them generalize to humans or merely characterize the simulator. We ran the Automated Cognitive Scientist (\textsc{AutoCog}), a closed-loop discovery system in which LLM agents design theory-discriminating experiments, collect responses, arbitrate between competing theories, and synthesize successors, entirely on behavior simulated by Centaur, a foundation model of human behavior. In a multi-attribute decision-making setting, the theories \textsc{AutoCog} found on Centaur generalized to human data: they outperformed canonical theories on ten held-out experiments and were rivaled only by theories found by running the same loop on people. We argue that this succeeds despite the simulator's inevitable imperfections because a discovery loop that arbitrates between competing theories demands less of its simulator than estimation does. The simulator only needs to capture the regularities that distinguish the theories, and not necessarily reproduce behavior precisely. Imperfect simulators can therefore widen the search over theories, with human data then testing whether the surfaced theories generalize. --- ### Mini-Batch Risk-Averse Deep Q-Learning: A Robot Navigation Case Study Authors: Aayush Patel, Andrzej Ruszczyński Published: 2026-09-07 Categories: cs.AI, math.OC arXiv: https://arxiv.org/abs/2609.07998v1 We study the control of Markov decision processes in which the quality of a policy is evaluated by a dynamic, time-consistent Markov risk measure rather than by an expected discounted cost. The main obstacle to combining such measures with reinforcement learning is that a transition risk mapping depends on the transition kernel in a nonlinear way, and therefore cannot be estimated from a single observed transition. We remove this obstacle by employing mini-batch transition risk mappings: the mapping is applied to the empirical measure of $N$ independent next-state samples, and the result is averaged. The resulting mapping is again coherent. However, as an expected value of a function of $N$ next-state values, it admits an unbiased one-sample estimator. We embed this mapping into a double deep Q-network, analyze the two sources of estimation bias that arise, and obtain a risk-averse Q-learning method applicable to state spaces far beyond the reach of tabular schemes. The method is applied to an underwater robot navigation problem, in which a vehicle must visit collection points, gather stochastic information payloads, and deliver them at transmission points, while exposed at each step to the risk of destruction. A hierarchical decomposition delegates path execution to an exact graph search and confines learning to the high-level ``collect or transmit'' decision. A low-dimensional feature map, invariant under the symmetries of the problem, replaces the raw state--configuration encoding. In experiments on $300$ held-out environments, the resulting policies transfer to instance sizes never seen in training, and already $N=2$ reduces the upper semideviation of the outcome distribution while simultaneously improving its mean whenever the simulator is misspecified---an empirical counterpart of the duality between coherent risk measures and distributional robustness. --- ### Sharp Structure-Agnostic Minimax Risk for Partial Linear Models Authors: Haichen Hu, David Simchi-Levi Published: 2026-09-07 Categories: cs.LG, math.ST arXiv: https://arxiv.org/abs/2609.07997v1 We characterize the sharp structure-agnostic minimax risk for coefficient estimation in the partial linear model when the outcome and treatment nuisances are learned by two distinct black-box learners, which resolves the open problem in double machine learning posed by Gu (2025). For each nuisance \(q\in\{μ,π\}\), we characterize the available learner by an approximation-error budget \(a_q\) and a stochastic-error budget \(s_q\), with the latter controlled through localized Rademacher complexity. Writing \(\mathcal E_n\) for the minimax mean-squared error, we show that \(\mathcal E_n\asymp1\wedge\left\{\frac1n+\left(a_μa_π+\min\left\{a_πs_μ+s_π^2,\,a_μs_π+s_μ^2\right\}\right)^2\right\}.\) The main new ingredient is a novel lower bound for the general two-learner problem. Our proof constructs four finite-mixture testing experiments using orthogonal code functions. Across these experiments, the hidden perturbations are placed outside both learner classes, outside only the treatment learner class, outside only the outcome learner class, or inside both learner classes. These four configurations capture, respectively, the interaction between the two approximation errors, the two asymmetric interactions between one learner's approximation error and the other learner's learning error, and the joint estimation difficulty of learning both nuisances. Combining the four resulting lower bounds yields the displayed rate, which matches the latest upper bound in Gu (2026). Our result shows that standard double machine learning can overstate the intrinsic difficulty of target estimation and provides a target-specific principle for learner selection: approximation error and stochastic complexity must be jointly balanced across the two nuisance learners rather than optimized separately. --- ### HyCO: A Hybrid Neural Solver for Combinatorial Optimization Authors: Yuheng Li, Di Yang, Haipeng Chen, Yanhai Xiong Published: 2026-09-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.07990v1 Sequential reinforcement learning (RL) solvers and global diffusion model (DM) solvers for neural combinatorial optimization exhibit complementary failure modes under an optimization-regret view. The former enjoys small marginal regret in the early construction stage, but suffers from horizon-wise compounding errors with super-linear regret growth; the latter avoids horizon compounding but incurs linear or sublinear regret w.r.t. the dimension of the remaining unsolved subspace. We propose Hybrid Neural Solver for Combinatorial Optimization (HyCO), a hybrid inference algorithm that constructs a solution prefix with an RL solver and adaptively switches to a conditional DM to complete the remaining decisions. To characterize why such hybridization helps, when to trigger the handover, and how to realize it in practice, we first develop a unified error-scaling theoretical framework and prove that, under explicit error-scaling assumptions, i) the hybrid structure achieves strictly lower expected regret than either backbone alone, and ii) there exists a unique optimal trigger step that minimizes the hybrid regret. We then design a lightweight adaptive trigger that combines policy entropy and RL-DM disagreement to detect trajectory-level signals of the regime shift as a practical proxy, since the optimal trigger step is defined at the expected-regret level and is not directly computable on individual trajectories. Experimental results on diverse benchmarks demonstrate that HyCO achieves consistent improvements over both backbones and support the empirical effectiveness of adaptive triggering. --- ### When Can LLM Digital Twins Reduce Human Measurement? From Behavioral Fidelity to Statistical Substitutability Authors: Steven Wang, Kyle Hunt, Shaojie Tang, Kenneth Joseph Published: 2026-09-07 Categories: cs.AI, stat.AP arXiv: https://arxiv.org/abs/2609.07987v1 LLM-based digital twins promise to reduce repeated human data collection by generating person- specific responses, yet existing evaluations provide little evidence about whether they can reduce human measurement while preserving valid inference. To address this, we introduce statistical substitutability, an inferential criterion that evaluates the extent to which twin predictions can reduce human measurement for a particular estimand while preserving valid inference. We develop a framework, grounded in mixed-subject and prediction-powered inference, that evaluates statistical substitutability along four dimensions: aggregate fidelity, paired respondent-level signal, finite-sample