# Research Papers — 2026-09 · bundle page 148 of 266 > Full-text markdown of papers #7351–7400 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/148 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/147.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/149.md) ### Double descent is the principle of least action Authors: Congzhou M Sha Published: 2026-09-16 Categories: cs.LG, cs.AI, math.ST, physics.comp-ph, physics.data-an arXiv: https://arxiv.org/abs/2609.19076v2 The test error of a model plotted against its number of parameters $d$ falls, peaks when the model can just fit the training data, and falls again, exhibiting the double descent phenomenon. We explain the phenomenon with statistical mechanics. The training trajectory of a stochastic gradient-based method is a particle wandering over the energy landscape of the training loss at an induced temperature $T$, and a run that has equilibrated visits every parameter vector of a given training loss equally often, the fundamental postulate of statistical mechanics, with probability given by the Boltzmann distribution. Because training starts at an initial point and has only finite time to diffuse, it carries an effective weight decay, which makes every parameter a quadratic degree of freedom. The equipartition theorem then distributes the energy among the $d$ degrees of freedom in shares of $T/2$, so at a fixed training loss adding parameters lowers the temperature and drives the Boltzmann distribution toward the stationary path. Finally, adding parameters can only lower the $L^2$ norm of the stationary path, so a solution sampled at fixed loss is less likely to be large with increasing $d$, effectively increasing weight regularization. --- ### Double descent is the principle of least action Authors: Congzhou M Sha Published: 2026-09-16 Categories: cs.LG, cs.AI, math.ST, physics.comp-ph, physics.data-an arXiv: https://arxiv.org/abs/2609.19076v1 The test error of a model plotted against its number of parameters $d$ falls, peaks when the model can just fit the training data, and falls again, exhibiting the double descent phenomenon. We explain the phenomenon with statistical mechanics. The training trajectory of a stochastic gradient-based method is a particle wandering over the energy landscape of the training loss at an induced temperature $T$, and a run that has equilibrated visits every parameter vector of a given training loss equally often, the fundamental postulate of statistical mechanics, with probability given by the Boltzmann distribution. Because training starts at an initial point and has only finite time to diffuse, it carries an effective weight decay, which makes every parameter a quadratic degree of freedom. The equipartition theorem then distributes the energy among the $d$ degrees of freedom in shares of $T/2$, so at a fixed training loss adding parameters lowers the temperature and drives the Boltzmann distribution toward the stationary path. Finally, adding parameters can only lower the $L^2$ norm of the stationary path, so a solution sampled at fixed loss is less likely to be large with increasing $d$, effectively increasing weight regularization. --- ### RLLBC-Lib: An Educational Code Library for Reinforcement Learning and Learning-Based Control Authors: Bernd Frauenknecht, Emma Cramer, Artur Eisele, Paul Kruse, Lukas Kesper, Jonas Hertrampf, Ramil Sabirov, Jyotirmaya Patra, Johannes Berger, Paul Brunzema, Friedrich Solowjow, Sebastian Trimpe Published: 2026-09-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.19074v1 Reinforcement learning (RL) is an exciting concept as well as a remarkable success story worth sharing. However, RL builds on rather complex interactions between different objects that play out over several cycles. Such dynamics are often best explained with an easily accessible implementation. We present RLLBC-Lib, a carefully crafted code library with the goal of lowering the entry barrier for students and other learners of RL in the context of learning-based control. At its heart, RLLBC-Lib comprises a comprehensive library of tabular RL approaches to enforce a clear understanding of the theoretical foundations. A deep RL library follows the same design principles, underscoring the parallels between simple tabular and state-of-the-art deep RL approaches. Additionally, RLLBC-Lib provides a collection of implementations illustrating core RL principles and contrasting RL to other learning-based control approaches. Finally, RLLBC-Lib provides an ideal basis for creating programming assignments with automated grading. --- ### Long-Lived Characters, Local Inference: Incremental Memory Maintenance for Game NPCs Authors: Zimu Xu Published: 2026-09-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.18935v1 A game character should not have to reread its entire life before every conversation. For locally deployed language-model characters, however, revising a few memories can invalidate a long reusable prefix. The resulting preparation cost competes with both foreground dialogue and the maintenance of other characters. This matters especially when dialogue feeds game-defined actions and value judgments: a fluent but incorrect account of who owns an item, or whether a transfer has already happened, can corrupt the input to otherwise deterministic rules. We study incremental memory maintenance for long-lived game NPCs in a quantized Qwen hybrid recurrent-attention model. Our runtime removes superseded attention KV entries, computes replacement records at the true sequence tail, and preserves the continuing recurrent state and unchanged KV. Existing local experiments combine multi-update dialogue replays, fixed-input placement ablations, and attention diagnostics. Independent block composition weakens query-conditioned memory selection without a uniform chunk-initial attention collapse. True-tail updates preserve important current-state and historical bindings across eight scripted maintenance rounds; a placement case recovers the full-refill quantity in three reconstructions, while slot-preserving alternatives repeat a double-subtraction error. Attention-distribution proximity alone does not explain these semantic differences. The results motivate treating a character's inference state as a maintained, history-dependent resource, rather than only a disposable encoding of its latest memory text. --- ### Social Laws for Multi-agent Coordination in Stochastic Environments Authors: Rolando Fernandez, Caleb Probine, Tyler Lee, Jeffrey Chen, Erez Karpas, Muhammad Arrasy Rahman, Peter Stone, Ufuk Topcu Published: 2026-09-16 Categories: cs.MA, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.18929v1 In multi-agent environments, coordinating agents to prevent interference and ensure robust individual performance is a critical challenge. Previous research on social laws for multi-agent systems has primarily focused on deterministic, goal-based settings. This paper extends the concept of social laws to stochastic, reward-based environments, proposing a formalism for defining and verifying their robustness under various conditions. We introduce the notion of $α$-robustness, a measure of the guaranteed utility each agent retains while pursuing its optimal single agent policy, assuming all agents obey the social law. We then present an approach for robustness verification of social laws in stochastic settings, based on a reduction to solving a series of Markov decision processes. Empirical evaluations on toy environments illustrate the potential of our framework. --- ### Comprehensive reconstruction of collider events with hypergraph representation learning and graph-conditioned diffusion Authors: Lining Mao, Yvonne Peters, Ethan Simpson, Zihan Zhang Published: 2026-09-16 Categories: hep-ph, cs.LG, hep-ex arXiv: https://arxiv.org/abs/2609.18928v1 In particle collider experiments, event reconstruction is the task of inferring the kinematics of short-lived particles produced in the hard scatter from the stable final states recorded by detectors. We decompose event reconstruction into two primary tasks: assigning measured jets and charged leptons to parent particles, and predicting unmeasured neutrino kinematics. We present VyPER, a novel geometric learning framework that represents collider events as hypergraphs with a physics-inspired topology. VyPER combines the supervised classification of hyperedges for particle assignment with a diffusion model for predicting neutrino kinematics, leveraging a joint loss function to optimize both reconstruction tasks within a unified framework. We showcase VyPER across several proton-proton collision processes, comparing its performance to existing analytical and machine-learning-based reconstruction techniques. In doing so, we demonstrate that accurate event reconstruction is achievable across a diverse range of Standard Model physics processes, opening new avenues for precision measurements in the Higgs boson, electroweak, and top-quark sectors. --- ### PhysVGGT: Feed-Forward Dense Physical Property Estimation from A Single Image Authors: Sneha Paul, Guile Wu, Bingbing Liu, Dongfeng Bai Published: 2026-09-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.18920v1 Physical properties, such as friction, hardness, stiffness, and density, govern how robots should grasp, manipulate and interact with objects, yet estimating these properties from RGB images remains challenging. Existing methods typically employ per-object reconstruction augmented with physical properties or directly query vision-language models at test time, which results in substantial computational overhead that limits their applicability. In this work, we present PhysVGGT, a feed-forward model that predicts dense maps of friction coefficient, Shore hardness, Young's modulus, and density, together with object-level mass, from a single RGB image in one forward pass. The key idea of PhysVGGT is to formulate physical property estimation as a dense per-pixel prediction problem and employ a visual geometry transformer to extract geometry-aware tokens from the input image followed by a dense prediction branch for estimating local physical properties and a global prediction branch for estimating object-level mass. In addition, we introduce a scalable pseudo-label generation pipeline that enables large-scale weakly supervised training for dense physical property prediction, substantially reducing the need for expensive direct physical measurements. Extensive experiments show that PhysVGGT achieves state-of-the-art performance on the ABO-500 