# Research Papers — 2026-09 · bundle page 47 of 266 > Full-text markdown of papers #2301–2350 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/47 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/46.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/48.md) ### LLN: Learnable Lens Networks for Parameter-Efficient Long-Horizon Dynamical Prediction Authors: Binbin Yong, Zhao Su, Lan Guo, Haoran Li, Jun Shen, Qingguo Zhou Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34493v1 Explicit residual connections of the form (x+f(x)), often combined with normalization layers, have become a standard strategy for training very deep neural networks. However, residual addition primarily provides an algebraic shortcut for gradient propagation, while leaving the evolution of feature geometry across layers largely unconstrained. We introduce Learnable Lens Networks (LLN), a physics-inspired architecture that replaces direct feature-space residual accumulation with learnable optical transport in an augmented position-angle phase space. Each layer alternates between free propagation, which provides an implicit transport path, and a learnable lens field that performs nonlinear trajectory transformation and focusing. Theoretically, we establish that LLN transport is globally invertible and volume-preserving for any differentiable lens field, with the implemented coordinate-wise Gaussian transport further satisfying symplecticity. Importantly, these structural constraints do not limit expressivity: with unrestricted embeddings and readouts, LLN retain universal approximation of continuous end-to-end maps. Experiments across diverse dynamical systems demonstrate that LLN improves long-horizon prediction while using substantially fewer parameters than same-depth comparators. Further analysis reveals stable depth-wise gradient transport and interpretable learned dynamics under the coupled propagation and refraction design. --- ### PowerBench: A Benchmark for Agentic Retrieval and Reasoning in Power Systems Authors: Xijing Wang, Yinsheng Yao, Jinru Ding, Yidong Jiang, Ziwen Xu, Yiwen Jiang, Jie Xu, Dawei Cheng Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34492v1 Large language model (LLM) agents offer new opportunities for automated analysis in industry. However, rigorous evaluation of such agents-for example, within power system scenarios-remains hindered: real operational data are confidential, and existing public resources fail to fully capture the chained dependencies and heterogeneous evidence. To address this gap, we propose PowerBench, comprising (1) a generation framework that derives interconnected heterogeneous operational data through a common dependency chain, and (2) a synthetic dataset generated by this framework. The dataset covers 761 devices across 100 device types, with 13.35 million hourly telemetry records spanning two years and 24,939 operational documents. Building on this dataset, we construct 300 questions across three task families that evaluate frontier LLMs' ability to complete analysis tasks that require autonomous evidence retrieval and reasoning across interconnected and heterogeneous data under restricted tool calls and time budgets. Results demonstrate that the evaluated frontier LLMs remain challenged on these tasks: the best model reaches only 74.2% joint accuracy. Our trace analysis further reveals that model performance varies across evidence discovery, content retrieval, tool use, reasoning over evidence, and answer submission. These findings provide detailed insights for evaluating LLM agents and guiding their reliable deployment in industry. The framework, dataset, and benchmark tasks are available at https://github.com/open-compass/PowerBench. --- ### M3OS: A Monte Carlo Graph Search-Orchestrated Multi-Agent LLM System for Evidence-Traced Molecular Optimization Authors: Junjie Wang, Yaowei Jin, Ruohui Tang, Guonan Cui, Haojie Wang, Penglei Wang, Dingyan Wang, Duo An, Shuangjia Zheng, Qian Shi Published: 2026-09-28 Categories: cs.LG, q-bio.BM arXiv: https://arxiv.org/abs/2609.34491v1 Small-molecule optimization integrates medicinal-chemistry reasoning and computational evidence through iterative, multi-objective decisions. When large language models (LLMs) reason over optimization histories stored primarily in conversational context, they must recover candidate identities, prior evaluations, and task constraints to guide subsequent decisions. We present M3OS, a multi-agent LLM system that decouples molecular-design reasoning from optimization-state management through Monte Carlo graph search. A persistent graph links evaluated candidates, parent-child transformations and evaluation evidence, while rewards and visit statistics guide LLM-assisted parent selection. Two branches combine tool-driven candidate generation with knowledge- and case-guided medicinal-chemistry editing. An execution harness controls graph updates through structured output extraction, molecular validation and task-bound evaluation. Agents receive role-specific contexts, while the graph preserves optimization trajectories beyond their active contexts. Across three molecular optimization benchmarks, M3OS achieves higher success rates than baselines, supporting the integration of persistent search state, specialized agents and controlled execution for multi-constraint optimization. --- ### HPMD: A Historical Persian Manuscript Dataset for Word Spotting with Line-Level Annotation Authors: Saeid Firouzi Daghigh, Majid Iranpour Mobarakeh Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34490v1 Large collections of historical Persian manuscripts have been digitized, but searching them is still slow and mostly manual. Historians usually want to find where a specific name, date, event, or topic appears, which is a word spotting problem. Progress on this task is limited by two things. First, there is almost no public dataset of historical Persian handwriting; the only notable resource, OpenITI MAKHZAN, contains a relatively small Persian portion. Second, word spotting models usually need word-level bounding boxes, which are very expensive to annotate. In this paper we introduce a new dataset of 223 pages, 3,678 lines, 37,631 words, and 130,630 characters, collected from diverse historical Persian books of poetry and prose and annotated at the region, line, and text level. We also propose a baseline that is trained only with line-level annotations but returns word-level locations. A fine-tuned line detector finds text lines, and a fine-tuned CRNN recognizer trained with CTC produces a frame-by-character posterior matrix for each line. Instead of decoding the most probable character at each frame, the query is scored directly against this matrix, so visually similar characters in Persian such as be and pe no longer cause hard failures. The frame alignment also gives the horizontal position of the word inside the line. On the test set, the fine-tuned line detector reaches an F1 of 0.892, and posterior-based search raises the word spotting F1 from 0.487 to 0.558 compared with exact matching on the decoded text, with the decision threshold selected on a held-out validation set. A PHOC attribute-embedding baseline that additionally receives oracle word boundaries at test time reaches an F1 of 0.449, below the proposed method. We also report a distributional analysis of the dataset, a taxonomy of retrieval errors, and a per-conditionbreakdown of performance. --- ### When Less Data Favors Smaller Teachers: Rethinking Teacher Capacity and Data Selection for Knowledge Distillation Authors: Minjae Park, Taesun Yeom, Jaeho Lee Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34489v1 Data pruning reduces the training cost of knowledge distillation (KD). However, the preferred teacher capacity changes with the data budget: smaller teachers can outperform larger ones when limited training data are available. Understanding what drives this shift is important not only for teacher choice but also for identifying which samples are useful for distillation. We analyze teacher supervision by decomposing it into relational ordering---the ranking of classes---and score geometry---the magnitudes and margins of class probabilities---and show that the small-teacher advantage in the low-data regime arises not only from score geometry but also from relational ordering. Beyond understanding teacher capacity, our analysis reveals two properties of effective subsets: samples should match the difficulty appropriate for the available budget, and their relational signals should be diverse rather than redundant. Based on these findings, we propose DVA (Difficulty- and Volume-Aware data selection for KD), a training-dynamics-free method, which uses a small teacher as a proxy for budget-aware difficulty filtering and class-conditional relational volume maximization. Despite requiring no training dynamics statistics, our method remains competitive with training-dynamics-based methods while consistently outperforming training-dynamics-free baselines. --- ### FlexLoop: Depth-Elastic Looped Policies for Adaptive Test-Time Computation in Deep RL Authors: Xun Wang, Ruishuo Chen, Yu Chen, Zhuoran Li, Longbo Huang Published: 2026-09-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.34488v1 Looped architectures scale computation by reusing the same parameters across recurrent steps, and recent work shows that they substantially improve deep reinforcement learning policies on long-horizon tasks. Since recurrent depth directly controls computation, one may expect looped policies to naturally support elastic inference across recurrent depths. Surprisingly, we find that pretrained looped policies exhibit severe recurrent-depth specialization: reliable decisions are concentrated near the full trained depth, tying deployment computation to this depth even when less computation may suffice. Achieving depth elasticity, i.e., reliable decisions across recurrent depths with adaptive computation at deployment, therefore remains a key challenge. To address this, we propose FlexLoop, a novel post-training framework that converts pretrained fixed-depth looped policies into depth-elastic policies. FlexLoop keeps training on the original RL objective to preserve