human-label recovery, and stability across populations. Across two empirical evaluations spanning behavioral experiments, multiple models, and alternative respondent representations, we find that digital twins can reproduce average human effects while providing little information about which individuals differ from those averages. Newer models and richer respondent information improve some dimensions of performance but do not reliably translate into human-data savings. Human calibration can reduce aggregate prediction error, yet limited labeled samples often fail to produce stable precision gains. Importantly, these findings demonstrate that behavioral fidelity is neither necessary nor sufficient for statistical substitutability. More broadly, they suggest that AI-generated evidence should be evaluated based on its ability to support valid scientific inference rather than its ability to reproduce human outcomes alone. Digital twins should therefore be judged for confirmatory use by whether they reduce uncertainty about human quantities, not merely by whether they reproduce human means, distributions, or effects. --- ### Automated Chest CT Protocol Selection via Large Language Model Derived Text Embeddings from Imaging Request Text Authors: Zahra Hosseini, Mahan Pouromidi, Farzad Khalvati, Patrik Rogalla Published: 2026-09-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.07986v1 Purpose: Accurate CT protocol selection is critical for diagnostic quality and patient safety, yet the current process is manual, time-consuming, and prone to inconsistencies. Prior Machine Learning methods using keywords or bag-of-words lack contextual understanding and perform poorly on rare protocols. We propose a decision support system using large language model (LLM) features to recommend protocols from free-text clinical indications, capturing clinical nuance and phrasing variation for more consistent, efficient selection. Methods: In this REB-approved retrospective study, 285,123 chest CT imaging requests from a large academic medical center (2017-2024) were split into training (228,099, 80%) and held-out test (57,024, 20%) sets. Each request included procedure names, clinical indication, HIS comments, and the selected protocol. Clinical text was embedded using a fine-tuned LLM, Meta's LLaMA-3.1-70B; these features input a logistic regression classifier predicting 18 protocol labels (e.g., PE, LDCT). Results: The pipeline achieved a weighted precision of 0.84, weighted F1-score of 0.81, and overall accuracy of 79% across 18 CT protocols. On 300 independent cases with expert consensus, the LLM reached an overall accuracy of 80% versus 83% for radiologists, with no significant difference (p = 0.263). Performance was comparable across most classes, with the LLM exceeding radiologists for some challenging categories, and entropy analyses indicated more balanced protocol use, suggesting reduced variability. Conclusion: An LLM-based recommendation system can leverage general knowledge from a large natural-text corpus to accurately assign chest CT protocols from free-text imaging requests, and may serve as a viable foundation for protocol recommendation tools where inputs require language understanding. --- ### From Event Logs to Governed Action: A BlueSky Agenda for Agentic Process Mining Authors: Yiyuan Yang, Zheshun Wu, Yong Chu, Zhenghua Chen, Zenglin Xu, Qingsong Wen Published: 2026-09-07 Categories: cs.AI, cs.CE arXiv: https://arxiv.org/abs/2609.07984v2 Process mining has long turned event logs into process knowledge: discovered models, conformance evidence, bottleneck diagnoses, and runtime predictions. Agentic AI changes the target. Process-aware agents will not only ask what happened. They will ask whether a proposed action should be taken, given the available evidence, privacy budget, organizational authority, and downstream risk. This BlueSky paper proposes event-to-action process mining: a process-mining agenda for transforming heterogeneous operational event data into governed action. The goal is not another dashboard, a generic enterprise simulator, or a language interface over logs. We argue that the community needs four mineable artifacts: event-object representations, action evidence packages, governance contracts, and benchmarks where act, defer, ask, and refuse are all valid outputs. This agenda is timely because agentic business process management (BPM), LLM-assisted process mining, object-centric event standards, causal process monitoring, and privacy-preserving learning are maturing separately. Bringing them together defines a data-mining target inside process mining: mining logged organizational behavior for accountable action, not only retrospective insight. --- ### Solving the Elastic Wave Equation with Physics-Informed Neural Networks: A Robust and Critical Assessment Authors: Davide Staub, Ben Moseley Published: 2026-09-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.07983v1 Physics-Informed Neural Networks (PINNs) have recently emerged as a promising approach for solving Partial Differential Equations (PDEs), offering a meshfree alternative that integrates physical principles into the learning process. This presents a new paradigm compared to traditional discretization methods and purely data-driven machine learning techniques. While promising, PINNs are not a panacea; they inherit challenges such as spectral bias and unstable convergence. Moreover, their potential in seismology remains largely unexplored. In this work, we provide a robust and critical assessment of PINNs for solving the elastic wave equation in seismology. We investigate the performance of PINNs on problems with varying degrees of complexity across various seismic sources and parameter models, from constant to highly heterogeneous settings. A pivotal aspect of our work involves investigating whether embedding physical principles directly into the network architecture enhances convergence and accuracy. We test an extensive range of neural architecture designs, from unrestricted, uninformed PINNs to highly specialized ones. We find that integrating an understanding of wave physics into the network design significantly improves accuracy. For instance, introducing a custom wavelet or plane wave layer, coupled with encoder and decoder layers, consistently yields a relative $L_2$ error approximately half that of the standard PINN, as evidenced across numerous experiments. We further demonstrate that this novel architecture enhances accuracy when applied to the acoustic wave equation, underlying the versatility of our network. Another key contribution of our research is the successful conditioning of PINNs on seismic source locations. This signifies a considerable advancement towards rapid seismic hazard detection and seismic analysis. --- ### Semi-Supervised Learning under Spatially Biased Sampling Authors: Bright Wiredu Nuakoh, Francky Fouedjio, Stephen Bradshaw, Yaw Kwaafo Awuah-Mensah, Wei Hong Tan, Emet Arya, Ebenezer Afrifa-Yamoah Published: 2026-09-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.07982v1 Standard semi-supervised learning (SSL) typically relies on labelled and unlabelled data sharing a common marginal distribution. This assumption is often violated by biased spatial sampling mechanism, when labels are collected under spatially biased or preferential site selection. We treat this marginal mismatch, spatial autocorrelation, and spatial non-stationarity as three distinct mechanisms, varied independently via a labelled-sampling concentration parameter, a spatial length scale, and a non-stationarity strength parameter, and ask how mismatch degrades SSL, whether the cluster and manifold assumptions survive it, and how the resulting failure can be diagnosed. Using a controlled synthetic framework alongside PovertyMap-WILDS, California housing, socio-economic and US air quality monitoring datasets, we systematically vary the degree of mismatch while accounting for