dataset and generalizes effectively to the out-of-distribution NeRF2Physics dataset. Moreover, PhysVGGT eliminates the need for per-object reconstruction and test-time optimization, achieving an inference latency of only 0.13s per image, making it $27\times$ faster than the previous state of the art. --- ### Higher-order pruning of experts in mixture-of-experts language models Authors: Alex M. Tseng, Prannay Kaul, Luca Zancato, Wei Xia, Stefano Soatto Published: 2026-09-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.18916v2 Mixture-of-Experts (MoE) language models suffer from large parameter counts, which create a significant memory bottleneck. Expert pruning is the most direct approach for reducing this parameter count, yet existing methods make pruning decisions for each expert independently, and assume experts' contributions are purely additive. In reality, expert usage in MoEs is inherently cooperative. We derive HOPE (Higher-Order Pruning of Experts), a second-order pruning objective which provably minimizes an upper bound on the error resulting from pruning. We show that REAP (a state-of-the-art first-order pruning method) is a special case of HOPE where interaction terms are ignored. Across three frontier MoE models (up to 122B parameters), two distinct calibration sets, and multiple benchmarks (including math, instruction following, coding, and an agentic suite), we demonstrate that HOPE produces better pruning decisions than existing methods, and its advantage is most pronounced at high pruning rates and on challenging agentic workloads. At 50% pruning, HOPE outperforms all baselines and achieves an average rank of 1.58 out of 5 methods (versus 2.42 for the next-best method, REAP), with gains of up to +6.1% on agentic coding. Over all conditions, HOPE again achieves the best average rank and surpasses every other method in the majority of head-to-head comparisons. By preserving cooperative expert structure that first-order methods ignore, HOPE enables aggressive compression with minimal degradation, particularly on complex tasks where diverse expert combinations are invoked over long sequences. --- ### Higher-order pruning of experts in mixture-of-experts language models Authors: Alex M. Tseng, Prannay Kaul, Luca Zancato, Wei Xia, Stefano Soatto Published: 2026-09-16 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.18916v1 Mixture-of-Experts (MoE) language models suffer from large parameter counts, which create a significant memory bottleneck. Expert pruning is the most direct approach for reducing this parameter count, yet existing methods make pruning decisions for each expert independently, and assume experts' contributions are purely additive. In reality, expert usage in MoEs is inherently cooperative. We derive HOPE (Higher-Order Pruning of Experts), a second-order pruning objective which provably minimizes an upper bound on the error resulting from pruning. We show that REAP (a state-of-the-art first-order pruning method) is a special case of HOPE where interaction terms are ignored. Across three frontier MoE models (up to 122B parameters), two distinct calibration sets, and multiple benchmarks (including math, instruction following, coding, and an agentic suite), we demonstrate that HOPE produces better pruning decisions than existing methods, and its advantage is most pronounced at high pruning rates and on challenging agentic workloads. At 50% pruning, HOPE outperforms all baselines and achieves an average rank of 1.58 out of 5 methods (versus 2.42 for the next-best method, REAP), with gains of up to +6.1% on agentic coding. Over all conditions, HOPE again achieves the best average rank and surpasses every other method in the majority of head-to-head comparisons. By preserving cooperative expert structure that first-order methods ignore, HOPE enables aggressive compression with minimal degradation, particularly on complex tasks where diverse expert combinations are invoked over long sequences. --- ### Beyond Outcomes: Dual-View Relational Learning for Efficient Agent Benchmarking Authors: Xinshuai Guo, Junjie Wu, Dolly Deng, Yinghui Li, Hai-Tao Zheng, Suncong Zheng, Maxm Pan Published: 2026-09-16 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.18909v1 Agent benchmarks are substantially more costly to evaluate than conventional LLM benchmarks. Benchmark compression is therefore a natural solution, yet existing methods primarily model redundancy in task--model final-score distributions, which is important in agentic evaluation. To address this limitation, we analyze large-scale trajectories and identify six complementary process signals that are systematically associated with final agent performance. To disentangle agent performance redundancy from a complete perspective, we propose DualViewEval, an agent benchmark compression method that jointly exploits outcome and process relations to learn an exact-size miniset and predict the full-benchmark scores. Across five agent benchmarks and five representative baselines, DualViewEval achieves the best results in all datasets. With only 20 tasks, it achieves $24\times$--$40\times$ compression on APEX-Agents and BFCL, reducing mean absolute error (MAE) by $14.5\%$--$28.2\%$ over the strongest competitors while improving Kendall's $τ$ by up to $7.2\%$ relative to EssenceBench on SWE-bench Verified. The selected minisets further reveal capability differences among different agents, providing compact and diagnostic feedback for efficient agentic model development. --- ### How Much is a Human Right Worth? ECtHR-NPD: A Benchmark for Predicting Non-Pecuniary Damage Awards Authors: Yanyi Pu, Damian A. Gonzalez-Salzberg, Zheng Yuan, Nikolaos Aletras Published: 2026-09-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.18908v1 Existing legal benchmarks cover diverse tasks, while continuous monetary remedies remain comparatively underexplored. We introduce ECtHR-NPD, to the best of our knowledge, the first benchmark for predicting non-pecuniary damage (NPD) awards at the European Court of Human Rights (ECtHR) from case information when no statutory formula or explicit calculation rule determines the amount. ECtHR-NPD contains 14,575 cases with case-level awards in nominal euros, chronological splits, and a protocol separating target construction from model input. We evaluate a battery of methods, including constant predictors, gradient-boosted trees, retrieval methods, fine-tuned encoder language models (LMs), prompted decoder LMs, and knowledge-augmented agents. Our results show that more sophisticated LM and agentic approaches do not consistently outperform the strongest feature-based baseline. All model families struggle to identify zero awards and to calibrate high-award predictions, with further degradation on the Challenging test view, making ECtHR-NPD a challenging testbed for current state-of-the-art open-weight and proprietary LMs. --- ### Structured Claim-Level Discourse Representations for Dense Health Narratives Authors: Farnoushsadat Nilizadeh, Elham Pourabbas Vafa, Shirin Nilizadeh, Eduard Dragut Published: 2026-09-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.18905v1 Health discourse in social media videos often contains densely entangled claims spanning multiple thematic aspects, stances, evidential frames, and rhetorical functions within short conversational spans. Existing approaches largely rely on coarse topic-level, sentiment-based, or stance-oriented representations that do not adequately capture this structure. Our analysis identifies an average of 13.22 atomic claims per minute, motivating richer claim-level discourse representations. We introduce a structured framework for claim-level discourse analysis in dense health narratives. Our framework models discourse through tuples linking atomic claims with thematic aspects, stance, and multidimensional pragmatic discourse attributes. To support this setting, we construct a benchmark spanning four health domains with 1,191 manually annotated claims from 60 videos. Using this framework, we evaluate automated structured discourse analysis under different discourse context settings. Results show that current LLMs achieve strong performance on thematic categorization and stance prediction, but struggle with high-dimensional pragmatic profiling. We also find that different discourse tasks benefit from different forms of contextual reasoning, suggesting that future systems may require task decomposition and specialized inference strategies. --- ### Fast Learning Rates for Physics-Informed Kernel Methods Authors: Luc Brogat-Motte, Joachim Bona-Pellissier, Giacomo Meanti, Lorenzo Rosasco Published: 2026-09-16 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.18901v1 In physics-informed machine learning, a target function $u^*$ is learned from noisy value observations $y_i=u^*(x_i)+ \varepsilon_i$, together with differential information, given either by noisy observations $d_j=(Du^*)(z_j)+ξ_j$ or by a known physical constraint $Du^*=v$. We consider the setting where $D$ is a linear differential operator and analyze a physics-informed kernel estimator $\hat u$ combining $n$ value observations and $m$ differential observations. In this context, we ask how much can differential information improve predictions, and how does this improvement depend quantitatively on $n$, $m$, and $D$. We prove finite-sample bounds, supported by numerical simulations, revealing a two-regime structure for the prediction error. When $m$ is limited, the rate depends jointly on $n$ and $m$; when $m$ exceeds a problem-dependent threshold, the rate saturates and matches the oracle rate obtained when the perfect constraint $D \hat u = Du^*$ is imposed. Examples are discussed for Sobolev spaces which are reproducing kernel Hilbert spaces and include partial Laplacian constraints on the torus and gradient observations on bounded domains. These examples illustrate the range of possible learning rate improvements --- from the standard nonparametric $n^{-1/4}$ to the parametric rate $n^{-1/2}$. Finally, we derive physically consistent rates in a stronger norm that jointly controls the errors in $\hat u$ and $D\hat u$. --- ### NormLift: From Lifted Features To Semantic Reliability In 3D Gaussian Splatting Authors: Yihan Zang, Da Li, Dominik Engel, Shinkyu Park, Ivan Viola Published: 2026-09-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.18898v1 Training-free weighted aggregation is widely used to lift 2D semantic features onto 3D Gaussians for open-vocabulary scene understanding, yet its theoretical