full-depth capability while performing adjacent-depth policy distillation to progressively transfer decision quality from deeper to shallower recurrent steps. The resulting policy supports reliable inference across recurrent depths and enables state-wise adaptive inference through recurrent-depth consistency. Experiments on $30$ online and offline long-horizon goal-conditioned environments show that FlexLoop preserves full-depth performance while making shallower depths effective. Keeping competitive performance, FlexLoop reduces average recurrent depth by up to $\bf{43\%}$ and achieves up to $\bf{1.34\times}$ wall-clock speedup in a stress test. --- ### PACER: Progressive Availability-Conditioned Evidence Routing for Radiology Report Generation under Incomplete Clinical Context Authors: Yulong Chen, Yadong Liu, Haoyu Cao, Sen Xu, Yueying Wang, Jie Wen Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34487v1 Radiology report generation (RRG) increasingly incorporates heterogeneous clinical evidence, such as multi-view radiographs and previous reports, whose availability varies across examinations. However, accommodating different input combinations does not ensure effective evidence use: generated reports may still omit or inaccurately describe clinically relevant findings. To address this problem, we propose PACER, a Progressive Availability-Conditioned Evidence Routing framework for structured incomplete-context RRG that follows a Refine-Calibrate-Commit pipeline. It first refines observed visual representations through endpoint-preserving patchwise routing across frozen encoder depths, incorporating complementary cues while retaining the pretrained terminal representation. It then calibrates the language-model prefix according to the observed evidence and availability state, adapting the shared generator's conditioning as the available source set changes. Finally, it generates polarity-structured clinical commitments before the report in the same autoregressive trajectory, providing structured clinical context for subsequent generation. Experiments demonstrate state-of-the-art clinical efficacy across all four MIMIC-RG4 settings and strong MIMIC-CXR performance, while maintaining competitive language-generation quality. --- ### ARS: Agentic Reward System for Robot Learning Authors: Sheng Hu, Weiyi Lu, Lingbing Zeng, Gan Weng, Weiwei Zhang, Kai Xie, Xiaofeng Mou, Yi Xu Published: 2026-09-28 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2609.34484v1 Progress reward modeling is the problem of estimating how a robot's behavior changes task progress over time. Reliable estimation requires distinguishing meaningful state changes from failed attempts and task-irrelevant actions. We introduce the Agentic Reward System (ARS), an inference framework for progress reward modeling with general-purpose vision-language models (VLMs), without additional reward-model training. Given an offline trajectory and a task instruction, ARS uses adaptive visual inspection for both event proposal and verification. A subagent proposes a task-relevant event timeline, which a primary agent verifies and revises before estimating per-frame progress. ARS can incorporate optional terminal outcome labels and visual references to inform its judgments. It can also audit progress estimates from external reward models. We evaluate ARS with a 27B VLM on a controlled semantic-mismatch benchmark and downstream policy learning in simulation and on a real robot. The benchmark reveals that several evaluated reward baselines assign spurious progress to wrong-object manipulation even in simple pick-and-place scenes. ARS better suppresses these errors and outperforms these baselines in simulation policy learning. We further demonstrate that ARS supports long-horizon policy learning from mixed-quality offline experience on real-robot multi-screw fastening in a full-scale laboratory replica of an industrial washing-machine assembly line. These results suggest that structured inference and verification can improve the usefulness of general-purpose VLMs for robot reward modeling. Code is at https://github.com/midea-ai/ars --- ### CARDAMOM: A Micro-Dialectal Arabic Speech Dataset for ASR Authors: Bashar Talafha, Samar M. Magdy, Aisha Alansari, Alaa Alkhawaldeh, Abdurrahman Juma, Sharaf Makahleh, Nour Gamal, Omar Attia, Hanaa Kurdi, Najwa Rizk, Maysa Anaya, Hessah Altimyat et al. Published: 2026-09-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.34481v1 We present Cardamom, a micro-dialectal Arabic speech dataset designed to support fine-grained evaluation and adaptation of automatic speech recognition (ASR) systems. Community-curated by native speakers familiar with the represented varieties, Cardamom contains approximately 40 hours of transcribed YouTube speech spanning 21 micro-dialects across Egypt, Jordan, Lebanon, Mauritania, Palestine, and Saudi Arabia. Each segment is annotated with one or more operational micro-dialect labels, code-switching information, and utterance-level perceived gender, enabling analysis of sub-country variation that is obscured by conventional country-level labels. We describe the collection and annotation process, motivate the micro-dialect inventory linguistically, and benchmark four multilingual ASR systems in zero-shot and adapted settings. The strongest zero-shot system obtains 43.47% aggregate WER, with particularly high error rates on Mauritanian and Lebanese varieties; adaptation on Cardamom reduces its WER to 35.21%. Audio-based identification experiments further show that the annotations provide a learnable prediction target, with a dedicated classifier reaching 85.57% accuracy on 21-way micro-dialect identification. Cardamom provides a resource for studying localized dialectal variation and developing Arabic speech systems with broader regional coverage. --- ### When Does an Image Determine the Answer? Benchmarking Visual Answerability across Charts and Scenes Authors: Sungguk Cha, Mintae Kim, Youngsub Han, Byoung-Ki Jeon, Sangyeob Lee Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34480v1 Reliable visual question answering requires correct answers when evidence is sufficient and abstention when it is not. We introduce a benchmark that connects complete-question evaluation with explicit evidence for its labels across PlotQA charts, CLEVR rendered scenes, and GQA photographs. Each question groups original and edited images, presented independently; success requires every supported answer and every required abstention to be correct. For chart missing-information labels, executable witnesses establish that admissible complete charts give different answers but identical pixels after masking. Scene labels follow source programs and edits, with a residual-cue analysis for photographs. Across 72,000 responses from six model configurations, the highest observed complete task success rates are 57.0%, 43.5%, and 33.7%, respectively. On charts, the strongest configuration achieves 96.2% per-view decision accuracy, yet 265 of its 835 groups with every decision correct still contain incorrect answers. Evaluating supported answers and necessary abstentions together exposes failures that answerability decisions alone conceal. --- ### SentZero: An Enhanced Sentence-Centric Vision-Language Pretraining for Multi-Task Zero-Shot Chest X-Ray Analysis Authors: Hangyul Yoon, Hyungyung Lee, Edward Choi, Eunho Yang Published: 2026-09-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.34479v1 Vision-language (VL) pretraining using paired chest X-ray (CXR) images and radiology reports has shown strong potential for medical image understanding. However, existing methods often remain dependent on task-specific finetuning because radiology reports are lengthy, clinically dense, and difficult to align with simple zero-shot prompts. Recent sentence-level approaches partially address this limitation using clinical phrases extracted by large language models (LLMs), but they largely overlook the intrinsic characteristics of radiology discourse. In particular, limited positive-pair diversity constrains further gains, while clinically equivalent sentences frequently recur across patients, creating false negatives in contrastive learning. To address these issues, we propose SentZero, an enhanced sentence-centric VL pretraining framework for zero-shot, multi-task CXR analysis. SentZero introduces LLM-based abstract-level sentence structuring and mapping to expand positive-pair diversity, together with an additional loss term to mitigate false negatives. We further introduce sentence-conditioned residual modulation of visual embeddings, enabling visual features to adapt to the semantic characteristics of each input sentence. Across diverse downstream tasks and datasets, SentZero improves zero-shot generalization and outperforms prior multi-task zero-shot methods. --- ### Learn Here, Move Less Elsewhere: Input-Conditioned Plasticity from Retained-Domain Activation Atlases Authors: Jiangtao Lin, Bangyang Wei, Yihang Ding, Siyi Liu, Yuhan Dong Published: 2026-09-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.34478v1 Task-specific fine-tuning can rewrite a language model's answers beyond the training task, complicating updates that must preserve existing behavior. We introduce ATLAS, which turns retained-domain representations into an input-dependent rule for task adaptation. An activation atlas supplies local reference centers and directional filters to a shared low-rank residual. Target supervision learns the residual, while retained geometry shapes its action throughout training and inference. On Qwen3-8B, ATLAS achieves lower mean retained-output Kullback-Leibler (KL) divergence than all seven published baselines at shared coding-performance requirements, with consistent advantages across multiple training seeds. Structural comparisons identify the contributions of retained reference states and directional conditioning, and answer-level analyses show fewer rewritten mathematical answers and more stable commonsense choices. Experiments spanning five backbones and two retained domains further demonstrate coding gains with reduced retained-output movement. With compact storage and modest decoding overhead, ATLAS provides a practical