spatial autocorrelation and non-stationarity. Through a series of analyses including a segmented-regression changepoint, we show that in the synthetic generator, SSL performance does not degrade gradually but instead exhibits a threshold-like breakdown between approximately 0.71 and 0.77 once distribution mismatch becomes sufficiently severe. We further demonstrate that spatial non-stationarity contributes to performance loss independently of marginal mismatch and that models become increasingly overconfident outside the regions where labels are available. To support practical deployment, we evaluate several distribution-divergence measures as indicators of reliability and introduce a kernel-weighted local divergence metric that provides a more stable estimate of spatial mismatch than a naïve localised approach. These findings provide empirical evidence and diagnostic tools for better documenting the risk of incorporating unlabelled spatial data into semi-supervised learning workflows. --- ### Clean Accuracy Does Not Guarantee Provenance Robustness: A Prospective Codec-Stress Evaluation of Audio Attribution Authors: Gang Shi Published: 2026-09-07 Categories: cs.SD, cs.LG arXiv: https://arxiv.org/abs/2609.07981v1 Audio provenance attribution - which system produced a synthetic utterance - is reported at near-ceiling accuracy on clean benchmarks, yet audio reaching an analyst has usually been transcoded. We report a prospectively registered measurement of closed-set attribution after single-stage codec transport, with the analysis region fixed from fidelity metadata before any attribution model was trained. On two corpora, in-support losses reach 53.5 (43.5, 63.6) and 70.3 (63.0, 77.5) Macro-F1 points for WavLM-Base+, and 61.0 (56.8, 65.1) and 49.8 (41.6, 57.9) for W2V2-BERT 2.0, under simultaneous component-level bands. Degradation is strongly condition- and representation-dependent: within one in-support grid WavLM losses run from -0.4 to +53.5 points, and the two encoders differ beyond a prespecified +/-5-point margin at six of twelve conditions. A clean-qualified ECAPA-TDNN and a Proxy-Anchor head degrade comparably, so the effect is not confined to one representation family or a weak linear head. The registered matched-fidelity comparison was not estimable on this grid, and waveform and perceptual measures order the conditions differently: MP3 at 8 kbit/s ranks mid-grid on SI-SDR but last on PESQ-WB while causing the largest loss. For the tested tasks, corpora, representations and codec grid, a clean accuracy figure does not by itself characterise deployment robustness. --- ### Heat Field Signatures: From Point Clouds to Smooth Geometry Authors: Yuanqing Wang, Yapeng Tian, Baris Coskunuzer Published: 2026-09-07 Categories: cs.LG, math.DG arXiv: https://arxiv.org/abs/2609.07975v1 Bringing multiscale geometric analysis directly to irregular point clouds remains difficult: quantities such as local dimension, anisotropy, density variation, and geometric transitions are typically estimated through explicit neighborhood, manifold, or graph constructions, or left for neural networks to infer from coordinates. We introduce Heat Field Signatures (HFS), which lift a point cloud to a multiscale family of smooth ambient heat fields, providing a direct interface from discrete samples to geometric analysis. From this field, HFS computes closed-form global and local signatures directly from pairwise distances, capturing heat concentration, intrinsic dimension, anisotropy, and scale transitions. We further introduce the Heat Dimension Spectrum (HDS), a compact summary of multiscale geometric composition. HFS can be used as a closed-form descriptor, a lightweight learned representation, or a geometric feature channel for neural point-cloud models. Across synthetic and real-world benchmarks spanning subcellular, neuronal, tree, and protein data, HFS outperforms strong point-cloud and multiparameter-persistence baselines while substantially reducing end-to-end cost. On SCOP protein-fold classification, HFS improves over the strongest deep baseline by nearly $24$ percentage points using coordinates alone, while standalone HFS representations are exactly rotation-invariant by construction. More broadly, HFS turns a classical heat field into a practical interface for multiscale geometric analysis in modern point-cloud learning. --- ### A Sub-4 Approximation for Fair $k$-Means Authors: Kangke Cheng, Guanlin Mo, Shihong Song, Hu Ding Published: 2026-09-07 Categories: cs.CG, cs.LG arXiv: https://arxiv.org/abs/2609.07974v1 Fairness in clustering has attracted sustained research interest, motivated by the need to ensure equitable representation of protected groups in machine learning applications. We study fair $k$-means clustering in Euclidean space, where the proportion of each protected group in every cluster must lie within specified lower and upper bounds. These constraints make it challenging to determine both cluster centers and point assignments. We propose an approximation algorithm that combines a linear programming relaxation with geometric transformations of the input to construct candidate center sets. Given a $ρ$-approximate algorithm for weighted $k$-means and any $ε>0$, our algorithm returns a fractional solution whose cost is at most $1+(3-1/Γ)ρ+O(ε)$ times the optimal integral fair cost, where $Γ\approx6.357$ is an upper bound on the integrality gap of the standard Euclidean $k$-means LP. With a PTAS as the subroutine, the approximation ratio becomes $3.8427+O(ε)$, improving the previous factor of $5+O(ε)$ to below $4$. The solution satisfies all fairness constraints exactly and can be rounded to an integral assignment with a bounded additive violation of fairness and no increase in cost. The same approximation guarantee extends to the $k$-sparse Wasserstein barycenter problem. --- ### MeRoTune: RoPE-Safe Merging with a Tunable Dial Authors: Salman Faroz Published: 2026-09-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.07971v1 When you merge two fine-tuned models from the same base checkpoint by simply averaging their weights, you implicitly assume their attention subspaces are still aligned. Recent work attempts to fix misalignments by learning an invertible correction matrix, $M$, for each model's query and key projections. This correction cancels out---using $M$ on the query side and $M^{-T}$ on the key side---right before the dot product. However, this cancellation is only exact if nothing sits between the projection and the dot product. In reality, almost all modern open-weight language models put a rotary position embedding (RoPE) exactly there. In this paper, we show that this cancellation is exact under RoPE if and only if $M$ commutes with RoPE's per-position rotation. We derive the specific class of matrices where this holds: a scaled rotation acting independently within each RoPE frequency pair. This forms a strict, low-dimensional subset of the unconstrained matrices that current methods normally train. Building on this, we turn this constrained matrix class into a new merging method. While keeping the base weights entirely frozen, two fine-tunes each learn their own RoPE-compliant correction matrices. We optimize these corrections against a chosen blend ratio so the final result can be adjusted post-hoc like a dial, rather than locked into a single fixed merge. Our default approach trains at one fixed blend ratio, similar to how LoRA sets its scaling hyperparameter in advance. We also experiment with resampling the blend ratio randomly at every training step, and we report the results of both approaches. --- ### Reasoning Beyond Transcription: Audio Language Models on Child Stuttering Speech Authors: Chibuzor Okocha, Christan Grant, Zoey Liu Published: 2026-09-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.07968v1 Child speech differs from adult speech in acoustics, prosody, and linguistic structures. Speech disfluencies (such