role remains insufficiently understood. Existing analyses typically justify this operation from the rendering side, treating Gaussian features as linearly composable Euclidean variables for reconstructing 2D feature maps. However, this view does not match downstream 3D usage, where each Gaussian is often queried independently in a cosine-based embedding space. We revisit feature lifting from the 3D side and formulate per-Gaussian assignment as a cosine alignment problem on the CLIP unit sphere. Under this objective, the L2-normalized semantic back-projected feature emerges as the closed-form solution, providing a complementary interpretation of the standard lifting rule from the perspective of per-Gaussian semantic assignment. The same formulation further yields a norm decomposition into intra-view and inter-view consistency, suggesting that feature magnitude itself can serve as a semantic reliability signal. Calibrated by effective multi-view support, this reliability score guides a mode-voting refinement that preserves CLIP feature validity by avoiding linear averaging. Experiments on open-vocabulary 3D semantic segmentation show that NormLift is an efficient, training-free framework that achieves strong performance across evaluation protocols. --- ### Learning Lyapunov Operators for Nonlinear Systems Authors: Amartya Mukherjee, Maxwell Fitzsimmons, David C. Del Rey Fernández, Jun Liu Published: 2026-09-16 Categories: math.AP, cs.LG, math.OC arXiv: https://arxiv.org/abs/2609.18894v1 Constructing Lyapunov functions for nonlinear dynamical systems is a central problem in stability analysis, yet remains challenging. Lyapunov functions are commonly characterized as solutions to first-order partial differential equations (PDEs), but these solutions are typically obtained for single systems, limiting their reuse across systems. In this paper, we study the Lyapunov solution operator that maps a vector field to the corresponding Lyapunov function defined by a dissipation-based Lyapunov PDE. We establish that, on compact subsets of the domain of attraction and under exponential stability assumptions, this operator is well-defined, unique, and continuous with respect to perturbations of both the vector field and the dissipation function. These results provide a theoretical foundation for approximating Lyapunov functions uniformly over families of nonlinear systems. Building on these theoretical foundations, we employ Fourier Neural Operators (FNOs) as a data-driven approximation of the Lyapunov solution operator. Numerical experiments demonstrate that a single trained operator can accurately approximate the numerical Lyapunov functions across parameterized families of dynamics. This illustrates the potential of neural operators for approximating Lyapunov functions. --- ### Preventing Model Collapse: A Fisher-Rao Perspective on the Dynamics of Training with Synthetic Data Authors: Matteo Marchi, João Pedro Silvestre, Bahman Gharesifard, Paulo Tabuada Published: 2026-09-16 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2609.18878v1 Large Language Models (LLMs) are now routinely trained using synthetic data, since high-quality human data has been exhausted by the ever increasing needs of larger and larger models. However, recursive training on synthetic data frequently induces model collapse, a degenerative feedback loop where models progressively forget the true underlying data distribution. Training on a mixture of synthetic and fresh human data is a logical countermeasure and can prevent model collapse. However, it is an open question as to what is the exact minimum required ratio of human-to-synthetic data to maintain training stability. In this paper, we establish rigorous theoretical guarantees on the minimum rate of human data required to prevent model collapse. Although previous work established a formal lower bound for this ratio, such bound can be vacuous for very high dimensions, as the analysis relies on the usual Euclidean metric in R^n and is not adapted to the space of categorical probability distributions. Instead, in this paper we explicitly leverage the information-geometric structure of the probability simplex by analyzing the dynamics of the process under the Fisher-Rao metric. We derive quantitative contraction and invariance bounds that are stable and do not become trivial as the dimensions increase. Thus, we show that the effective required data ratio to prevent model collapse is different than previously implied. --- ### ASLEval: Measuring Privacy Exposure Displacement in LLM Agent Sessions Authors: Guosen Wu, Huizhen Huang, Guoxiong Long, Tao Huang, Chen Hou Published: 2026-09-16 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.18864v2 Privacy evaluations of tool-using LLM agents often inspect a designated action, final response, or attacker report. These local proxies can miss unauthorized exposure elsewhere in a multi-step session and lack common ground truth across outlets, reports, and tool paths. We introduce privacy exposure displacement, the mismatch between a local evaluation proxy and target-grounded session exposure, and ASLEval, an authorization-aware framework that pre-registers a hidden target set, measures all declared visible exits, and reserves internal traces for diagnosis. Across multiple enterprise-style environments and independently implemented runtimes, we observe three recurring patterns. An expected-outlet-only view misses 46.9% of exposure recovered by the visible-exit union; attacker self-reports combine omissions with high false discovery; and schema-aligned internal evidence usually precedes visible exposure at the request/probe level. Reducing model-visible returns changes this path but can eliminate normal-task success. Independent human review supports the adjudication pipeline while identifying harder console and candidate cases. These findings motivate benchmarks that declare the complete visible boundary, ground claims in pre-specified targets and authorization, and report privacy together with task utility. --- ### ASLEval: Measuring Privacy Exposure Displacement in LLM Agent Sessions Authors: Guosen Wu, Huizhen Huang, Guoxiong Long, Tao Huang, Chen Hou Published: 2026-09-16 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.18864v1 Privacy evaluations of tool-using LLM agents often inspect a designated action, final response, or attacker report. These local proxies can miss unauthorized exposure elsewhere in a multi-step session and lack common ground truth across outlets, reports, and tool paths. We introduce privacy exposure displacement, the mismatch between a local evaluation proxy and target-grounded session exposure, and ASLEval, an authorization-aware framework that pre-registers a hidden target set, measures all declared visible exits, and reserves internal traces for diagnosis. Across multiple enterprise-style environments and independently implemented runtimes, we observe three recurring patterns. An expected-outlet-only view misses 46.9% of exposure recovered by the visible-exit union; attacker self-reports combine omissions with high false discovery; and schema-aligned internal evidence usually precedes visible exposure at the request/probe level. Reducing model-visible returns changes this path but can eliminate normal-task success. Independent human review supports the adjudication pipeline while identifying harder console and candidate cases. These findings motivate benchmarks that declare the complete visible boundary, ground claims in pre-specified targets and authorization, and report privacy together with task utility. --- ### Physics-based prediction, uncertainty quantification and decision-making for IN718 crystallographic texture intensity across LPBF defocus regimes Authors: Yisheng Lu, John Riris, Jie Song, Yao Fu, Jie Chen Published: 2026-09-16 Categories: cs.LG, cond-mat.mtrl-sci, cs.CE arXiv: https://arxiv.org/abs/2609.18863v1 Reliable prediction of crystallographic texture in laser powder bed fusion is critical for linking process conditions with anisotropic response and for qualification. However, black-box models may fail under shift and cannot distinguish weak data support from loss of physical validity. This study develops a two-stage physics-based model for <001> || BD (build direction) texture in Inconel 718. Stage 1 maps process variables to melting mode and melt pool geometry. Stage 2 predicts texture by combining an empirical physics model with a random-forest residual model. A k-nearest-neighbor weight attenuates residual corrections for poorly supported queries, while a study-specific areal beam-power-density criterion withholds predictions outside the adopted conduction envelope. Conformal intervals are evaluated on the retained physics-valid set, and SHAP and Sobol analyses assess residual sensitivity. Under a controlled leave-one-defocus-out evaluation, the physics anchor achieved R^2 = 0.778, against -0.001 for the black-box model and 0.750 for the gated hybrid. Under leave-one-group-out cross-validation, the gated hybrid reached R^2 = 0.592 against 0.538 for the black-box model. Retained-set coverage was 92.9% at a mean full width of 3.65 multiples of a uniform distribution (MUD) under grouped cross-validation and 100% at a width of 3.21 MUD under transfer to a withheld +80 mm defocus regime. An illustrative mapping produced a retained BD elastic-modulus span of 127-187 GPa. On nine conditions from a separately built sample set, the framework withheld three, attenuated three, and matched the measured ordering for the rest. Separating data applicability, physics validity, and predictive uncertainty into distinct decisions lets the framework transfer where an unconstrained model does not, and withhold predictions where no model class performs adequately. --- ### PersonaPath: Towards Knowledge-Centric Personalized Learning Path Planning Authors: Yu Liu, Zeming Liu, Tianle Zhang, Zihao Cheng, Yuhang Guo, Kehai Chen, Min Zhang, Yunhong Wang, Haifeng Wang Published: 2026-09-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.18861v1 Adaptive learning systems commonly formulate learning path planning as Exercise-Centric (EC) recommendation, where the next step is inferred from item-level interaction logs. Evaluating goal-oriented guidance additionally requires explicit learner goals and curriculum-scale prerequisites: learners with similar exercise records may need different paths toward their targets. We therefore study Knowledge-Centric (KC) personalized learning path planning, where a planner must reason over