mechanism for acquiring specialized skills while maintaining continuity in existing responses. --- ### Causal Routing for Unlearning Authors: Bardh Prenkaj, Andrea D'Angelo, Davide Mottin, Federico Fontana, Davide Gabrielli, Paola Velardi, Stefano Faralli Published: 2026-09-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.34475v1 LLMs cannot forget the way we delete a file. Strangely, we are asked to remove something that was never put anywhere in particular. What the model took from a piece of text is now smeared across billions of weights. Existing methods rewrite all of them to change one thing, and none of them say which part produced that change. To address this, we introduce Causal Routing for Unlearning (CRU) by asking where the concept is expressed in the model and suppressing only that part. One untrained forward pass over the forget set ranks neurons by how their activations vary. Then, small routing modules on those neurons gate and suppress only the concepts that need to be forgotten. In CRU, the base model is frozen, and any change in behavior is caused only by the gated neurons; hence, why the routing is causal. Due to our parameter efficiency (only ~0.01% as many parameters as the base model), unlearning a concept costs 14 GiB, whereas the baselines require 71 GiB. On TOFU, CRU is indistinguishable from the retained model (p > 0.05, KS test) and is never Pareto-dominated, whereas every compared baseline matches its forgetting on the larger-forget batches only by collapsing utility. On RWKU, it achieves an adversarial-probe recall of 0.052, compared to 0.250 for the strongest baseline, meaning the knowledge is gone, not merely harder to reach. Thus, deciding on the intervention at query time, rather than fixing it beforehand, is the axis along which we argue that unlearning should proceed. --- ### Deep kernel hedging Authors: Jean-Loup Dupret, Donatien Hainaut, Edouard Motte Published: 2026-09-28 Categories: cs.LG, math.FA, q-fin.CP, q-fin.RM arXiv: https://arxiv.org/abs/2609.34474v1 We introduce a deep kernel hedging framework that combines the flexibility of deep learning with the structural inductive bias of kernel methods. The hedging functional is restricted to a reproducing kernel Hilbert space whose kernel is parameterized through a neural network embedding of the input features. The framework minimizes a regularized empirical risk under convex loss functions and can accommodate path-dependent information through truncated time-augmented signature features. We derive a generalized representer theorem for the joint hedging problem, reducing the empirical optimization to a finite-dimensional problem. To further reduce the computational cost associated with large kernel matrices, we develop a scalable random Fourier feature approximation and establish convergence guarantees. The random Fourier parameters are sampled once and remain fixed throughout training, while the deep kernel adapts to market data through the learned neural representation. We evaluate the performance of the proposed deep kernel approach on both synthetic and real data and compare it with standard kernel methods and classical deep hedging architectures. Numerical results indicate competitive and robust hedging performance, particularly in low-data regimes, which highlights the benefits of combining expressive neural representations with the inductive bias of kernel methods. --- ### VL-AcneSeg: A Vision-Language Framework for Region-Aware Acne Lesion Segmentation Authors: Sukju Oh, Soo Ick Cho, Dae Hun Suh, Sukkyu Sun Published: 2026-09-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.34472v1 Acne assessment is crucial for clinical decision-making, yet traditional grading and counting are subjective and fail to account for lesion size. While area-based assessment has emerged as a promising alternative, acne segmentation has continued to rely on general-purpose architectures. To address this gap, we propose VL-AcneSeg, a multimodal framework for acne lesion segmentation that leverages CLIP and region-level text prompts to incorporate spatial priors, enabling lesions to be localized across the whole face. Because region-level prompts indicate which facial areas contain lesions, we report a single global prompt, which requires no such information, as our primary setting. On our internal clinical dataset, VL-AcneSeg achieves a Dice score of 0.5082 and an IoU of 0.3407 under this protocol, the highest among all compared methods, including recent vision-language segmentation methods that are themselves given region-level prompts; region-level prompting raises these to 0.5296 and 0.3602. Moreover, lesion area measurements derived from our segmentation correlate with IGA scores at a level comparable to expert annotations (Pearson r = 0.719 versus 0.658). Notably, our framework maintains consistent performance across external validation datasets, performing reliably even on uncontrolled smartphone images without requiring additional training or fine-tuning. By pairing a protocol that requires no lesion-location information with area-based severity estimation, this work provides a foundation for objective acne assessment outside the clinic. Our implementation is publicly available at: https://github.com/sukjuoh/VL-AcneSeg --- ### Precise Editing and Flexible Referencing for Interactable Worlds Authors: Xinyao Liao, Xianfang Zeng, Zhu Liang, Zhoujie Fu, Qianxun Xu, Jiachi Liu, Gang Yu, Guosheng Lin Published: 2026-09-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.34470v1 We present EditWorld, a video world model for precise editing and flexible referencing in interactable worlds. Existing video world models primarily focus on navigation, letting users explore generated worlds but offering limited control over how existing world content is modified. EditWorld extends world modeling from exploration to precise modification by streaming editing instructions and reference images during autoregressive generation. To support these capabilities, EditWorld introduces Gated Causal Attention for temporally varying editing conditions and reference images, together with a Sparse Context mechanism that maintains a bounded historical context for long-horizon inference. We further adopt joint autoregressive and bidirectional training with annealed self-resampling, and construct a dedicated data synthesis and annotation pipeline that provides supervision for world editing. We also present WBench-Editing to systematically evaluate streaming world editing capabilities. EditWorld achieves the best overall performance on WBench-Editing with an overall score of 73.8 and an editing score of 80.0, substantially outperforming existing methods on editing-related metrics. https://github.com/leoisufa/EditWorld --- ### The Last Mile Is the File: OfficeEditBench for Preservation-Aware Office Editing Authors: Zhiwen Wu, Chengxu Wu Published: 2026-09-28 Categories: cs.SE, cs.CL arXiv: https://arxiv.org/abs/2609.34469v1 A small Office edit creates two obligations: propagate every required update and leave protected state untouched. Updating too little leaves dependencies inconsistent; updating too much changes content the user did not authorize. We introduce OfficeEditBench, a 170-task benchmark for change-scoped maintenance of spreadsheets, presentations, and documents. Task contracts specify required updates, protected state, native structures, and applicable interaction requirements. Across 510 archived task-system outcomes from WorkBuddy, Doubao, and Codex, we distinguish file delivery, target completion, and verifier-defined acceptance. Hard package-valid delivery ranges from 92% to 100%, yet no selected output satisfies the complete contract. Case analysis highlights why local correctness is insufficient: an updated value can lose its generating formula, a revised rule can fail to reach related conclusions, and a new deadline can omit a retained prerequisite. These mechanisms connect artifact-level checks to the continued maintainability of Office files. We analyze maintenance failures while distinguishing frozen automatic verdicts from human acceptability. OfficeEditBench provides a testbed for completing required changes while preserving the logic and scope of existing work. --- ### Alignment-Guided Flow Transformer for Efficient Vision-Language-Action Policy Learning Authors: Shengchao Hu, Peng Wang, Qiyang Zhou, Guodong Zheng, Yuqi Huang, Li Shen, Ya Zhang, Dacheng Tao Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34467v2 Recent advances in Vision-Language-Action (VLA) models point toward general-purpose robotic intelligence by unifying perception, instruction, and control. Despite impressive progress, existing VLA models often adapt poorly due to \emph{tri-modal misalignment} among vision, language, and action, which weakens action grounding and hurts generalization and fine-tuning efficiency. In this work, we present Alignment-Guided Flow Transformer (AGFT), a novel framework that explicitly enforces tri-modal alignment through a dedicated alignment loss, bridging the representational gap across modalities and enhancing task adaptation. While prior research has predominantly emphasized bi-modal vision--language alignment, we systematically formalize and study tri-modal alignment in VLA models, and provide both ablations and analysis to isolate its role in improving adaptation and robustness. To further accelerate deployment, we adopt a flow-matching objective, enabling substantially fewer inference steps than diffusion-based policies while maintaining accuracy. Theoretically, we establish a quantitative connection between the tri-modal alignment gap and the optimization tightness of flow matching; empirically, experiments on the extensive benchmark show that AGFT achieves superior success rates and lower inference latency compared to SOTA baselines, underscoring tri-modal alignment as a key ingredient for scaling robust VLA manipulation. --- ### PhysioTRACE: Provenance-Aware Stress Tests for Physiological Foundation Models Authors: Ayana Mussabayeva, Anuar Aimoldin, Olivier Oullier, Xue Liu, Kun Zhang Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34466v1 Physiological foundation models encode how a signal was recorded