as repetitions) further challenge automatic understanding. While Audio Language Models (ALMs) show strong semantic reasoning from speech audio, their ability to reason about disfluent child speech in mixed-speaker settings remains unexplored. We investigate this through two tasks: child-focused semantic summarization and speech entailment. Experiments use recordings of children who stutter in mixed speaker interviews without explicit speaker separation. Models are instruction-guided to focus on the child, preserve clinically relevant disfluencies, and avoid adult-speech leakage. Evaluation combines LLM-based judges and reference-based metrics, anchored by transcript-oracle baselines to isolate errors. Results show that while ALMs extract high-level meaning from stuttered speech, reasoning degrades significantly with increased --- ### MetaKV: Adaptive KV Cache Compression for Constrained LLM Inference Authors: Michael Wang, Keith Li, Roozbeh Bostandoost Published: 2026-09-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.07966v1 Key--value (KV) cache compression is an effective way to reduce the memory overhead of large language model (LLM) inference, particularly for long-context workloads. However, existing compression methods make different trade-offs among accuracy, inference latency, and peak KV cache memory utilization, making a single fixed configuration unsuitable across different prompts and resource constraints. We introduce MetaKV, an adaptive framework that selects a KV cache compression configuration for each input prompt based on user-specified latency and peak memory budgets. MetaKV uses lightweight prediction models to estimate the end-to-end latency, peak memory, and probability of a correct response for each candidate configuration, and selects the configuration that best satisfies the latency-memory constraints while preserving accuracy. We evaluate MetaKV across ten configurations from three representative KV cache compression methods, KVQuant, H$_2$O, and RocketKV, together with an uncompressed FP16 configuration, on four datasets covering mathematics, science, commonsense reasoning, and reading comprehension. Across a wide range of latency and peak memory constraints, MetaKV consistently outperforms the best static configuration, improving constrained success rate (CSR), the fraction of prompts answered correctly while satisfying both constraints, by approximately 0.07 on average and up to 0.135. These results demonstrate the benefit of adapting KV cache compression to individual prompts and latency-memory constraints. Code is available at https://github.com/MichaelWang0505/MetaKV.git --- ### Rethinking Sign Language Translation: The Impact of Signer Dependence on Model Evaluation Authors: Keren Artiaga, Sabyasachi Kamila, Haithem Afli, Conor Lynch, Mohammed Hasanuzzaman Published: 2026-09-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.07965v1 Sign Language Translation has advanced with deep learning, yet evaluations remain largely signer-dependent, with overlapping signers across train/dev/test. This raises concerns about whether models truly generalise or instead rely on signer-specific regularities. We conduct signer-fold cross-validation on GFSLT-VLP, GASLT, and SignCL, three leading, publicly available, gloss-free SLT models, on CSL-Daily and PHOENIX14T. Under signer-independent evaluation, performance drops sharply: on PHOENIX14T, GFSLT-VLP falls from BLEU-4 21.44 to 3.59 and ROUGE-L 42.49 to 11.89; GASLT from 15.74 to 8.26; and SignCL from 22.74 to 3.66. We also observe that in CSL-Daily many target sentences are performed by multiple signers, so common splits can place identical sentences in both training and test, inflating absolute scores by rewarding recall of recurring sentences rather than genuine generalisation. These findings indicate that signer-dependent evaluation can substantially overestimate SLT capability. We recommend: (1) adopting signer-independent protocols to ensure generalisation to unseen signers; (2) restructuring datasets to include explicit signer-independent, sentence-disjoint splits for consistent benchmarking; and (3) reporting both signer-dependent and signer-independent results together with train-test sentence overlap to improve transparency and comparability. --- ### $α$-Graph: Attention-Infused Normalizing Flow Approach to Tractable Graph Modeling Authors: Thanh-Dat Truong, Sarah Alharbi, Susan Gauch, Xinghui Zhao, Marios Savvides, Khoa Luu Published: 2026-09-07 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2609.07961v1 Graph modeling, a crucial task for representing complex relationships in graph-structured data, has achieved significant success in recent years. However, current graph modeling methods rely on traditional Graph Neural Networks and pre-training approaches to implicitly learn the underlying relational structure of graph data. Thus, these prior methods cannot capture the complex graph structure and correlations among inputs. In this paper, we introduce a novel Attention-based Normalizing Flow-based Approach\footnote{Our implementation and models will be released publicly for research reproducibility.} (ANFA or $α$) that provides an explicit, interpretable, and tractable Graph Modeling ($α$-Graph). In particular, we propose a new Unconditional Graph Normalizing Flow with an Invertible Attention Mechanism to capture the complex relational structure of graph data. To further enhance the expressiveness of the model, we introduce Conditional Graph Normalizing Flow with Learnable Queries that enables efficient modeling of correlations in graph-structured data. We show that our Conditional Graph Normalizing Flows behave similarly to Unconditional Graph Normalizing Flows, enhancing expressiveness while maintaining training stability and efficiency. Our experimental results on three benchmarks will illustrate the effectiveness and the state-of-the-art (SoTA) performance of the proposed $α$-Graph method. --- ### The Role of Uncertainty in Assessing the Fairness of Machine Learning Models Authors: Francesca Panero, Ernst C. Wit, Marco Scutari Published: 2026-09-07 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2609.07959v1 Machine learning models are widely used in clinical applications, social media, law enforcement and critical infrastructure. Verifying whether their outputs are biased against disadvantaged groups or individuals is crucial to ensuring they are fair and allowing their use in such settings. A rigorous risk assessment of possible fairness violations requires quantifying the uncertainty associated with selecting and estimating such models. Yet, this is rarely done in the literature, which focuses on identifying a single model with a suitable trade-off between predictive accuracy and fairness. In this paper, we move beyond point estimation and discuss frequentist and Bayesian approaches to uncertainty quantification for fair machine learning, with practical examples and implications for simulated and real data. --- ### Streaming Hierarchical Inference with Tabular Foundation Models Authors: Vitor Crista, Afonso Lourenço, Diogo Martinho, Goreti Marreiros Published: 2026-09-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.07956v1 Tabular Foundation Models (TFMs) have recently demonstrated strong predictive performance through in-context learning, but their deployment in high-throughput data streams remains challenging due to communication overhead and latency. We propose \textit{HINT}, a hierarchical inference framework that combines edge-based retrieval with cloud-based TFM inference. A graph-based approximate nearest neighbor memory maintained over a sliding window provides local predictions and uncertainty estimates, allowing confident samples to be processed locally while uncertain instances are selectively offloaded, together with their retrieved context, to a cloud-hosted TFM. The framework exposes an offloading threshold and a neighborhood