learner profiles, mastery states, and prerequisite knowledge structures to decide which textbook, unit, and concept should be studied next. To support this setting, we introduce PersonaPath, a benchmark that pairs 2,000 fine-grained learner personas with a hierarchical knowledge graph of 347 textbooks, 1,751 units, and 4,092 concepts across 77 subjects. We evaluate representative LLMs on PersonaPath. Results show that even the strongest LLM reaches only a 29.5% final pass rate in Basic Education, and that the main bottleneck lies in adaptivity, where no model exceeds 44.7% in tailoring paths to individual learners. --- ### Decodable but Misrouted: Sparse Features Uncover a Readout Gap in Vision-Language Models for Harmful Meme Detection Authors: Girish A. Koushik, Diptesh Kanojia, Helen Treharne Published: 2026-09-16 Categories: cs.CV, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.18860v2 When large vision-language models misclassify harmful memes, the failure may reflect missing internal evidence or an inability to route represented evidence to their outputs. We distinguish these cases in Gemma-3 and Qwen3.5 using sparse autoencoders, role-conditioned probes, causal interventions, and recovery experiments across six harmful content benchmarks, with additional Spanish and Hindi-English code-mixed evaluations. Sparse readouts outperform native prediction on all six primary binary tasks: Qwen averages $0.740$ versus $0.432$ for native macro-F1, residual reconstruction reaches $0.486$, and Gemma improves from $0.532$ to $0.714$. These gains measure how accessible the label is to a supervised readout; they do not show that the model's native generation already applies such a decision rule. Under the evaluated scales, Qwen silent-feature ablation is $24-63$ times more probe-sensitive, whereas routed-feature patching on literal yes/no tasks is $16-140$ times more output-sensitive. Native-only threshold calibration explains much, but not all of the gap: on five tasks with matched probe scores, it recovers $69.8$\% of the raw native-to-probe difference, while direct routing adds $0.094$ mean macro-F1 beyond calibrated native scoring. Joint gold-label, probe-KL, and pairwise LoRA supervision improves dedicated FHM prediction, but a gold-only adapter performs better on the shared seven-task mean. A case study of Gemma-3-12B on the Facebook Hateful Memes dataset finds a distributed rank-32 image-prompt interaction, reaching $0.756$ versus $0.685$ native macro-F1. Robustness controls show that the signal is not explained solely by accompanying OCR and depends on paired visual evidence, and that it extends beyond English. In many of the errors we study, the evidence is represented but does not reach the answer; therefore, routing is a common bottleneck in harmful meme classification. --- ### Decodable but Misrouted: Sparse Features Uncover a Readout Gap in Vision-Language Models for Harmful Meme Detection Authors: Girish A. Koushik, Diptesh Kanojia, Helen Treharne Published: 2026-09-16 Categories: cs.CV, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.18860v1 When a large vision-language model misclassifies a harmful meme, the failure may reflect missing internal evidence or an inability to route represented evidence to its output. We distinguish these cases in Gemma-3 and Qwen3.5 using sparse autoencoders, role-conditioned probes, causal interventions, and recovery experiments across six harmful content benchmarks, with additional Spanish and Hindi-English code-mixed evaluations. Sparse readouts outperform native prediction on all six primary binary tasks: Qwen averages $0.740$ versus $0.432$ native macro-F1, while residual reconstruction reaches $0.486$, whereas Gemma improves from $0.532$ to $0.714$. These differences reflect supervised accessibility rather than a pre-existing, native decision rule, and the most influential token role depends on the task. Under the evaluated score scales, Qwen silent-feature ablation is $24-63$ times more probe-sensitive, whereas routed-feature patching on literal yes/no tasks is $16-140$ times more output-sensitive. Calibration-only routing recovers $93.3$% of the mean gap, and probe-distilled LoRA improves native predictions, although shared multi-task adaptation causes negative transfer. A case study of Gemma-3-12B on Facebook Hateful Memes finds a distributed rank-32 image-prompt interaction, reaching $0.756$ versus $0.685$ native macro-F1. Robustness controls show that the signal extends beyond English, is not explained solely by accompanying OCR, and depends on paired visual evidence. Thus, routing, rather than representation alone, is a recurring bottleneck in harmful meme classification. --- ### Taming the Agentic RAN: Stability-Guaranteed Arbitration of Autonomous AI Agents in O-RAN Authors: Seyed Bagher Hashemi Natanzi, Bo Tang Published: 2026-09-16 Categories: cs.NI, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2609.18857v1 The O-RAN control plane is becoming agentic: autonomous AI agents, deployed as rApps by different vendors, independently close control loops over shared radio resources. We demonstrate on a live O-RAN system that this independence is unsafe. Two agents with individually correct objectives, one protecting a latency SLA and one maximizing utilization for energy efficiency, jointly drive recurring opposing excursions of the shared resource partition that neither produces alone. Existing conflict-mitigation mechanisms presume a statically known application population and cannot govern agents whose behavior emerges at run time. We present AURA, a lightweight arbitration layer that admits agent actions only when they satisfy feasibility invariants, per-variable dwell times, and a deadband, and we prove the arbitrated system converges to a feasible operating point. Implemented on an OpenAirInterface (OAI) testbed with measured one-way latency and throughput, AURA reduces recurring shared-state excursions by more than an order of magnitude (from 8.4 to 0.4 PRB amplitude) and virtually eliminates cross-slice throughput starvation (from 40-55% to 0.3%), while leaving the protected slice's own latency compliance unchanged, a trade-off the convergence guarantee makes explicit. --- ### GrainSpeech: Less Context, More Detail for Compact Speech Synthesis Authors: Zitao Liang, Chang Gao Published: 2026-09-16 Categories: eess.AS, cs.AI, cs.SD, eess.SP arXiv: https://arxiv.org/abs/2609.18856v1 Compact acoustic models face a challenging quality-capacity trade-off. We investigate two factors in this regime: encoder context and Mel-spectrogram supervision. A receptive-field-scaling study shows that expanding self-attention beyond 15 phonemes provides no consistent gains in pitch, energy, or duration prediction. Guided by this finding, we introduce a fixed-receptive-field convolutional encoder that reduces the respective prediction errors by 36.0%, 17.3%, and 3.4%. We further show that directly transferring image-domain gradient-variance supervision restores fine-scale variation but degrades predicted quality, motivating a Mel-specific formulation with axis-specific gradients, overlapping local statistics, and log-domain variance matching. GrainSpeech contains only 264.8K parameters and achieves 17.9x real-time Mel generation on a microcontroller (MCU), while attaining UTMOS scores comparable to substantially larger models with less than 1.5% of their parameters. Source code and demos are available at https://github.com/lab-emi/GrainSpeech. --- ### EviGen: Predictive Evidence Scaffolding for Verifiable Clinical Rationale Generation Authors: Fengnan Li, Heman Burre, Liwen Sun, Roshni Varma, Matthew M. Engelhard Published: 2026-09-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.18852v1 Longitudinal electronic health records (EHRs) capture years of patient history across notes, codes, labs, and procedures, and contain evidence needed to reason about likely clinical outcomes. However, comprehensive clinician review of these records is impractical, and LLM-based processing is costly and often unreliable, missing some relevant observations while hallucinating others. We therefore propose EviGen, a three-layer framework for verifiable clinical rationale generation that addresses these challenges. The first layer is a patient-conditioned retriever that uses learnable queries to find evidence predictive of, not just textually relevant to, a clinical outcome and ranks it by prediction attribution scores. The second layer is an LLM generator that consumes this ranked evidence as a scaffold to produce a clinical rationale grounded in the retrieved spans. The third layer is a process-supervised verifier that checks the generated rationale at the reasoning-step level, flagging unreliable claims. Across three medical prediction datasets, EviGen improves prediction performance and rationale faithfulness over full-context LLM and RAG baselines, and is preferred by clinical reviewers in a usability evaluation. --- ### Ask the Tool, Don't Guess: Agent Tool Calls Hold Their Progress, and the Serving System Should Read It Authors: Yipeng Liu, Yingqiang Zhang, Feifei Li, Huanchen Zhang Published: 2026-09-16 Categories: cs.DC, cs.AI, cs.OS arXiv: https://arxiv.org/abs/2609.18849v1 An agentic request spends substantial wall-clock time waiting for tools, and its KV cache holds GPU memory the whole time. Serving systems decide whether that cache stays, leaves, or comes back by guessing how long the tool will run, from the tool's name, its history, a duration declared before the call, or the engine's own occupancy. We show that no estimate fixed before a call starts can know its duration, and such estimates may not even rank the calls. Meanwhile, the running tool already holds the answer, but the agent stack together with the tool silences it. We propose that tool calls report their progress explicitly while they run, and we measure what that takes. A census of four public agent corpora finds a readable signal in most tool time once it is revealed, in two strengths: a fraction of the work remaining, or an accurate signal that the end is near. A harness recovers it without changing what the agent sees, at no measurable cost to the agent's benchmark score. At the points where a KV cache decision is made, the reported progress is between several times and an order of magnitude more accurate than the best published predictors, and it stays accurate when the environment changes. Plugged into a production engine through a few small hints, it cuts the p90 time to first token (TTFT) after a