alongside the physiology it reflects. When recording conditions are associated with diagnosis, this acquisition provenance can become a shortcut, yet the usual evidence, shifted transfer and provenance decodability, does not show whether a predictor uses it. We introduce PhysioTRACE, a four-axis behavioral audit for frozen encoders that separates what a probe can decode from what a fixed task head relies on. Recover scores how decodable provenance is; Stress reverses only the provenance-target association on the same held-out records; Intervene removes a train-localized provenance component; and Verify certifies that removal only if it beats matched random projections within a declared utility margin. Each audit thus ends in one of three verdicts: no reliance, or reliance with the remedy certified or refused. Across EEG and ECG, five training objectives, and five frozen foundation models, the relation between Recover's calibrated score and out-of-distribution utility changes sign between datasets, so neither can stand in for a reliance test. On paired EEG views where the shortcut is known by construction, the audit detects it (the exposed head loses about 0.2 AUROC when the association is reversed, while a control head is unaffected) and certifies removal of a rank-two component that restores control-level behavior without measurable utility loss, for both encoder objectives tested. On real ECG device metadata it returns all three verdicts: it certifies a remedy that removes 91% of one model's excess vulnerability, finds no reliance where device and diagnosis are barely associated, and refuses the remedy for a second model whose localized direction also carries task signal. Robustness to how inputs were recorded therefore needs a behavioral test, and PhysioTRACE provides one that can pass, fail, or refuse a remedy. --- ### CoDeL: Co-Evolutionary Defense against Indirect Prompt Injection in LLM-based Agents Authors: Xiao Yang, Yangchen Ou, Yuhan Gao, Le Wang, Zonghao Ying, Aishan Liu Published: 2026-09-28 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.34463v1 Large language model (LLM)-based agents increasingly rely on external tools and content, exposing them to indirect prompt injection (IPI). This threat has motivated a wide range of defenses, among which training-based defenses are often regarded as most reliable. However, existing training-based defenses are typically optimized on a static distribution of explicit injections. They learn surface-form cues rather than the boundary between serving the user and obeying an injected objective, and therefore fail when malicious intent is folded into a plausible workflow and deferred for several turns. We present CoDeL, a defense that hardens agent against an attack distribution it reshapes as it trains. The defender is updated each round via LoRA-based GDPO under a decoupled reward over safety, task progress, and format compliance, so refusing injections and completing the user's task jointly define fitness. To keep supplying it with the failures worth learning from, a co-evolving prober searches over injection rounds, attack methods, and payloads for injections that still penetrate the current defender, guided jointly by attack success and attack latency so that it preferentially mines breaches the defender notices too late. Each defender update invalidates part of the attack population and forces the next round onto a new frontier, turning the defender's own failures into a moving curriculum. Extensive experiments on three IPI benchmarks, nine baselines, and two base models show that CoDeL reduces attack success rate (ASR) by 88.5% and outperforms other baselines largely (+38.0%). Codes are available. --- ### When Does Structured Knowledge Help Neural Theorem Proving? Authors: Sareh Nabi, Roland Vogl, Marzieh Nabi Published: 2026-09-28 Categories: cs.AI, cs.LG, cs.LO arXiv: https://arxiv.org/abs/2609.34460v1 Does structured mathematical knowledge help LLMs prove theorems in Lean 4? If so, for which models, and does the answer vary by problem? Formal libraries such as Mathlib encode 285,000+ verified theorems with syntactic dependencies, but the semantic layer mathematicians rely on for discovery (analogies, generalizations, cross-domain bridges) remains implicit. We introduce MathAgent, which builds this layer as a knowledge graph, MathKG, and uses it to augment LLM theorem provers. MathKG connects 364 Mathlib theorems and definitions by 9,434 typed semantic edges inferred via LLM-based relation extraction anchored to verified Mathlib declarations. We run a controlled ablation across four augmentation modes (no context, knowledge-graph context, Mathlib retrieval, both) and five models: Qwen3-8B/32B, their Lean-specialized derivatives Goedel-Prover-V2-8B/32B, and Claude Sonnet 4.6, on miniF2F, plus PutnamBench and MathOlympiadBench for Sonnet. Three findings emerge. (i) Specialization dominates augmentation: Lean fine-tuning adds 33-38 percentage points of solve rate in every mode, and a specialized 8B model beats a $4\times$ larger general one by 29-35 points, while no augmentation mode improves solve rate by more than 3 points. (ii) Augmentation is capability-conditioned: knowledge-graph context helps small models but hurts large ones, with the specialized model gaining more relative to its general base at every scale. (iii) Yet the augmentation modes solve different problems: an oracle selecting the best mode per problem solves 6% to 58% more than the unaugmented prover, a complementarity effect that strengthens on harder problems (32% more on PutnamBench). These results motivate adaptive strategies that select augmentation by model capability and problem. Code, data, and artifacts are available at https://github.com/sarehnabi/mathagent --- ### Escaping Local Views: Discovering Latent Concepts for Interpretable Multi-Agent Reinforcement Learning Authors: Yijie Sun, Sanquan Sun, Yanda Zhu, Yuanyang Zhu, Yaohua Hu, Chunlin Chen Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34459v1 Efficient cooperation is challenging due to the usual partial observability of each agent in multi-agent reinforcement learning. Recurrent networks encode local interaction histories, but their hidden representations provide limited insight into the information underlying individual decisions. To address these challenges, we propose a novel interpretable framework, called escaping local views (ELV), which introduces semantically structured latent concepts to render policy decisions transparent. Specifically, each agent extracts low-dimensional semantic concepts from its local observation and action-observation trajectory. These concepts are jointly encoded into a contextual latent variable via a variational autoencoder (VAE), which builds a bridge between local views and global semantics. To explicitly model the decision of each agent, we employ a dual-path attention mechanism in which one module estimates the salience of individual concepts relative to the global context, while the other captures higher-order cooperative patterns with pairwise concept interactions. Furthermore, we incorporate a concept prediction module that derives an intrinsic reward from next-concept prediction errors, which incentivizes agents to explore regions of semantic novelty. Experiments in multiple environments verify that ELV not only achieves competitive performance but also explicitly provides how agents reason about their decisions. --- ### Understanding Generalization Requires Universal Induction Authors: Aram Ebtekar, Marcus Hutter, Danica J. Sutherland Published: 2026-09-28 Categories: stat.ML, cs.AI, cs.IT, cs.LG, math.ST arXiv: https://arxiv.org/abs/2609.34458v1 Classical statistical theory is insufficient to explain the successes of general-purpose AI models, because it depends on handcrafted inductive biases that it cannot justify. No Free Lunch (NFL) theorems force any learner that beats chance on some environments to underperform on others. We might hope that past experience informs which environments to expect, but NFL applies equally to meta-learning. Thus, any method that makes meaningful predictions necessarily begins with an inductive bias external to the data. Choosing to bias toward short programs yields Solomonoff induction (SI), whose performance is competitive against all computable learners - albeit up to "constants" that become large when comparing against specialized methods that exploit background information. We therefore relativize SI to an information vantage point, biasing toward short programs with access to all preexisting information. This reframes the inductive bias: instead of seeking some absolute notion of simplicity, we favor accessibility with respect to our vantage point. An algorithm can only outpredict the relativized SI to the extent that its code contains additional information about the data, and no algorithm can generate such information. While SI is incomputable and hence not a practical algorithm, it provides a formal optimum for inference in the limit of infinite compute, and there is evidence to suggest that frontier AI systems roughly approximate it. Thus, the only known answer to meta-NFL is rooted in algorithmic information theory, which we should expect to play a fundamental role in explaining the generalization behavior of modern (and future) AI systems. --- ### ZonoGPT: Towards An Abstract Domain for Verifying Large GPT Models Authors: Hai Duong, Thanh Le, ThanhVu Nguyen Published: 2026-09-28 Categories: cs.LG, cs.SE arXiv: https://arxiv.org/abs/2609.34457v1 Transformer-based models are widely used for reasoning, coding, and multimodal agentic tasks. To provide formal assurance of desirable behaviors, such as robustness, safety, and fairness, neural network verification techniques prove required properties and provide auditable guarantees before deployment. However, prior work remains limited to small or restricted Transformers, and maintaining precision across deep models remains challenging. In this work, we introduce ZonoGPT, an abstract domain for verifying large transformers that maintains a space complexity independent of network depth. ZonoGPT uses a structured zonotope and a generator reduction mechanism to efficiently preserve