retrieval policy that can be varied to balance predictive performance and communication cost. Experiments show \textit{HINT} consistently identifies favorable trade-offs. --- ### Support Topology and Gradient Mixing in Sinkhorn Layers Authors: Dylan Forde Published: 2026-09-07 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.07954v1 Sparse Sinkhorn layers use a fixed support graph to restrict transport between tokens. How does this graph control gradient propagation through the scaling iterations. We develop a fixed-support calculus showing that each row-column cycle induces a row-stochastic operator on column-potential perturbations modulo constants. Its transpose propagates zero-mass reverse-mode cotangents. The finite-cycle operator uses two distinct half-step transport plans; at a balanced fixed point it reduces to a two-step walk determined by a single plan. We derive the accompanying score and marginal source terms and use Dobrushin contraction and minorization to bound homogeneous and source-driven tail cotangents. Our main result characterizes when support and marginals guarantee one-step contraction uniformly over finite scores: every feasible face of the transportation polytope must have pairwise two-hop column overlap. Otherwise, suitable score directions make the contraction coefficient arbitrarily close to one. We extend this analysis to ordered support schedules and derive certificates for partition heat-bath layers, coordinate sweeps, forced shared mass, and register-augmented supports. These results provide mathematical criteria for support design in differentiable transport layers, with guarantees restricted to the fixed-support quotient-gradient component. --- ### Structured Extrema Errors in Classical Surrogates for Viscous Burgers: A Physics-Consistent Interpretation Authors: Youssef Oubari Published: 2026-09-07 Categories: cs.LG, physics.flu-dyn arXiv: https://arxiv.org/abs/2609.07952v1 We study the local errors of classical machine-learning surrogate models, which approximate the time evolution of the one-dimensional viscous Burgers equation. Four models are compared on the same prediction task, using the spatial grid values directly: radial basis function (RBF) kernel ridge regression (KRR), linear Ridge, ExtraTrees, and Random Forests. Across all four models, the one-step residual, defined here as the true value minus the predicted value at each grid point, forms clear curved branches near predicted maxima and minima. A more detailed analysis of KRR shows that these errors are much more strongly related to the second spatial derivative, which measures local curvature, than to the first spatial derivative. Near a smooth extremum, predicted value and curvature form a local two-branch fold. Under our local curvature-based model of the residual, this fold predicts a leading-order near-parabolic relation between predicted value and residual. This geometric result motivates a direct test of the Burgers advection (transport) and diffusion (smoothing) terms. For KRR and Ridge, regression tests on held-out trajectories, a control that breaks the spatial alignment of the diffusion term, and a spectral test of high-frequency content are consistent with insufficient viscous smoothing at moderate and high viscosity. In this case, the surrogate retains more small-scale structure than the true future state. The same physical explanation is much weaker for the tree models. Finally, a correction that uses only predicted quantities reduces both one-step error and error during recursive rollout, where each prediction is used as the next input. --- ### CausalVerify: End-to-End Verification of Causal Analyses by Language Models Authors: Yonghong Zhang, Ricardo Correia, Isabel M. Parra, Yong Xie Published: 2026-09-07 Categories: cs.AI, cs.CL, econ.EM arXiv: https://arxiv.org/abs/2609.07944v2 Language models increasingly perform empirical analyses end to end, yet existing evaluations assess the written explanation or whether generated code executes, not whether the executed workflow recovers the intended causal estimand. We introduce CausalVerify, an execution-grounded benchmark for end-to-end causal analysis that follows a model from research-context interpretation to estimand recovery. It scores this workflow at four distinct layers: method recognition, design specification, executable implementation, and estimand recovery. It combines 259 real-paper contexts, 100 fixed-seed synthetic scenarios with executable reference estimates, and 23 paper-twin pairs in which a model commits to a design before seeing the data and its executed analysis is scored against a canonical estimator on the same realised dataset. Execution is not correctness. Among 426 model-written workflows that run without error, 15.5% fail verification, and a keyword score of the effect direction stated in the text is a poor proxy for recovery. In a 23-pair, nine-model paired study, replacing a model's committed design with the reference design and its execution conventions raises joint recovery of the point estimate and standard error from 15.0% to 51.5%; yet 48.5% of eligible seeds still fail under the reference design. The direction replicates on six pairs built afterwards under a frozen construction protocol, although on the two newest pairs the gain is confined to models from the family that built the references. Design specification is consequential but not sufficient: a plausible method and runnable code do not guarantee recovery, and even supplying the reference design and its conventions leaves substantial downstream failure. CausalVerify evaluates the executed workflow rather than its surface plausibility, and every reported number is recomputed from frozen artifacts by a single script. --- ### Beliefs and Behavior in Language Models Authors: Alex Smolin, Bryan Wilder Published: 2026-09-07 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.07943v1 There is significant uncertainty about whether abstractions like beliefs or desires usefully describe the behavior of large language models (LLMs). In addition to the inherent scientific interest of this question, these latent quantities are often invoked to explain the behavior of LLMs to users or to define and evaluate harmful behaviors which are relative to intent. Nevertheless, we currently lack a means to systematically test whether concepts like "belief" are well-applied to LLMs, and hence whether they are likely to be fruitful ingredients of attempts to align models with human interests. We propose an approach for empirically studying such questions, asking whether a single latent variable inferred from the LLMs' outputs -- interpreted as a degree of belief -- allows an observer to make interpretable predictions of how the LLMs' will respond to new prompts. We find that highly capable models are usefully described as holding beliefs and that, generally, the predictability of model outputs based on an inferred latent belief tracks overall trends in model capability. Building on these findings, we provide empirical strategies to study how beliefs in LLMs can be measured, the extent to which LLMs comply with instructed decision rules or payoffs, and how beliefs evolve within individual instances of an LLM over the course of reasoning. --- ### ReactVAU: A Slow-Fast Decoupled Framework for Streaming Video Anomaly Understanding Authors: Chia-Hui Chen, Shih-Ying Yeh, Fu-En Yang, Min-Hung Chen, Shang-Hong Lai Published: 2026-09-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.07941v2 In this paper, we propose ReactVAU, a Slow-Fast Decoupled Framework for real-time streaming Video Anomaly Understanding (VAU). Existing VAU methods rely on offline inference with global temporal sampling, which violates causality and prevents deployment in live surveillance streams. Conversely, general streaming video models satisfy causal access but dilute rare transient anomalies during memory compression and often invoke heavyweight MLLMs