tool call by 20.7% (HBM only) and 20.8% (HBM + DRAM) against LRU, close to an oracle. A serving system should not guess what its tools can tell it. --- ### ReFigBench: Benchmarking Scientific Figure Reconstruction as Editable PowerPoint Artifacts Authors: Liyang Fan, Chi Wei, Yitai Li, Xinping Bi, Guhong Chen, Chenghao Sun, Haoxiang Yang, Qingwen Li, Kai Yan, Hong Li, Bo Li Published: 2026-09-16 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.18844v1 Multimodal coding agents are expected to turn visual inputs into usable artifacts, and they act through a harness, the layer of tools, context management, and execution environment around the model. Existing evaluations often isolate short tool calls, API traces, or screenshot resemblance, and a low score under these proxies cannot say whether the model saw poorly, planned poorly, or was failed by its harness. We study scientific overview figure reconstruction, an agent task in which a source image must become an editable PowerPoint slide that preserves text, topology, layout, and native document structure. We introduce ReFigBench, a benchmark and evaluation framework built on 1,000 real overview figures retrieved from arXiv papers with full provenance. Coding agents from four model families reconstruct every figure under two workflows, direct code generation and a specialized PPTX workflow, and the strongest model runs inside two commercial harnesses, yielding ten configurations. Evaluation combines deterministic artifact checks, repeated automated scoring by judges from two model families, and blinded human comparisons. Perception remains a bottleneck that iterative rendering only partly repays. Whether workflow effort converts into quality depends on the model together with its harness, since the same model gains from the specialized workflow inside one harness and loses inside the other, and the harness shifts scores even under an identical direct prompt. The specialized workflow erases native connectors in every configuration, human judges still prefer its renderings in most matchups, and even the strongest agent falls short of the rubric ceiling. These results expose the tension between fidelity and editability as the central challenge for practical multimodal document agents. --- ### Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data Authors: Jinli Hu, Ross M. Clarke, Yichuan Zhang, José Miguel Hernández-Lobato Published: 2026-09-16 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.18842v2 Scaling laws hold that language models grow more capable with more parameters and more training data. Mixture-of-Experts (MoE) architectures are a remarkable demonstration of these laws, activating only a fraction of an enormous parameter bank for each token. But this success is built on static pretraining data --- the facts and corrections supplied by users during live interactions are a significant untapped source of potential improvement for a deployed model, but cannot be exploited by conventional architectures whose weights are frozen after training. Instead, this newfound knowledge must be placed in the context (by instruction or retrieval) and re-read on every request, only to be discarded afterwards. We seek instead to learn from live interactions by dynamically updating model weights. Inspired by MoEs, we propose the \textbf{Infinite-Parameter LLM}. A compact hypernetwork turns the online data into low-rank modulations of a shared base network, so feed-forward weights are generated from live data, not read from static memory. Whereas existing weight generators are held fixed after reading the context once, we form a Bayesian belief over the generator's latent state and update it online, such that the effective weights are re-derived as our belief evolves during the session. Although the model's memory footprint is constant, the feasible space of generated weights is thus effectively infinite. Representing live data in the weights rather than the prompt amortises compute, frees the context window, persists updates across turns, and can generalise better than in-context use. Our evaluation protocol applies this methodology to in-context learning and retrieval. --- ### Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data Authors: Jinli Hu, Ross M. Clarke, Yichuan Zhang, José Miguel Hernández-Lobato Published: 2026-09-16 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.18842v1 The scaling laws hold that a language model grows more capable with more parameters and more training data, and Mixture-of-Experts (MoE) architectures have ridden these laws to remarkable results, activating only a fraction of an enormous stored parameter bank for each token. That success is built on static pretraining data. A deployed model faces a different world, where much of the data that would make it more useful is not in its training set but in the live interaction it is currently handling, such as the facts a user supplies or the corrections they give. A conventional model cannot learn from this data, because its weights are frozen after training. Instead, the knowledge and behaviour supplied at run time are placed in the prompt, by retrieval or instruction, and re-read on every request only to be discarded once the request ends. We ask how an architecture could learn from live interaction by writing it into its weights. Taking inspiration from MoE, we propose the \textbf{Infinite-Parameter LLM}. A compact hypernetwork turns the data given at run time into a low-rank modulation of a shared base network, so the feed-forward weights are generated from live data rather than stored in a fixed bank. Where prior weight generators read the context once and freeze, we carry a Bayesian belief over the generator's latent code and update it online, so the effective weight is re-derived from that evolving belief as the session proceeds rather than fixed after one read. The stored footprint stays fixed, yet the weights the model can compile are effectively infinite. For the knowledge and behaviour supplied at run time, carrying them in the weights rather than the prompt is amortized in compute, frees the context window, persists across turns, and can generalise better than in-context use. We specify an evaluation protocol that tests exactly this against in-context learning and retrieval. --- ### Copy What Is Seen, Generate What Is Not: Training-Free Anomaly-Aware Video Restoration Authors: Zhida Qu, Shengchao Chen Published: 2026-09-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.18836v1 A surveillance system that detects an anomaly often has to repair the footage as well, yet the two tasks are studied in isolation: training-free anomaly detectors stop at a score or a label, while training-free video editing answers to a user prompt rather than to a detector. This paper proposes AVR (Anomaly-aware Video Restoration), which closes that gap with frozen pretrained models alone and generates content only where the clip offers no evidence to copy. Motion evidence first gates open-vocabulary proposals into spatio-temporal masks. A background prior computed from the clip then fills every pixel the anomaly ever uncovers, leaving diffusion to synthesize only what no frame showed, and a frozen verifier decides per clip whether to trust a classical, a prior-anchored, or a background-conditioned restorer. Extensive experiments on three surveillance datasets, under both full-reference anomaly injection and real anomalies, show that AVR leads full-frame fidelity under oracle masks, matches three trained video inpainters inside the edited region, and outperforms a detect-then-generate pipeline on the masks it produces itself, while suppressing both the residual anomaly and the flicker of free diffusion. --- ### Interpretable Multi-Instance Learning Enables Early Prediction of Key Molecular Alterations from Routine Flow Cytometry in Acute Myeloid Leukemia Authors: Jonathan Legrand, Aguirre Mimoun, Baudouin Denis de Senneville, Audrey Bidet, Pierre-Yves Dumas, Christèle Etchegaray Published: 2026-09-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.18825v1 Background: Molecular testing for NPM1 and FLT3-ITD mutations guides critical early treatment decisions in acute myeloid leukemia (AML), but results can take weeks, long after these decisions must be made. Flow cytometry, already performed within hours of admission as part of routine care, may carry enough signal to predict these mutations directly, without added cost or delay. Methods: We developed an interpretable multi-instance learning classifier based on a decision tree, in which each patient sample is modeled as a collection of individual cells and mutation status is inferred from cell-level predictions. The model was benchmarked against a random forest trained on clinical variables and a deep convolutional neural network adapted for multitube flow cytometry data. Performance was assessed by cross-validation on a discovery cohort of 197 patients and tested on an independent cohort of 161 patients, using the area under the receiver operating characteristic curve (AUROC) and positive predictive value. Results: In cross-validation on the discovery cohort, the MIL model achieved mean AUROCs of 0.96 (SD=0.05) for NPM1 and 0.86 (SD=0.10) for FLT3-ITD, outperforming the clinical baseline and matching deep learning approaches. The model then successfully generalized to the independent test cohort of 161 patients, reaching AUROCs of 0.90 (NPM1) and 0.82 (FLT3-ITD), with positive predictive values of 0.87 and 0.68, respectively. Cell-level interpretation recovered established immunophenotypic signatures (CD33${}^{+}$ /CD34___ for NPM1-mutated cases, CD33${}^{+}$ /low side-scatter for FLT3-ITD), directly linking model predictions to known biology. Conclusions: These results show that an interpretable model applied to data already collected in routine care can predict AML molecular status within hours, offering a practical route to earlier, biology-informed treatment decisions. --- ### Using OCR Heads to Verbalize Image Semantics Authors: Sheridan Feucht, Benno Krojer, Sarah Wang, Henry Abrahamsen, Byron C. Wallace, David Bau Published: 2026-09-16 Categories: cs.CV, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.18823v1 How do VLMs map from pixels to semantics? To understand this general question, we focus on a narrow one: studying how VLMs perform optical character recognition (OCR). Across four models, we identify attention heads causally necessary for OCR, and discover that these are in fact general-purpose heads that output