correlations. To maintain precision, it introduces block-specific fused transformations for Attention and LayerNorm that retain feature relations, along with an affine transform for GELU that preserves generator relations. These mechanisms enable \tool{} to be the first approach to verify standard architectures, scaling to official HuggingFace models up to GPT-2 Medium (24 blocks, 300M+ parameters) and successfully verifying 1,339 instances across text and vision tasks. --- ### RGDT-Bench: Benchmarking LLM Reasoning for Rule-Governed Decisions and Their Justifications Authors: Jianpeng Zhao, Haihua Xu, Haoyang Zhang, Shuang Qian, Yixiang Tang, Xintao Wang, Kun Sun, Pei Wu, Shuhan Zhong, Pengyang Wang Published: 2026-09-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.34455v1 We study reasoning in Rule-Governed Decision Tasks (RGDTs), where models apply external rules to case facts and justify decisions, as required in policy, contract, and compliance settings. Beyond the deductive capability emphasized by standard mathematical and logical reasoning tasks, RGDTs require interpreting rules and their applicability, assessing conditions from evidence, combining judgments under rules and exceptions, and providing checkable justifications. These demands motivate a benchmark assessing both decisions and their stated grounds. We introduce RGDT-Bench, providing 202.1K condition-level supervision slots across four task tracks and eight supported task-probe combinations that vary access to supporting information. Label-blind extraction and deterministic checks produce labels for warrant completeness: source-referenced coverage and consistency of stated decision grounds. The benchmark attributes failures to four process layers: rule use, condition, evidence, and aggregation, and checks the final outcome. Among evaluable correct responses, warrant incompleteness averages 40.2% across six evaluated LLMs and supported task-probe combinations. Such warrant incompleteness poses potential safety risks and remains difficult to detect: the best of seventeen existing evaluators reaches only 57.69% (random: 50%) task-averaged area under the receiver operating characteristic curve (AUROC). To address this difficulty, we train a simple reward model with warrant supervision. It achieves 69.24% task-averaged AUROC among correct answers, exceeding the matched outcome-supervised baseline by 10.37 pp (percentage points) and the best existing evaluator by 11.55 pp. Beyond completeness assessment, the model outperforms both outcome-supervised baselines across nearly all response-selection comparisons, supporting RGDT-Bench's warrant supervision for RGDT reasoning. --- ### When Words Fall Short: Iterative Synergy Between Verbalized Reasoning and Hidden Features for LLM Confidence Estimation Authors: Yekun Xu, Ante Wang, Jingyi Ren, Xuanyi Chen, Weizhi Ma, Yang Liu Published: 2026-09-28 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.34454v1 Confidence estimation is crucial for developing trustworthy large language models (LLMs), with most methods following estimator-based or verbalization-based paradigms. While recent research increasingly focuses on improving verbalized self-reports of confidence, we challenge the prevailing view that this approach surpasses independent confidence estimators. Our empirical study shows that a dedicated confidence estimator can substantially outperform verbalized confidence, indicating that LLMs' internal representations contain richer confidence signals. Building on this finding, we propose Iterative Policy-Estimator Training (IPoET), a framework that synergizes the complementary strengths of verbalized reasoning traces and informative representations. IPoET alternates policy optimization with estimator updating, integrating estimator-derived confidence feedback into policy learning and refreshing the estimator on new policy rollouts. Experiments across diverse datasets and Qwen and Llama backbones demonstrate that, by iteratively exploiting richer hidden features and adapting to the evolving policy distribution, IPoET consistently outperforms both estimator- and verbalization-based baselines in-domain and achieves superior or comparable results across all out-of-domain metrics. For more details, refer to https://github.com/xyk829/ipoet. --- ### SPACE-LoRA: Allocating Activation-Subspace Protection for Continual Learning Authors: Seunghyun Yoo, Kiseok Kim, Hyeontae Joo, Junyeop Bang, Hwangnam Kim Published: 2026-09-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.34453v1 This study addresses the catastrophic forgetting problem that occurs when sequentially learning successive tasks using Low-Rank Adaptation (LoRA) from a lifelong learning perspective. While existing approaches have primarily constrained parameter updates or learning subspaces to reduce interference with past knowledge, they have not fully considered additive interference. This occurs when a newly added residual adapter on top of a fixed past model generates non-zero responses along input directions important for old tasks, thereby altering previous predictions. To this end, we propose Subspace Protection with Allocated Capacity for Efficient Continual Adaptation (SPACE-LoRA). SPACE-LoRA directly suppresses the responses of the new residual branch along input activation directions that are important for old tasks and adaptively determines the protection coverage for each module based on past-task sensitivity estimated via a common Fisher sensitivity-based coverage target. Under a fixed LoRA rank, this approach adaptively adjusts module-specific protection coverage while suppressing interference along input directions sensitive to old tasks. We assess the effectiveness of activation-subspace protection in mitigating catastrophic forgetting and examine the role of sensitivity-guided protection in continual learning across diverse tasks. Code is available at https://anonymous.4open.science/r/SPACE-LoRA-7864. --- ### Modeling Whole-Slide Images as Dynamic Tumor Microenvironment Fields Authors: Lei Wu, Jiashuai Liu, Di Zhang, Zhangpeng Gong, Yingkang Zhan, Yi Niu, Jiusong Ge, Chunze Yang, Kai Yi, Mireia Crispin-Ortuzar, Chen Li, Zeyu Gao Published: 2026-09-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.34451v1 Due to the gigapixel-scale nature of whole-slide images (WSIs), weakly supervised WSI analysis is commonly formulated as a multiple instance learning (MIL) problem, where patch-level features are aggregated into slide-level representations. However, diagnostic and prognostic evidence often arises from spatially coherent tumor microenvironment regions and their interactions, rather than isolated patches alone. Existing patch-level or static region-based methods usually overlook how tissue regions should be adaptively formed and subsequently evolved through microenvironment interactions across heterogeneous boundaries. In this paper, we propose Concept-Guided Tumor Microenvironment Evolution (TMEvolve), a reaction-diffusion-inspired framework that models WSIs as latent tumor microenvironment fields over discrete patch graphs. TMEvolve instantiates this view as a learnable graph-discretized evolution process over patch neighborhoods. It first forms adaptive soft tissue regions as coherent microenvironment units, then performs pseudo-time evolution through two complementary local dynamics: intra-region diffusion, which stabilizes latent states within coherent tissue compartments, and concept-guided boundary flux, which propagates visual feature signals and language-derived concept signals across heterogeneous region interfaces. The evolved microenvironment regions are finally aggregated for slide-level prediction. We evaluate TMEvolve on six datasets across three weakly supervised WSI tasks: survival prediction, gene expression prediction, and histological subtype classification. TMEvolve consistently improves over representative MIL methods, pathology foundation models, and concept-guided baselines. Ablation studies and visualizations further support the effectiveness and interpretability of TMEvolve, highlighting the value of dynamic region modeling and boundary interaction. --- ### CORTEX: Learning to Share and Specialize in Dense Language Models Authors: Chuiyang Meng, Ming Tang, Vincent W. S. Wong Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34449v1 Large language models are trained on heterogeneous data mixtures, where different knowledge domains require both shared knowledge and specialization. Existing modular approaches typically impose explicit components or discover modules through interpretability analysis after training. In this work, we propose CORTEX, a learning dynamics-inspired framework that learns internal modularization within dense language models. CORTEX partitions trainable matrices into parameter groups and learns module assignments from domain-conditioned gradient and cross-domain gradient similarity. We introduce the selective lesion score and module-domain mutual information to characterize the target-domain lesion effects and alignment, and analyze how module assignment affects the trade-off between assignment bias and update magnitude. Experiments with 160M, Qwen3-8B, and Qwen3-32B backbone models show that CORTEX achieves the highest synthetic-domain exact match and largest average perplexity reduction, while remaining competitive on real-domain evaluations and forming identifiable modules. --- ### Unbiased Top-$k$ Estimation for On-Policy Distillation Authors: Linjian Meng, Siyuan Gan, YuHan Li, Xiran Wang, Ziyang Ding, Ditang Gou, Yiming Wu, Zhen Zhao Published: 2026-09-28 Categories: cs.CL, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.34447v1 On-policy distillation (OPD) is becoming an important component of large language model (LLM) post-training for transferring the reasoning capability of a strong teacher LLM to a weaker student LLM. OPD trains the student by minimizing the reverse KL divergence between the teacher and the student via rollouts generated by the student's policy. However, estimating the gradient of the reverse KL divergence in OPD remains a challenge. Using only the sampled token from the student-generated rollout is computationally cheap but provides