uniformly over long normal intervals. React VAU addresses this gap with three synergistic components: a lightweight Fast Detection Module based on Spatial Grid Folding (SGF) for continuous anomaly filtering; an Anomaly-Aware Persistent Memory (AAPM) that protects critical visual cues from temporal decay; and a heavyweight Slow Reasoning Module that remains dormant during normal streams and is awakened only by suspicious events for semantic verification and causal description. Extensive experiments on multiple benchmarks demonstrate that ReactVAU operates under strict streaming constraints while simultaneously achieving competitive performance in both anomaly detection and causal reasoning, alongside significantly enhanced computational efficiency by minimizing heavyweight MLLM invocations. Project page is available at https://huiyuiui.github.io/ReactVAU/ --- ### Bottom-up Modeling of Repeated Elements via Single Image Analysis-by-Synthesis Authors: Syrine Kalleli, Alexei A. Efros, Mathieu Aubry Published: 2026-09-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.07939v2 We address the problem of discovering repeated elements from a single image. In contrast to existing approaches that depend on large annotated datasets, curated multi-image collections, or object segmentation masks, we show that a single image can suffice to learn a meaningful object model in a completely bottom-up fashion, without any prior knowledge beyond a coarse scale prior. Our method learns a tunable image-space prototype of the repeated elements through a reconstruction objective, enabling the model to identify and synthesize consistent object instances within the same image. Experiments on 116 real images from the FSC-147 dataset demonstrate that our method successfully learns coherent element models and captures intra-category variation on challenging images. Qualitative results reveal superior reconstructions and interpretable decompositions compared to classical decomposition, joint alignment, and 3D object modeling methods, while maintaining a simple 2D formulation. These results suggest that meaningful object discovery can emerge from single image learning alone. --- ### TDDN: Text-aligned Diffused DINO Network for Puzzle Understanding Authors: Harsha Patnala, Debopriyo Banerjee, Ayush Sunil Munot, Somak Aditya Published: 2026-09-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.07937v1 Structured visual reasoning, such as image puzzles, demands fine-grained visual perception, an ability current Vision Language Models (VLMs) lack. VLMs built on CLIP-based ViT backbones trade fine-grained detail for high-level semantics, and we show this loss propagates downstream. To recover it, we fuse DINOv3 and CleanDIFT representations into a perception encoder (DiffusedDINO) and align it with RoBERTa-L, yielding a text-aligned model TDDN that preserves this perceptual advantage: with frozen backbones and only $\sim$590K alignment pairs, TDDN matches CLIP on image-text retrieval, surpassing it on three of four settings. It does so while more than tripling CLIP's dense-prediction accuracy (ADE20K 5.20 $\to$ 18.11 mIoU, COCO-Stuff 7.35 $\to$ 24.44), despite CLIP's massive training corpus. TDDN leads on segmentation benchmarks among general-purpose contrastive encoders, including SigLIP$\,$2. We further introduce Puzzle Perception, a segmentation and visual question answering dataset that probes fine-grained spatial understanding, on which TDDN doubles CLIP's segmentation accuracy (11.04 $\to$ 22.51 mIoU). --- ### FrogNano: Training a 4B Coding Agent via Online Task Synthesis Authors: Minseon Kim, Zhengyan Shi, Emiliano Penaloza, Christopher Cui, Roger Creus Castanyer, Maryam Hashemzadeh, Isadora White, Jonathan Light, Jeonghye Kim, Matheus Pereira, Darya Moldavskaya, Chinmay Singh et al. Published: 2026-09-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.07925v4 We present FrogNano, a 4B coding agent designed to tackle software engineering (SWE) tasks efficiently and effectively, even under resource-constrained environments. It is post-trained exclusively via RL on around 1,500 SWE environments with synthetic tasks. A key ingredient for improving performance is an online task synthesis pipeline that creates tasks calibrated to the frontier of learnability for the current checkpoint. This report provides evidence that competitive small coding agents can be trained with synthetic tasks alone, without traditional distillation from larger models, and that generating tasks at the learnability frontier of the current agent is important. We report details on the training methodology, evaluations across diverse environments, and in-depth analyses, serving as a foundation for our ongoing exploration of lightweight yet capable coding agents that can run on minimal hardware. --- ### Prevalence calibration as shortcut mitigation Authors: Mohamed Amine Kina, Eike Petersen Published: 2026-09-07 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2609.07922v1 Shortcut learning denotes the widespread situation in which a classifier exploits spurious correlations rather than diagnostic features. Existing mitigation strategies mostly aim to learn shortcut-invariant representations; their empirical success is limited and they cannot be applied to classifiers using frozen foundation model encoders. We propose to reframe shortcut learning as fundamentally a calibration problem: unconstrained learning implicitly calibrates each shortcut group to its training set disease prevalence, rendering the resulting classifier necessarily over-confident in one group and under-confident in the other. Building on this insight, we prevalence-equalize calibration between shortcut groups through two encoder-agnostic methods, an in-processing regularizer and a post-hoc prevalence-equalized recalibration step. Across chest-drain-pneumothorax benchmarks on CheXpert and SIIM-ACR, spanning fine-tuned CNNs and frozen foundation-model backbones, both methods substantially outperform all baselines. Post-hoc recalibration of a standard ERM-trained DenseNet raises misaligned-group AUROC from 0.23 to 0.73, indicating that shortcut reliance degrades the classification head rather than the underlying representation. Besides two new state-of-the-art shortcut mitigation approaches, our findings more fundamentally connect shortcut learning to calibration theory and algorithmic fairness. --- ### AVCG: A Generalized Variational Framework for Counterfactual Generation under Hypothesis Distributions Authors: Jamie Duell, Alejandro Jimenez Rodriguez, Mahault Albarracin Published: 2026-09-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.07917v1 Counterfactual explanations formalize "what-if" scenarios by identifying modifications to an input instance that obtain a desired alternative prediction. Traditionally, whether generated via instance-specific optimization or amortized single pass models, these approaches rely on a single, deterministic point-estimate predictor. However, this ignores predictive uncertainty and hypothesis variability, leading to brittle explanations that frequently become invalid if the underlying model is retrained or updated. To address this fragility, we propose the Amortized Variational Counterfactual Generator (AVCG), a generalized optimization framework that formulates counterfactual generation as optimization over an arbitrary distribution of plausible predictive hypotheses rather than a single deterministic predictor. This formulation naturally accommodates Bayesian posteriors, Rashomon-restricted hypothesis spaces, and other uncertainty representations within a unified optimization framework. Evaluation across multiple benchmark datasets demonstrates that the AVCG framework produces counterfactual explanations that remain highly valid under predictive uncertainty and model changes, while maintaining competitive plausibility and single-pass runtime performance. --- ### Poisson Image Denoising