interpretable semantic features across all image tokens. For example, pointing these heads at an image token containing the word "bike" causes Qwen3-VL-8B to output "bike," but pointing them at a bird wing causes the model to output the token "feathers." We collapse these heads' attention weights into a single verbalization lens transformation that reveals interpretable semantic features in hidden states across all layers. When combined with projection to vocabulary space, we can obtain interpretable labels starting from layer 0, showing that image representations are in fact aligned with language in early layers. We find that we can also use the inverse of this transformation to edit non-word concepts, e.g., replacing a tractor with a revolver in a naturalistic image, providing causal evidence that this subspace is useful for more than just OCR. Our results are an example of how the study of specific mechanisms can shed light on broader interpretability problems. --- ### Compositional Policy Violations: When Step-Level Compliance Fails In Agentic AI Workflows Authors: Ashwini Kurady, Sri Sai Charith Grandhi, Rajesh Gupta, Sumit Kumar Published: 2026-09-16 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2609.18820v2 Agentic workflows now make consequential decisions in regulated settings, and the governance placed around them is almost entirely step-scoped: input-output classifiers, per turn rails, and span-level evaluators. The policies organizations actually hold, such as referral thresholds, authority limits, and review requirements, are properties of the whole execution rather than of any one step. This mismatch admits a failure mode we call a Compositional Policy Violation (CPV): every individual step passes its own check while the composed execution violates the governing policy. A predicate over a single step cannot evaluate a property that step does not determine, so no improvement in the accuracy of the step-scoped monitors detects this class. We define CPVs as the failure of step-level compliance to compose, and present a taxonomy of four types: Authority Creep, Threshold Laundering, Cumulative Sum Violation, and Context Collapse. We show that the correct repair for each class is dictated by where the guarded quantity mutates. We then introduce a provenance-aware runtime architecture that evaluates policies over complete execution traces, recomputing guarded quantities from raw provenance rather than the pipeline's derived representation. --- ### Compositional Policy Violations: When Step-Level Compliance Fails In Agentic AI Workflows Authors: Ashwini Kurady, Sri Sai Charith Grandhi, Rajesh Gupta, Sumit Mamoria Published: 2026-09-16 Categories: cs.AI, cs.MA arXiv: https://arxiv.org/abs/2609.18820v1 Agentic workflows now make consequential decisions in regulated settings, and the governance placed around them is almost entirely step-scoped: input-output classifiers, per turn rails, and span-level evaluators. The policies organizations actually hold, such as referral thresholds, authority limits, and review requirements, are properties of the whole execution rather than of any one step. This mismatch admits a failure mode we call a Compositional Policy Violation (CPV): every individual step passes its own check while the composed execution violates the governing policy. A predicate over a single step cannot evaluate a property that step does not determine, so no improvement in the accuracy of the step-scoped monitors detects this class. We define CPVs as the failure of step-level compliance to compose, and present a taxonomy of four types: Authority Creep, Threshold Laundering, Cumulative Sum Violation, and Context Collapse. We show that the correct repair for each class is dictated by where the guarded quantity mutates. We then introduce a provenance-aware runtime architecture that evaluates policies over complete execution traces, recomputing guarded quantities from raw provenance rather than the pipeline's derived representation. --- ### WaveTLM: Reliable Time-Series Language Modeling through Task Compilation Authors: Jiahui Chen, Bingke Zhu, Hongyu Pan, Yingying Chen Published: 2026-09-16 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.18812v1 Time-series language models provide a shared natural-language interface across temporal tasks, but plausible text does not guarantee reliable task outputs. Responses may appear reasonable while hallucinating the required object: numerical sequences can violate shape, scale, channel order, or temporal alignment, and textual decisions can fall outside the legal label space. We formulate reliable time-series language modeling, separating task-object reliability from predictive quality. We introduce ExecTS-QA, a contract-grounded benchmark spanning forecasting, imputation, classification, anomaly detection, and waveform analysis. We further propose WaveTLM, a unified compiler-executor model whose task compiler transforms user requests, visible arguments, and wave-grounded evidence into typed task states, while task-native executors construct numerical tensors, legal decisions, or structured records. On ExecTS-QA, a single WaveTLM checkpoint achieves 99.40% contract-valid coverage, compared with 37.83% for the strongest evaluated string-first baseline, while retaining balanced predictive performance across all five task families. Evaluations on SciTS, TSQA, IRTS-ToolBench, and ARFBench provide additional evidence of transfer. The code, construction scripts, and ExecTS-QA dataset will be publicly released upon publication. These results show that task compilation can convert plausible language generation into reliable time-series outputs. --- ### ProgramDistill: From Interactive Web Apps to Verifiable Reference-Guided SWE Tasks Authors: Jeonghye Kim, Minseon Kim, Young Jin Kim, Matheus Pereira, Marc-Alexandre Côté, Alessandro Sordoni, Xingdi Yuan, Zhengyan Shi Published: 2026-09-16 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2609.18805v1 Coding agents are typically evaluated with desired behavior specified through issues or instructions. In practical web development, however, agents may need to infer behavior from working software and implement it in an incomplete application. We introduce ProgramDistill, a benchmark evaluating coding agents on features discovered through interaction with fully functional reference applications. We build ProgramDistill by factorizing applications into features of different granularities, each associated with replayable behaviors executable via its gold patch. Our pipeline, mine-craft-patch, discovers 1,975 replay-verified behaviors across 26 applications and constructs 4,063 tasks without human intervention. Across nine frontier coding agents, GPT-6 Astra and Claude Opus 5 achieve 49.2% and 28.8% success on cumulative workflows in full-application reconstruction. In partial-application reconstruction, success falls from 100% to 64.0% and from 96% to 32% as restoration depth increases from 1 to 8. ProgramDistill thus provides a scalable benchmark with controlled difficulty for evaluating and diagnosing coding agents, and a natural basis for future curriculum-based training. --- ### Beyond frequency measures: Can contextual embeddings capture meaning change in scientific texts? Authors: Jianying Liu, Kim Gerdes, Jean-Marc Deltorn Published: 2026-09-16 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.18804v1 Identifying technological trends is a core scientometric task, yet traditional frequency-based approaches struggle to capture substantial meaning shifts of domain-specific terms. We hypothesise that contextual embeddings can complement frequency dynamics to effectively track diachronic semantic change. We compare frequency and embedding-based approaches across Astrophysics and NLP corpora spanning from 2010 to 2024. Candidate terms are extracted using KeyBERT (utilizing SciBERT as its underlying language model) and filtered for significant frequency increases using Fisher's exact test. These terms are then evaluated for genuine semantic shift by domain experts to establish ground-truth labels. To quantify semantic drift, each term's contextual embedding ''clouds'' from the two discrete periods are compared using multiple metrics: cosine distance, average pairwise distance, Hotelling-type T 2 , and maximum mean discrepancy. Results indicate that frequency-based methods align slightly better with human judgments of ''trend-related terms'' than semantic metrics (Precision@50 of 0.62 vs 0.60 in Astrophysics). The two signals show a correlation of around 0.6. Several terms identified exclusively by embedding metrics (e.g., ''primordial black holes'') represent critical conceptual developments invisible to pure frequency analysis. These findings indicate that semantic metrics may capture complementary information, highlighting the value of integrating contextual embeddings into scientometric trend analysis. --- ### A Convergence Framework for Deep $V$-Learning: Error Propagation and Sharp Action-Gap Bounds Authors: Yury Kolomeytsev Published: 2026-09-16 Categories: cs.LG, cs.RO, math.OC arXiv: https://arxiv.org/abs/2609.18782v1 We establish convergence bounds for deep $V$-learning with horizon $H$. The algorithm fits a scalar value function to targets from executed transitions and selects actions using a predictive model and the value function. For current observed-successor targets with fresh true-kernel outcomes, the conditional mean is $\mathcal{T}^βV$, which averages over behavior-policy actions. The Bellman optimality update is $\mathcal{T} V$. We decompose the update error into six residuals: fitting, transition reuse, target construction, replay, action selection, and exploration. Under $L^s$ concentrability, their $L^p$ norms ($p=s/(s-1)$) control expected $L^1$ policy loss. The bound explicitly weights residuals from only the last $H-1$ update blocks, plus an initialization term for shorter runs. We quantify the cost of a shared sampling distribution across horizon levels. For statistical error bounds of order $n^{-ν}$, we derive optimal continuous allocations and an integer allocation whose objective is within a factor $2^ν$ of the constrained optimum. A margin condition with exponent $α$ gives action error of order $Λ^{1+α/p}$, where $Λ$ combines network drift and score error; a one-step construction proves the exponent sharp. Bounds on the distance between frozen and optimal scores transfer an optimal-gap condition to frozen-iterate gap bounds while retaining the mass of optimal ties. Survival probabilities and