limited distributional supervision, which will degrade accuracy. In addition, using the full vocabulary provides complete distributional supervision but is computationally expensive. Therefore, recent works propose Top-$k$ OPD (TK-OPD) that use selected top-$k$ tokens, which provides richer distributional supervision than sampled-token estimation at substantially lower computational cost than full-vocabulary estimation. Unfortunately, using only the selected top-$k$ tokens induces bias, leading to accuracy degradation, as the probability mass outside the selected top-$k$ tokens is discarded. To address the bias of TK-OPD, we propose Tail-Corrected Top-$k$ On-Policy Distillation (TT-OPD). It preserves the advantages of TK-OPD, including rich distributional supervision and low computational cost, while providing an unbiased estimator of the gradient of the reverse KL divergence. The key insight of TT-OPD is to use not only the selected top-$k$ tokens, but also the sampled token from the student-generated rollout, thereby recovering the discarded probability mass in expectation, avoiding the bias. Experimental results demonstrate that TT-OPD significantly outperforms other tested OPD variants. --- ### Livin' on a Prior: Likelihood Score Approximation for Inverse Problems Authors: Rostislav Makarov, Tal Peer, Danilo de Oliveira, Timo Gerkmann Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34446v1 Generative models have found great success as data-driven methods of solving inverse problems. Two popular approaches work either by combining a pretrained generative prior with a known degradation model, or by training a conditional generative model directly from paired data. We target a setting that spans both regimes: unknown degradations can be learned from few paired examples, while known degradations can be learned from self-generated samples. We introduce Likelihood Score Approximation (LSA), a generative framework that keeps a pretrained unconditional model fixed and learns an observation-conditioned model that approximates the likelihood score from paired samples. Within a conditional stochastic-interpolant framework, LSA can be trained in either score or velocity coordinates, independently of the unconditional model's native parameterization, and supports both deterministic and stochastic sampling. We further show empirically that the prior model can be swapped post-training while keeping the same LSA model. Across speech and image inverse problems, LSA operates effectively even at roughly 0.01% of the full training dataset. On the ImageNet-256 benchmark it achieves competitive or better restoration quality than strong posterior-sampling baselines while requiring up to several orders of magnitude fewer network evaluations. --- ### Social Circuits behind Multi-agent Echo Chambers Authors: Chuiyang Meng, Wenlu Yu, Ming Tang, Cheng Li Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34444v1 Language-model agents exchange messages to combine evidence, but their communication can also create echo chambers that reinforce shared errors. However, overall task performance does not explain how a message changes the receiving agent's internal activations and affects its decision. In this work, we introduce Social Circuits, a framework for tracing message effects through receiver activations. We compare the receiver's answers before and after changing a message. Then, we restore selected activations recorded under the original message to determine how much of the message effect these activations reproduce. Based on Social Circuits, we propose Circuit-Guided Deliberation (CGD), which learns to select useful messages using receiver activation changes. We establish when activation replacement preserves receiver decisions and bound the gap between CGD's task performance and the best achievable through message selection. Experiments show that receiver activation changes explain the message effects and guide message selection that improves the task performance. Across three models and four datasets, CGD achieves the highest or joint-highest average accuracy in our main comparisons while generating fewer tokens than multi-agent baselines. --- ### Making LLMs Truly Forget: Deep Unlearning by Searching, Selecting, and Severing Knowledge Paths Authors: Jialu Wang, Peizhi Niu, Haoteng Yin, Hans Hao-Hsun Hsu, Pan Li, Rongzhe Wei Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34442v1 While an unlearned language model may no longer recall a fact directly, the fact often remains recoverable through multi-hop reasoning over related knowledge. Most existing unlearning techniques overlook this vulnerability, targeting facts in isolation while leaving their supporting knowledge intact. To achieve true forgetting, we propose a general deep unlearning framework compatible with existing unlearning algorithms. Our approach adaptively explores both explicit responses and latent internal representations to discover valid reasoning paths, compiles them into a confidence-aware supporting subgraph, and we apply a graph minimum cut to sever all recovery paths while preserving unrelated knowledge. To rigorously evaluate deep unlearning, we introduce a model-specific pipeline that extracts and completes knowledge graphs from raw text, filtering them by calibrated model confidence to reflect what the model genuinely retains. Comprehensive experiments demonstrate that selectively unlearning supporting knowledge yields substantially deeper forgetting than superficial methods while preserving model utility, highlighting that genuine unlearning requires breaking the relational structures that enable factual reconstruction. --- ### When the Score Becomes the Target: Rethinking Metric Validity in Autonomous Driving Authors: Morui Zhu, Deyuan Qu, Qi Chen, Kentaro Oguchi, Qing Yang Published: 2026-09-28 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.34440v1 Driving benchmark scores are increasingly used not only for evaluation but also as optimization targets. This raises a fundamental question: do score gains remain reliable evidence of driving improvement once the score itself is optimized? We address this question by examining how the scoring process responds to changes in driving behavior and whether the resulting gains persist under repeated execution and replanning. We decompose the process into execution, measurement, subscore mapping, and aggregation. Controlled interventions reveal substantial behavioral changes that receive little score response because distinctions are omitted, thresholded, or attenuated between requested and executed motion. Closed-loop comparisons further show that optimization gains can reverse when the execution interface changes, demonstrating their dependence on how requests are executed and returned as feedback. Together, these findings connect the behavioral distinctions preserved by a metric to the conditions under which its gains transfer. Metric validity under optimization therefore requires examining both what the scoring process measures and how the optimized behavior is executed. --- ### Clinical Trajectory Alignment for Medical Vision-Language Pre-training Authors: Huimin Yan, Xian Yang, Zhi Wang, Liang Bai Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34439v1 Medical vision-language pre-training largely follows a visit-level image-report matching paradigm, aligning paired images and reports at individual visits. While effective for static cross-modal correspondence, this paradigm provides limited supervision for longitudinal clinical change, such as whether abnormalities improve, remain stable, or worsen over time. Learning such change is challenging because temporal semantics are implicit in free-text reports, and different abnormalities within the same patient may evolve asynchronously or even in opposite directions. We propose MedCTA, which reframes medical vision-language pre-training from visit-level cross-modal matching to learning clinical change. Rather than compressing a patient history into a single temporal representation, MedCTA models clinical change at two complementary scopes. At the abnormality scope, clinically grounded queries construct abnormality-conditioned visual and textual trajectories to capture heterogeneous abnormality evolution. At the patient-course scope, global image and report sequences are modeled to capture overall clinical progression beyond any individual abnormality. Structured trend supervision is extracted from longitudinal reports by an offline LLM parser, removing the need for manual temporal annotations. Combined with static image-report alignment, MedCTA learns representations that preserve visit-level cross-modal correspondence while encoding longitudinal change semantics. Experiments on temporal image classification, image-text retrieval, and zero-shot classification show consistent gains over strong medical vision-language baselines. --- ### Remember by Asking: Retrieval-Induced Memory Evolution for LLM Agents Authors: Wanqi Zhou, Jiawei Lu, Yang Wang, Zhaolong Xing, Zhen Chen, Ai Han, Haoyue Shi Published: 2026-09-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.34438v1 Long-term memory is essential for language agents to maintain coherent and effective behavior over extended, multi-session interactions. Existing memory systems mainly use retrieval at read time, while write-time memory formation still relies on direct extraction or compression. However, when future information needs are unknown, compressing an entire interaction in one pass can overlook locally important details that may matter later. To this end, we introduce RIME, a retrieval-induced memory framework that shifts memory construction from monolithic compression toward evidence-centered integration. RIME uses generic self-questions to retrieve focused dialogue evidence and grounds memory formation in both the retrieved evidence and relevant historical memories, which are jointly reconciled into an evolving memory bank with temporal and provenance information. At inference time, compressed memory serves as the primary rather than the sole source of evidence: when it cannot support an answer, RIME retrieves relevant source dialogue together with its local context to recover information