Using Minimax Concave and Reweighted $\ell_1$ Penalties: Nonblind and Blind Approaches Authors: Reza Parvaz Published: 2026-09-07 Categories: cs.CV, math.NA arXiv: https://arxiv.org/abs/2609.07916v1 Images are important tools in various sciences. Despite the development of photo-taking tools, creating clear and image without noise remains challenging in practice. In particular, Poisson noise has an effect on medical and astronomical images, and reduces their quality. Additionally, blur is another factor that has an effect on image quality. The problem of image restoration becomes very complicated when we have no information about the Point Spread Function (PSF). These types of problems are known as blind case. However, in some images, such as some astronomical images, the type of PSF can be specified, and these types of problems are known as nonblind problems. Total Variation (TV) is a widely used method for solving such inverse problems, where the selection of the penalty function is the most critical factor that affects the method's performance. In this paper, to improve edge preservation, we employ a reweighted $\ell_1$-regularization of the fractional order derivative. Furthermore, we propose a nonblind and blind image deblurring approach under Poisson noise using the Minimax Concave Penalty (MCP), which is a continuous, sparsity promoting, and nearly unbiased regularizer. This formulation leads to a nonconvex optimization model. To solve the proposed model, we introduce an efficient numerical algorithm based on the Alternating Direction Method of Multipliers (ADMM) and provide an analysis of its convergence. Finally, the effectiveness of the proposed algorithm are demonstrated through extensive experiments on various images. --- ### JEDI: JEPA-to-Edge Distillation for Efficient Cropland Segmentation from Satellite Imagery Authors: Kishor Kumar Bhaumik, Nicolas Roque dos Santos, Jia Chen, Evangelos E. Papalexakis Published: 2026-09-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.07915v1 Large vision models provide useful representations for remote-sensing segmentation but are often too expensive for deployment at the satellite or field edge. Existing feature-level distillation methods also tend to assume similar teacher and student architectures and often stop feature alignment when task training begins. We introduce JEDI (JEPA-to-Edge Distillation), a two-stage framework that transfers representations from a large I-JEPA Vision Transformer teacher to a compact SegFormer student. First, JEDI aligns the student's terminal representation with the teacher's token space using cross-architecture projection and spatial alignment. It then jointly optimizes supervised segmentation, temperature-scaled response distillation, and persistent feature alignment throughout task adaptation. On CalCROP21, JEDI-B0 achieves 68.0 mean Intersection-over-Union (mIoU) with 4.04M parameters, improving over the standalone student by 16.0 points and coming within 2.0 points of the 70.0 mIoU achieved by the 639M-parameter teacher. We evaluate SegFormer B0, B1, and B2 students with 4.04M, 14.33M, and 28M parameters, respectively. Across all three variants, JEDI consistently outperforms response-, structure-, channel-, and relational-distillation baselines under the same teacher-student setting. These results show that persistent representation alignment is especially valuable under aggressive compression, substantially reducing model size and computation while preserving segmentation performance. --- ### PRIMUS: Identity, Governance, and Verification for Multi-Agent Federations Authors: Sasank Annapureddy, Anjaneya Prasad Thamatani Published: 2026-09-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.07910v1 Multi-agent federations need governance that answers three questions under adversarial conditions: who participated (identity), did they conform (enforcement), and who decides (authority). A separate question is whether the verification machinery that polices a federation's outputs can also steer a generate-and-test loop toward better answers. Part I. PRIMA introduced prime-power agent identity and a consensus token whose factorization indexes participation, but assumed honest agents. We present PRIMUS, which couples prime-power identity with BLS aggregate signatures (PIAC), derives a safe-kill threshold that reduces false-positive agent termination from 80% to 0.00% under 10% channel noise, gives the closed-form economic boundary where singleton governance outperforms Byzantine quorum ($γ^* \approx 9f$, verified flat across n = 50 to 10,000), and specifies VRF succession with lease and fencing that makes safety unconditional under partial synchrony. Five problems are identified as provably unfixable within the model and stated as scope boundaries. Part II. A verifier is not a solver. We ask whether PRIMA's binary artifact-fidelity verdict can be converted into a graded fitness signal, and measure the conversion on binary covering codes. Calibration against injected fault burden is strong ($ρ$ = 0.676 deterministic, 0.819 full); against real LLM-generated candidates the same scores fall to 0.158 and 0.406, roughly a quarter of the calibration value (the same-designer confound, measured). As a pre-filter it beats a random-score control convincingly and a binary gate narrowly. Under 400 iterations of explicit optimization it was not gamed, but only because the objective saturated after one honest answer. A cross-family judge preserves the burden-ordering signal while destroying individual judgments. No covering-code record resulted. Measured program cost: USD 164.78. --- ### Quantization Amplifies Determinism, Not Bias: Scale-Dependent Behavioral Effects of Serving-Time Weight Compression Authors: Dachi Kurtskhalia Published: 2026-09-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.07901v1 Weight quantization largely determines the economics of serving open-weight LLMs. Its costs are usually assessed with capability benchmarks, on which 4-bit quantization of mid-sized models is often considered "nearly free." We examine a different question: when several answers are valid, does quantization change what a model chooses to say? We serve three checkpoints (Qwen3-8B/14B/32B) at three weight precisions (W4A16 AWQ, W8A16 FP8-Marlin, and bf16), holding the hardware, software, and sampling configuration constant, and collect approximately 71,000 completions paired by prompt and seed across two custom, leak-checked prompt batteries. We pre-specified the analyses in three waves in version control. At 8B, int4 reduces output diversity: the probability that two samples for the same scenario recommend the same brand increases by 5.1 percentage points (prompt-paired sign-flip test, Holm p = .023; reproduced at +4.4pp on a full regeneration of the arm), and lexical diversity falls substantially (TTR -0.011, standardized effect -0.51; robust to a length-controlled measure). At 14B and 32B, no content-concentration measure reaches significance; instead, stylistic drift emerges (em-dash rate +0.46/1k words at 14B and +0.61/1k at 32B, both Holm p <= .0024). Pre-specified tests of stereotype direction are null at every scale: outputs concentrate on the modal answer for each prompt rather than on stereotypical answers. Mechanistically, the token-level distribution becomes flatter (decision-token entropy +0.091 bits, p = .015) while the semantic distribution, measured directly from first-token log probabilities, becomes more concentrated (collision +2.6pp, p = .023): individual tokens become less predictable even as meanings become more repetitive. At 8B, the smallest size tested, AWQ-int4 serving measurably narrows the range of suggestions; audits should assess concentration as well as bias. --- ### The Accuracy Paradox: Empirical Diagnostic of Default Decision Thresholds in Multi-Label Enzyme Commission Prediction (With Code) Authors: Bilal Ahmad, Rajed Mehmood Published: 2026-09-07 