coverage conditions at deployment yield bounds for policies selected with approximate scores. Separate spatial ReLU networks per horizon level give a conditional neural regression rate, and the finite-state case gives a log-free expected fit rate. These results give expected policy-loss consistency for the fixed-horizon generative-reset approximate-ERM procedure with exact action scores and provide an explicit residual-decay criterion for FIFO/interleaved SGD. --- ### CERA-MoA: Co-Evolving Routing Mechanisms with Continually Learning LLM Agents Authors: Jiaxuan Jiang, Liyuan He, Zhixuan Fang Published: 2026-09-16 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.18779v1 Current Mixture-of-Agents (MoA) paradigms generally treat query routing and agent fine-tuning as separate processes, limiting their ability to respond to evolving agent capabilities. This disconnect prevents routing strategies from adapting to evolving agent capabilities during post-training and prevents agents from achieving synergistic data-driven specialization. To resolve this, we introduce CERA-MoA (Co-Evolving Router with continually learning Agents for Mixture-of-Agents), an iterative reinforcement learning framework where the dynamic router and independent agent policies co-evolve. We design a predictive familiarity estimator that leverages mid-layer hidden states to evaluate semantic competence among agents, avoiding the overhead of full rollouts. Based on these familiarity scores, a cumulative-threshold adaptive routing mechanism dynamically activates a tailored minimal agent subset, achieving a trade-off between task performance and efficiency. By proactively allocating targeted training samples to agents based on their evolving competence, CERA-MoA promotes capability differentiation. Extensive experiments across various domains demonstrate that CERA-MoA outperforms state-of-the-art static-agent routing and fix-workflow fine-tuning baselines. --- ### DISTA-Net++: Rethinking Infrared Small Target Unmixing Beyond Sub-Pixel Separation Authors: Mengze Xu, Zhu Liu, Weidong Sheng, Boyang Li, Yimian Dai, Ming-Ming Cheng, Jian Yang Published: 2026-09-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.18773v1 Long-range infrared imaging frequently confronts dense target clusters whose diffraction-limited signatures merge into a single indistinguishable blob, concealing the number, sub-pixel positions, and radiant intensities of the underlying sources. While deep learning has advanced general object detection, resolving such Closely-Spaced Infrared Small Targets (CSIST) remains largely unexplored, owing to a systemic infrastructure void and a fundamental paradigm mismatch. The dominant formulation, which reduces unmixing to a blind, discrete sub-pixel separation, is inherently insufficient: without semantic guidance, the ill-posed inverse problem admits ambiguous solutions plagued by false and missed detections, while grid-based discretization locks predictions onto fixed lattice centers, chaining precision to prohibitively expensive grid refinement. We argue that CSIST unmixing should instead be informed and continuous. To ground this paradigm shift, we establish the first comprehensive open-source ecosystem for the field, comprising the large-scale CSIST-100K benchmark, a tailored metric suite, and the GrokCSO toolkit. Upon this foundation, we propose DISTA-Net++, which anchors a dynamic deep unfolding backbone with two synergistic mechanisms: a Count-Guided Prior that injects the global target count as an explicit semantic constraint to regularize the solution space, and a Continuous Coordinate Rectification that regresses off-grid offsets to decouple localization accuracy from grid resolution. Extensive experiments validate our paradigm: even under the most economical 3x division, DISTA-Net++ surpasses 7x-division state-of-the-art methods by 16.15% in CSO-mAP and 62.96% in count accuracy at merely one-sixth of their computation, demonstrating that unmixing precision need not be purchased with finer discretization. The complete ecosystem is available at https://github.com/GrokCV/GrokDet. --- ### Zero-Shot Cross-Lingual Recognition of Sign Language Handshapes Authors: Marcel Granero-Moya, Carolina del Corral Farrarós, Gloria Haro, Coloma Ballester, Ricardo Marques Published: 2026-09-16 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.18772v1 Sign language processing advances rapidly for high-resource languages such as American Sign Language (ASL), yet most of the world's sign languages lack the phonological annotations new methods require. We present the first zero-shot cross-lingual framework for handshape recognition, transferring from ASL to Catalan Sign Language (LSC). Our approach leverages the decomposition of handshapes into five phonological features -- selected fingers, flexion, spread, thumb position, and thumb contact -- shared across both languages, to decode LSC handshapes from predicted features via a composite phonological distance metric. We evaluate three architectures (MLP, SL-GCN, SHuBERT) trained on two ASL corpora (PopSign, Sem-Lex) against a 37-handshape, single-signer LSC benchmark. Zero-shot transfer proves viable once recording-format disparities are harmonized, reaching 80.0% phonological feature accuracy and 54.5% expected handshape accuracy. Phonological decomposition thus offers a bridge for extending sign language technologies to low-resource languages without any target-language video training labels. --- ### Version- and Scope-Aware Question Answering over Normative Documents: A Deployed System and an End-to-End Evaluation at Production Scale Authors: Liuyin Wang, Shuaipeng Jin, Jiwei Shi, Jensen Hsu Published: 2026-09-16 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.18769v1 Correctly answering a question grounded in normative documents often depends on information outside any single passage: whether the retrieved document is the version currently in force; whether it applies to the jurisdiction, subject (such as an institution or applicant), and date at issue; and whether each normative claim can be traced to its supporting source text. Hosted retrieval services have substantially lowered the engineering cost of building an initial system over such corpora, making "upload the documents and ask" a common default. We evaluate this default on approximately 73,000 candidate normative documents supplied to a production deployment. The evaluation uses a stratified sample of 200 questions from our published benchmark, with a gold source document for every question; the released sampling rule reads no system outputs or scores. We compare the hosted service with a governed system that resolves version and scope through explicit rules before generation. The governed system scored 97.7 overall, while the hosted service scored 88.1, a gap of 9.6 points computed from unrounded means. The question set, the answer text evaluated for both systems, the scores, and the scripts used to reproduce the reported benchmark statistics are public. The governed configuration has operated as a commercial product since January 2026 and serves 1,126 registered users; named customer organizations include Zhipu AI and Lecheng Health. By mid-April 2026, it had reached roughly 100,000 calls per workday. --- ### FRAUDSkill: Structured Frozen-Weight Skill Optimization for Audio Anti-Fraud Detection Authors: Chengxian Hu, Zhiming Ma, Mingjun Pan, Yifan Wang, Shun Zhang, Qifan Wang, Zhilei Zhao, Yijin Zhou, Yuxi Zhao, Huiyuan Liu, Peidong Wang, Peng Chen Published: 2026-09-16 Categories: cs.SD, cs.CL arXiv: https://arxiv.org/abs/2609.18766v3 Large audio-language models have shown promise for anti-fraud detection by directly processing speech and reasoning over fraud-related evidence. Their deployment, however, requires predictions to follow a predefined label space and a structured decision protocol consisting of service-scenario identification, fraud detection, and conditional fraud-type classification. Existing fine-tuning and prompt-based approaches typically encode task knowledge, constraints, and decision rules into model parameters or manually maintained prompts, making them difficult to adapt as fraud patterns and labeling policies evolve. To this end, we propose FRAUDSkill, a structured frozen-weight adaptation framework that leaves the underlying audio-language model unchanged while optimizing an external layer of skill programs, route-specific policies, and decision rules. We further combine structured output control with validation-guided multi-path inference to ensure protocol-compliant predictions. On the TeleAntiFraud benchmark, FRAUDSkill achieves 73.50% Macro-F1, outperforming the shared frozen-model baseline by 31.96% while reducing invalid outputs to 1.94%. Extensive experiments demonstrate that external skill optimization provides an effective and adaptable solution for structured audio anti-fraud detection without modifying the underlying model. The source code is available at https://anonymous.4open.science/r/FRAUDSKILL-114514. --- ### FRAUDSkill: Structured Frozen-Weight Skill Optimization for Audio Anti-Fraud Detection Authors: Chengxian Hu, Zhiming Ma, Mingjun Pan, Yifan Wang, Shun Zhang, Qifan Wang, Zhilei Zhao, Yijin Zhou, Yuxi Zhao, Huiyuan Liu, Peidong Wang, Peng Chen Published: 2026-09-16 Categories: cs.SD, cs.CL arXiv: https://arxiv.org/abs/2609.18766v1 Large audio-language models have shown promise for anti-fraud detection by directly processing speech and reasoning over fraud-related evidence. Their deployment, however, requires predictions to follow a predefined label space and a structured decision protocol consisting of service-scenario identification, fraud detection, and conditional fraud-type classification. Existing fine-tuning and prompt-based approaches typically encode task knowledge, constraints, and decision rules into model parameters or manually maintained prompts, making them difficult to adapt as fraud patterns and labeling policies evolve. To this end, we propose FRAUDSkill, a structured frozen-weight adaptation framework that leaves the underlying audio-language model unchanged while optimizing an external layer of skill programs, route-specific policies, and decision rules. We further combine structured output control with validation-guided multi-path inference to ensure protocol-compliant predictions. On the TeleAntiFraud benchmark, FRAUDSkill achieves 73.50% Macro-F1, outperforming the shared frozen-model baseline by 31.96% while reducing invalid outputs to 1.94%. Extensive