omitted during memory formation, without resorting to full-history processing. Extensive experiments on LoCoMo with Qwen3-235B-A22B and GPT-5.6 Sol show that RIME consistently achieves the best performance across all three quality metrics among the compared methods, while requiring substantially fewer query-time LLM tokens. --- ### Admissible Diffusion for Multimodal Interventional Trajectories Authors: Xing Han, Shravan Chaudhari, Jiarui Shao, Paul Pu Liang, Suchi Saria Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34433v1 Generating a plausible clinical trajectory does not establish what would happen under a different treatment. We present ADMIT, a framework combining irregular multimodal representations, treatment-conditioned latent diffusion and explicit constraints on generated states or actions. We formulate its interventional target through sequential g-computation and distinguish causal assumptions from constraint satisfaction. Its admissibility mechanism translates physiological prior knowledge into explicit constraints on generated states and proposed actions. Treatment-exposure dynamics condition latent transitions, while state projection or action gating applies the constraints during rollout so that they influence subsequent trajectory generation. In our preliminary experiments, multimodal inputs improved supervised hidden-state recovery and reduced treatment-contrast error. In a simulated dosing-schedule experiment with leak-free history encoding, ADMIT predicted most of the tumor-volume change caused by redistributing a fixed total dose. An exposure input improved these predictions around a temporary dose reduction whether or not the assumed clearance rate was correct, but reduced the predicted size of a dose effect, and a deterministic recurrent baseline matched ADMIT's average predictions. Exposure projection reduced constraint violations, although enforcement remained incomplete. Semi-synthetic experiments using eICU context illustrated treatment-response generation under fixed and adaptive policies. Observational examples further characterize model treatment sensitivity. ADMIT provides a framework for testing whether complementary observations and physiological restrictions improve intervention trajectories, with representation recovery, effect accuracy and rule enforcement assessed separately. --- ### Harmonizing Spectral Evolution in Conditional Flow Matching for TTS Authors: Isha Pandey Varad Deshpande Abhijat Bharadwaj Ganesh Ramakrishnan Published: 2026-09-28 Categories: cs.LG, cs.SD arXiv: https://arxiv.org/abs/2609.34431v1 Conditional Flow Matching (CFM) models for text-to-speech (TTS) suffer from incoherent frequency evolution during inference. While similar spectral imbalances are addressed in diffusion models for other domains, those generic solutions fail to generalize to the inherently uncoordinated acoustic dynamics of CFM. We demonstrate that this issue can be effectively mitigated by introducing a novel training-free frequency-selective boosting strategy. Using the Discrete Wavelet Transform (DWT), our method dynamically modulates mel-spectrogram sub-bands during ODE integration, synchronizing spectral development by penalizing aggressive low-frequency growth and boosting lagging high-frequency details. Validated across diverse architectures (Matcha-TTS, F5-TTS, IndicF5), our approach reduces the required Number of Function Evaluations (NFE) from 32 to 26 and improves Frechet Audio Distance (FAD) by up to 61%, all without compromising mean opinion scores, speaker similarity, and speech intelligibility. --- ### HUMAN-TCI: Hierarchical Multi-Stream Motion-Aware Network with Torso-Centered Interaction for Text-to-Motion Retrieval Authors: Muhammad Islam, Euijoon Ahn, Usman Naseem, Tao Huang Published: 2026-09-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.34430v1 Accurate retrieval of human motions is a crucial first step in text-guided human motion modeling and synthesis, as it selects semantically relevant sequences from large datasets and provides grounded references for downstream tasks. Retrieving motions from natural language descriptions remains challenging because sentences can describe multiple actions, overlapping movements, and intricate dependencies between body parts. Existing methods often focus on simple, single-action descriptions and typically process body parts independently or by merely concatenating features, without explicitly modeling how torso movements influence other parts. In addition, their processing pipelines often rely on computationally heavy models, introducing considerable overhead, particularly when modeling longer or more complex motion sequences. This limits learning discriminative motion-pattern representations, reducing retrieval accuracy, interpretability, and efficiency in practical applications. To address these limitations, we propose HUMAN-TCI, a Hierarchical Multi-Stream Motion-Aware Network for text-guided human motion retrieval. HUMAN-TCI employs a three-stream architecture that separately models upper-body, lower-body, and torso motions while explicitly capturing their interactions, allowing torso-related movements to influence the positioning and dynamics of other body parts. By incorporating tailored torso attention, our model effectively recognizes complex human motion patterns, captures fine-grained motion relationships and handles complex multi-action descriptions. Our framework supports retrieval for both simple, single-action sentences and long, compositional descriptions containing sequential or overlapping actions without relying on complex models. --- ### AgentHop: A Diagnostic Benchmark for Agentic Multi-Hop Scientific Question Answering Authors: Chanhee Park, Jeongho Yoon, Sungbin Han, Hyeonseok Moon, Heuiseok Lim Published: 2026-09-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.34428v1 Agentic tasks require a large language model to interact with the world, navigating information and gathering evidence across multiple steps with restricted resources. Due to this complexity, agentic task failures arise from various sources, and pinpointing these failure causes is essential to diagnose and improve agentic systems. Existing benchmarks, however, tend to focus on a single leaderboard score, leaving the underlying failure modes opaque. To fill this gap, we introduce AgentHop, a diagnostic benchmark of 1,011 multiple-choice questions paired with a controlled seven-tool sandbox under fixed token, turn, and tool-call constraints. AgentHop reveals model vulnerabilities by dissecting a single accuracy score along four axes of agent operation: retrieval, synthesis, tool-call, and resource management. Across 19 models, we find that behavior clusters by model family, with tool-call signatures revealing distinct family fingerprints: GPT models commit early, Anthropic and GLM checkpoints verify before committing, DeepSeek and Kimi over-search, and Gemini-3 Pro stays balanced. Decomposed axes further expose within-family structure: Claude Opus 4.6 and Sonnet 4.6 land within one accuracy point yet diverge on retrieval-versus-synthesis emphasis, with Opus retrieving more and Sonnet synthesizing better. We release the full benchmark set and the harness to support diagnostic agent benchmarking. --- ### LLMs as Adaptive Meta-Solvers: Strategy-Diverse RL for Industrial-Scale Optimization Authors: Shihao Zhang, Weiting Liu, Siyu Shao, Yitian Chen, Jianfeng Feng, Dongdong Ge, Yinyu Ye Published: 2026-09-28 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.34427v1 Scaling LLM-based optimization from textbook-scale instances to real-world, industrial tasks remains a critical open challenge. Existing approaches are predominantly evaluated on small, self-contained textual problems and often commit to a solver-integrated paradigm, limiting their ability to handle the scale and structural diversity of practical optimization workloads. In this work, we propose a practical framework for training open-source LLMs to tackle real-world, industrial-scale optimization. We first show empirically that solver-integrated reasoning, exact combinatorial algorithm, and heuristic search exhibit complementary strengths across different problem structures and scales. Motivated by this, we introduce Strategy-Diverse Reinforcement Learning (SDRL), which trains LLMs as adaptive optimization meta-solvers. SDRL leverages this complementarity through a correctness-gated hierarchical diversity reward that promotes robust exploration across varying strategies and within each strategy, effectively preventing premature strategy collapse. We further introduce a mixed-format training scheme that jointly supports both self-contained textual problems and file-grounded instances. Across comprehensive evaluations, our framework outperforms existing fine-tuned methods and frontier models including DeepSeek-V4-Pro and GPT-5.5, both on average across benchmarks and on industrial-scale optimization tasks. --- ### Q-learning Penalized Transformer for Safe Offline Reinforcement Learning Authors: Shengchao Hu, Peng Wang, Jifeng Hu, Qiyang Zhou, Anning Hu, Li Shen, Ya Zhang, Dacheng Tao Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34426v2 This paper addresses the problem of safe offline reinforcement learning, which involves training a policy to satisfy safety constraints using an offline dataset. This problem is inherently challenging as it requires balancing three highly interconnected and competing objectives: satisfying safety constraints, maximizing rewards, and adhering to the behavior regularization imposed by the offline dataset. To tackle this trilogy challenge, we propose Q-learning Penalized Transformer policy (QPT), a \emph{training--inference consistent} framework that bridges conditional sequence modeling with constraint-aware value estimation. QPT trains a Transformer policy that generates actions conditioned on trajectory context and target return/cost, retaining strong behavior regularization. To inject explicit safety semantics during learning, we augment sequence-model training with a Q-shaped penalty using learned reward and cost Q-functions to favor high return under low constraint violation. At inference, the same Q-functions enforce