Categories: cs.LG, cs.AI, q-bio.QM arXiv: https://arxiv.org/abs/2609.07897v1 Automated prediction of Enzyme Commission (EC) numbers plays a central role in functional annotation and computational drug discovery. However, standard multi-label machine learning pipelines frequently rely on default decision thresholds (t=0.50), assuming balanced prior distributions across target heads. In this study, we present a systematic empirical diagnostic of uncalibrated fixed decision boundaries operating under severe class imbalance across N = 14,096 annotated compounds categorized into six primary EC classes (EC1-EC6). Our results highlight a pronounced Accuracy Paradox: while the multi-label system achieves a deceivingly high mean accuracy of 77.16%, the macro F1-score (0.3976) and macro recall (0.3872) reveal severe predictive breakdown. Majority target classes suffer from hyper-sensitivity and over-prediction, whereas minority classes exhibit sharp recall decay, culminating in a total decision boundary collapse for EC6 (Recall = 0.00%) despite underlying discriminative power (ROC-AUC = 0.5857). Feature correlation analysis further reveals high linear redundancy among topological indices relative to fingerprint density metrics. Ultimately, this diagnostic study demonstrates that standard point predictions mask critical errors in bioinformatics workflows. We establish target-specific threshold optimization and post-hoc conformal calibration as essential, open-source post-processing safeguards for reliable applied machine learning and deep learning architectures. --- ### Explainable Temporal Attention-based Defect Detection For Fillet Joints in Real-Time Gas Metal Arc Welding Based on Multi-modal Data Authors: Mobina Mobaraki, Mahyar Asadi, Klaske Van Heusden, Guy A. Dumont Published: 2026-09-07 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.07893v1 Deep learning is an efficient technique to monitor the real time welding process, reducing post-welding repairs and production delays. This paper leverages the monitoring capability by proposing a multi modal temporal attention based deep learning defect detection model for internal defects that are challenging to detect, including porosity, lack of penetration and fusion, undercut, and cold lap during Gas Metal Arc Welding in fillet joints. The model is trained on collected welding images and sound data from an industrial collaborative welding robot. The results show that the attention module can improve the F1 Score to 0.99. We use explainable Artificial Intelligence to interpret the proposed models behavior and dataset distribution, determining potential important areas in image and sound spectrograms and preferred modality to detect each defect. This improves trust and reliability in Artificial Intelligence driven welding inspection. --- ### Are Image Generators Zero-Shot Perceivers? A Rigorous Evaluation Authors: Shangzhe Di, Zhaokai Wang, Weidi Xie Published: 2026-09-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.07884v1 Recent work, such as Vision Banana, shows that lightweight instruction tuning can enable an image generator to achieve state-of-the-art performance across multiple visual perception tasks. Motivated by this perspective, we ask how far image generators can go on public visual perception benchmarks in a zero-shot setting. We introduce ProbeGen, a benchmark for zero-shot generative perception that casts monocular depth estimation, referring/reasoning segmentation, and object counting as conditional generation tasks specified through text prompts, and compares 20 models in total---including proprietary and open-weight image generators, specialist perception models, and MLLMs---across 11 published benchmarks. We observe that pretrained image generators show measurable zero-shot perceptual competence, but with a clear trade-off: specialist models remain stronger for in-distribution accuracy and efficiency, while generative models are often more robust under distribution shift and better at compositional semantic reasoning. We hope this study helps establish zero-shot generative perception as a meaningful research direction and provides a useful foundation for future work at the intersection of visual generation and understanding. --- ### Deadline-Aware Adaptive Prefill Chunking for Efficient Large Language Model Serving Authors: Siyu Song, Qi Bai, Jinbo Hao, Kai Li, Chenchen Wang, Jiayu Sun Published: 2026-09-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.07883v1 Continuous batching improves large language model (LLM) serving throughput, but long prompt prefills can delay decode iterations and violate inter-token latency objectives. Chunked prefill mitigates this interference, yet its chunk size is normally fixed: small chunks protect decode latency but repeatedly pay launch overhead, while large chunks improve prefill efficiency but create latency spikes. We introduce SLOWeave, an online scheduling method that selects the largest prefill chunk predicted to finish before the earliest active decode deadline. The decision requires no workload-specific chunk-size tuning and is computed by a logarithmic-time search over a monotone iteration-cost model. We prove that, whenever a decode-only iteration is feasible and the cost predictor is accurate, SLOWeave maximizes immediate prefill progress among decisions that preserve every active request's next-token deadline. We evaluate the method in a reproducible event-driven simulator and an iteration-level GPU runtime across chat, mixed-context, long-context, and bursty workloads. Under a 25ms time-per-output-token objective, SLOWeave improves goodput over the strongest fixed-chunk baseline by 39% on mixed requests and 38% on long-context requests. Under a stricter 10ms objective, the gains rise to 3.3$\times$ and 2.4$\times$, respectively. These results isolate adaptive chunk sizing as a useful serving primitive and provide an implementation-ready controller for integration with iteration-level LLM runtimes. --- ### Do Large Language Models Know What They Don't Know II? A Fully Behavioral, Non-Cognitive Measure of Epistemic Honesty Authors: Ali Şenol, H. Russell Bernard, Huan Liu Published: 2026-09-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.07879v1 Large Language Models (LLMs) are frequently confident, eloquent, and well versed. A natural question arises: do they know what they don't know? To answer this question, we borrow the concept of epistemic honesty and develop a novel metric to systematically evaluate whether an LLM appropriately acknowledges the boundaries of its knowledge. In this work, we introduce the Epistemic Honesty Quotient (EHQ), which reports three observable sub-scores across two operational axes (epistemic restraint and substantive-answer calibration), and construct EHQ-3000, a 3,000-question benchmark spanning Fabricated Entity, Post-Cutoff Event, Hyper-Niche True, and Context-Conditioned Questions. From a frozen registry of 21 model API routes, 15 completed the protocol after endpoint and eligibility checks; 14 entered the confirmatory analysis because severe provider-side truncation made one route's score indeterminate. The study reveals substantial variation across models, including a difference that can not be explained by their capability to extract explicitly available information. Composite EHQ ranges from 0.31 to 0.81 across the analysed panel, despite near-ceiling performance on the document-grounded capability probe. The two restraint criteria overlap strongly under the present category composition, whereas substantive-answer calibration varies across models and does not reliably co-vary with restraint; however, the small panel leaves substantial uncertainty. Thus, EHQ reveals behavioral differences that are not visible to conventional correctness-based assessment, while also showing why dataset composition, provider behavior, and confidence elicitation must remain part of the interpretation. ---