experiments demonstrate that external skill optimization provides an effective and adaptable solution for structured audio anti-fraud detection without modifying the underlying model. The source code is available at https://anonymous.4open.science/r/FRAUDSKILL-114514. --- ### Stable Filters for Generative Modeling of Graph Signals Authors: Martin Schmidt, Gonzalo Mateos Published: 2026-09-16 Categories: eess.SP, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.18759v1 Generating signals on graphs requires permutation-equivariant models that exhibit stability with respect to relative structural perturbations. While recent graph-aware Schrödinger bridge models incorporate topology information directly into their reference dynamics, it is unclear how perturbations of the graph propagate through these dynamics and affect the resulting generated distributions. In this paper, we analyze the structural stability of graph-aware continuous-time generative models whose drift combines a graph filter with a learned graph neural network. We derive explicit Wasserstein stability bounds that quantify the effect of relative graph perturbations on the generated distributions. Motivated by these bounds, we introduce a principled framework for designing stable graph filters that preserve the smoothing behavior of graph heat diffusion, while boosting structural stability. Experiments on synthetic and fMRI signals show our stable filters enhance structural robustness while matching or exceeding the generative quality of the heat equation baseline. --- ### Toward Markerless Video-based Tremor Analysis: Objective Quantification of Pathological Tremor in Mouse Preclinical Models Authors: Yota Koshimoto, Akihiro Tsukahara, Yasuhiro Moriwaki, Mariko Isogawa Published: 2026-09-16 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.18753v1 Tremor is a movement disorder characterized by involuntary, rhythmic oscillations of body parts and is a hallmark of several neurological conditions, including Parkinson's disease and essential tremor. Elucidating its underlying mechanisms relies heavily on mouse models, which offer genetic manipulability and translational relevance to human neural circuitry. Accordingly, these models are indispensable for studying tremor pathophysiology. So far, electromyography and accelerometers have been used as methods to quantitatively observe tremors in mice. However, these methods have several drawbacks, such as high costs and complex setups. In particular, the invasive surgical implantation of devices causes significant stress to the animals. Although RGB-based methods offer non-invasive and cost-effective alternatives, they often lack the sensitivity required to detect subtle tremors. Therefore, this paper addresses these challenges by achieving mouse tremor severity estimation using conventional RGB cameras only. To address the challenging task of isolating tremor-related vibrations while the mouse itself is also in motion, our pipeline incorporates segmentation-based pre-processing to extract the mouse region and a Tremor Score Estimation Module that captures subtle tremors with high sensitivity. In the experiments, we assessed tremors in unrestrained mice using a non-invasive method with two standard cameras. The results demonstrated a strong correlation with accelerometer measurements and confirmed that the method accurately captured the intensity-dependent characteristics of tremors. The project page is available at https://isogawalab.github.io/Video-based-Tremor-Analysis-Project/. --- ### TeleAntiFraud 2.0: A Refreshable, Profile-Grounded, and Audio-Based Benchmark for Telecom Fraud Detection Authors: Huiyuan Liu, Zhiming Ma, Yanxing Liu, Shun Zhang, Qifan Wang, Di Liu, Yifan Wang, Yuyang Deng, Haoyang Meng, Yijin Zhou, Yuxi Zhao, Chengxian Hu et al. Published: 2026-09-16 Categories: cs.SD, cs.CL arXiv: https://arxiv.org/abs/2609.18748v2 Telecom fraud scripts evolve rapidly and are often designed to resemble routine service conversations, creating two key requirements for audio-based telecom-fraud evaluation. First, benchmarks must incorporate newly observed scam patterns without overwriting previously established test sets. Second, they must distinguish fraud from lawful, near-domain calls rather than relying on topic-separated negative examples. We present TeleAntiFraud 2.0, constructed with our Mixed-Tree Anti-Fraud Generation Pipeline and evaluated under a monthly frozen evaluation protocol. The pipeline transforms online fraud-case abstracts into profile-grounded scenarios, expands them through mixed-tree generation, realizes fraud and non-fraud dialogue paths under shared contexts, renders validated dialogues as role-matched speech, and freezes the resulting audio, labels, prompts, manifests, and provenance records for each monthly evaluation set. Each frozen set contains 900 Chinese calls, comprising 600 fraud and 300 near-domain non-fraud cases. Controlled text experiments show that three classifiers achieve perfect macro-averaged F1 (Macro-F1) when evaluated against unrelated or ordinary negatives, but drop to 0.65-0.68 with near-domain sibling negatives. Full-set audio and automatic-speech-recognition plus large-language-model (ASR+LLM) evaluations further reveal class-prior shortcuts, prediction collapse, and snapshot sensitivity. Together, these findings establish near-domain construction and collapse-aware reporting as core requirements for evaluating audio-based telecom-fraud models under realistic confusable conditions. The accompanying research artifact includes the construction code, evaluation scripts, manifests, and documentation. Our dataset and code are available at https://anonymous.4open.science/r/TeleAntiFraud-2_0-EEB2/. --- ### TeleAntiFraud 2.0: A Refreshable, Profile-Grounded, and Audio-Based Benchmark for Telecom Fraud Detection Authors: Huiyuan Liu, Zhiming Ma, Yanxing Liu, Shun Zhang, Qifan Wang, Di Liu, Yifan Wang, Yuyang Deng, Haoyang Meng, Yijin Zhou, Yuxi Zhao, Chengxian Hu et al. Published: 2026-09-16 Categories: cs.SD, cs.CL arXiv: https://arxiv.org/abs/2609.18748v1 Telecom fraud scripts evolve rapidly and are often designed to resemble routine service conversations, creating two key requirements for audio-based telecom-fraud evaluation. First, benchmarks must incorporate newly observed scam patterns without overwriting previously established test sets. Second, they must distinguish fraud from lawful, near-domain calls rather than relying on topic-separated negative examples. We present TeleAntiFraud 2.0, constructed with our Mixed-Tree Anti-Fraud Generation Pipeline and evaluated under a monthly frozen evaluation protocol. The pipeline transforms online fraud-case abstracts into profile-grounded scenarios, expands them through mixed-tree generation, realizes fraud and non-fraud dialogue paths under shared contexts, renders validated dialogues as role-matched speech, and freezes the resulting audio, labels, prompts, manifests, and provenance records for each monthly evaluation set. Each frozen set contains 900 Chinese calls, comprising 600 fraud and 300 near-domain non-fraud cases. Controlled text experiments show that three classifiers achieve perfect macro-averaged F1 (Macro-F1) when evaluated against unrelated or ordinary negatives, but drop to 0.65-0.68 with near-domain sibling negatives. Full-set audio and automatic-speech-recognition plus large-language-model (ASR+LLM) evaluations further reveal class-prior shortcuts, prediction collapse, and snapshot sensitivity. Together, these findings establish near-domain construction and collapse-aware reporting as core requirements for evaluating audio-based telecom-fraud models under realistic confusable conditions. The accompanying research artifact includes the construction code, evaluation scripts, manifests, and documentation. Our dataset and code are available at https://anonymous.4open.science/r/TeleAntiFraud-2_0-EEB2/. --- ### When Edit Flows are Edit Jumps: replicating Edit Flows and EvoFlows Authors: Gabriel Bénédict, Melanie Buechler, Gerard Riera-Solà, Chloé de Ancos, Yves Gaetan Nana Teukam, Moritz Freidank Published: 2026-09-16 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.18745v1 Antibody lead optimization calls for a small, bounded set of edits to an existing candidate: substitutions, but also insertions and deletions. Edit-based generative models are the only ones that allocate such an edit budget without fixing the edit positions, the edit count, or the output length in advance. However, the existing approaches Edit Flows and EvoFlows did not release code or complete training specifications. Here, we show that both methods follow the same underlying process -- edits firing one at a time, at learned rates, in continuous time -- the pure-jump case of generator matching over finite sequences. With EditJumps we introduce the first open implementation of this framework, with a single generalist antibody editor trained on 1.66M Observed Antibody Space homolog pairs to propose homolog-like variants of a seed sequence, editing unseen leads zero-shot, without the per-family retraining original approaches require. Replicating this system from scratch exposes why open code is essential for generative biology: reconciling published edit distributions required reverse-engineering an undocumented rate-scaling hyperparameter that dictates realized mutation counts. Moreover, we show that published evaluation metrics are highly sensitive to reference sample size, frequently flipping method rankings. We release our full codebase, automated test suite, and configurations at: https://github.com/VisiumCH/editjumps --- ### A Scalable Framework for Automated NER Annotation Correction in Low-Resource Languages Authors: Toqeer Ehsan, Thamar Solorio Published: 2026-09-16 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.18739v1 Poor quality or noisy annotations in Named Entity Recognition (NER), as in any other NLP task, make it challenging to achieve state-of-the-art performance. In this paper, we present a multi-step framework to enhance the annotation quality of NER datasets by employing automated techniques. We propose a frequency-based iterative approach that leverages self-training and a dual-threshold mechanism to enhance inference confidence. Experimental evaluations on different NER datasets demonstrate significant improvements in NER performance with respect to the original datasets. This work further explores the potential of generative Large Language Models (LLMs) to perform NER for low-resource languages. ---