the cost threshold and choose the highest-reward feasible action, closing the loop between training and deployment. We provide a principled analysis under stylized near-deterministic CMDPs, characterizing how Q-penalized conditional generation improve safety and performance. Empirically, QPT consistently outperforms strong safe offline RL baselines across 38 tasks on the DSRL benchmark, and exhibits robust zero-shot adaptation to different constraint thresholds. --- ### Zero-Shot Cue-Grounded Topic Segmentation of Spoken Documents Authors: Suhwan Choi, Myeongho Jeon, Myungjoo Kang Published: 2026-09-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.34425v1 Topic segmentation structures spoken documents into coherent sections, facilitating navigation and downstream understanding. The appropriate granularity can vary substantially, ranging from broad thematic shifts to fine-grained subtopics. Existing LLM-based segmenters, however, often struggle to adapt to this variation, causing them to either merge distinct subtopics or over-segment coherent themes. To address this, we introduce Cue-Grounded Segmentation (CGS), a training-free framework that operates without any task-specific supervision. CGS first identifies phrases that explicitly signal the start of a new topic and uses their sentence positions as segment boundaries. When such cues are insufficient, it falls back to semantic segmentation, guided by the document structure inferred during cue extraction. Across six benchmarks and six LLM backbones, CGS consistently outperforms existing baselines, remains robust to noisy ASR transcripts, and achieves these gains with low API cost on proprietary models. --- ### On the Relation Between Interval Regret and Dynamic Regret Authors: Yi-Han Wang, Peng Zhao, Zhi-Hua Zhou Published: 2026-09-28 Categories: cs.LG, math.OC, stat.ML arXiv: https://arxiv.org/abs/2609.34423v1 Non-stationary online learning has attracted much attention in recent years, as static regret is insufficient to guide algorithm design in changing environments. To address this limitation, interval regret and dynamic regret have been introduced as two representative performance metrics that strengthen static regret in complementary directions. Interval regret requires an online algorithm to achieve competitive static regret over every local time interval, whereas dynamic regret evaluates performance against an arbitrary sequence of time-varying comparators. Despite their importance, the relation between these metrics has long remained unclear. Prior work has often regarded interval regret as the stronger notion, based on the intuition that local guarantees should naturally induce global guarantees. Consequently, it is widely conjectured that an algorithm with optimal interval regret should automatically attain optimal dynamic regret. In this paper, we first establish a negative result that refutes this intuition of a metric-level implication. Specifically, for both convex and curved functions (including exp-concave and strongly convex functions), we show that there exist instances in which an algorithm with optimal interval regret nevertheless fails to achieve optimal dynamic regret. We then show how to leverage local adaptivity to obtain optimal dynamic regret. In particular, optimal dynamic regret can be attained by invoking an interval regret minimization process over an enlarged Euclidean ball containing the original convex feasible domain and using a suitable domain-converted surrogate loss. This reduction applies to both convex and curved functions. As a byproduct, we obtain the first proper and efficient algorithm with optimal dynamic regret for exp-concave functions, improving prior results while significantly simplifying the analysis. --- ### Coding Agent Memory Post-training: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via Reinforcement Learning Authors: Lirui Luo, Kelong Mao, Heming Xia, Rongqing Li, Xinwei Yang, Luyu Chen, Kieran Wong, Yudong Guo, Xinrui Wang, Jiayin Zhu, Simiu Gu, Sulong Xu et al. Published: 2026-09-28 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.34422v1 Language-model agents increasingly tackle long-horizon tasks whose interaction histories exceed the model's active context. Recent work has begun to use reinforcement learning to make memory control part of the policy, often relying on predefined memory tools within domain-specific training environments of relatively short horizons. This setup ties learned memory behavior to environment-specific interfaces that lie outside the base model's pre-training and must be learned from scratch, so even after post-training, agents struggle to use memory in long-horizon tasks. To address these limitations, we introduce Coding Agent Memory Gym (CAMG), a suite of long-horizon agentic-RL environments spanning Shop, Coding, DeepResearch, and AutoResearch. Alongside each environment's native task interface, CAMG provides executable shell access and an episode-persistent workspace, enabling agents to create, revise, search, and reuse files as memory throughout an episode. We also introduce CAMG-RL, which trains a single policy jointly across all four environments with fully asynchronous PPO, learning this file-based memory behavior directly from downstream task reward, and we train CAMG-RL-4B and CAMG-RL-9B from Qwen3.5 models of matching size. On SWE-bench Verified and MLE-bench Lite, CAMG-RL-4B and CAMG-RL-9B are competitive with Qwen3.5-35B-A3B and Qwen3.5-122B-A10B, respectively. --- ### MegaGraph: Towards Efficient Training of Large-Scale Graph Transformers with Automated Hybrid Parallelism Authors: Tong Qiao, Ao Zhou, Yingjie Qi, Chunming Hu, Jianlei Yang Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34420v1 Graph Transformers (GTs) offer superior representation capabilities by overcoming the depth limitations and over-smoothing issues of traditional Graph Neural Networks (GNNs). However, scaling GTs to large graphs poses critical bottlenecks. Specifically, the attention score matrix and its associated topology-aware bias matrix jointly incur significant per-layer memory overhead, and heavy graph embedding layers result in severe workload imbalances. These characteristics are unique to GT training and are not addressed by parallelism techniques designed for either conventional GNNs or Transformers, making a dedicated solution necessary. This paper introduces MegaGraph, the first automated hybrid parallelism framework designed for efficient GT training. MegaGraph designs three specialized strategies, namely graph-aware context parallelism, heterogeneous pipeline parallelism, and hybrid data parallelism, to support efficient training on large-scale graphs. However, coordinating these three parallelism strategies yields an exponentially large configuration space. To address this complexity, an automatic search engine leverages precise cost models via a Profile - Model - Search workflow to identify the optimal parallelism configuration. Evaluations demonstrate that MegaGraph enables training on large-scale graphs where state-of-the-art baselines fail due to out-of-memory (OOM) errors. The framework reduces per-device peak memory by up to 77.8\% and achieves up to 4.51$\times$ training speedup while maintaining model accuracy. --- ### Beyond End-to-End Black Box Mapping: An Intentional Agent Framework for Cognitive-driven Facial Reaction Generation Authors: Hanzhong Zhang, Jindong Wang, Siyang Song Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34419v1 Automatic human-like facial reaction generation (FRG) is essential for building intelligent systems that can engage in human-computer interaction (HCI). While diverse and context-appropriate facial reactions can reflect latent appraisal and affective processes in human interaction, most existing FRG methods rely on end-to-end architectures that directly map speaker behaviours to listener expressions without an explicit intermediate internal state. We reformulate FRG as generation mediated by a structured internal-state process and propose the \textbf{Intentional Agent}, which shifts FRG from direct stimulus-response mapping to stimulus-grounded generation through explicit intermediate states. To represent temporal internal-state evolution, we propose an internal dynamics model that integrates emotional drives with an iterative Inner Thought Flow (ITF) within a structured intermediate state used for subsequent generation. This state can continue to update during conversational silences. Furthermore, to bridge abstract internal states with physiological actions, we formulate FRG as a downstream affective mapping from this latent thought flow to facial expressions. Experiments on the REACT 2025 dataset show an FRDist of 72.39 and an FRDiv of 0.5057; perceptual plausibility is evaluated separately through blinded human ratings. A blinded human evaluation of 96 reactions found no significant difference in mean score between Full and ground truth ($5.527$ vs.\ $5.195$, $p_{\mathrm{Holm}}=.076$), while Full significantly outperformed Event-Triggered and Heuristic-Only (both $p_{\mathrm{Holm}}<.001$). The Reaction Quality Scorer (RQS) correlated strongly with human judgements (Pearson $r=.855$; Spearman $ρ=.821$, both $p<.05$), supporting its use as an automatic metric. These results underscore the immense potential of endogenous dynamics in building highly autonomous, human-like agents. --- ### OSPD: On-Policy Self-Distillation for Persona-Consistent Dialogue Authors: Rui Xu, Yikai Zhang, Aili Chen, Zicheng Zhao, Xu Yinghui, Libo Wu Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34418v1 Maintaining persona consistency across multi-turn dialogues remains a core challenge for role-playing language models. Off-policy distillation from external teachers incurs distribution mismatch that compounds across dialogue turns, while reinforcement learning struggles with reward ambiguity inherent in subjective persona fidelity. We propose OSPD, an on-policy self-distillation framework where the same model serves as both teacher and student under asymmetric information: the teacher receives a complete character profile while the student sees only a brief summary, and the student generates trajectories from its own policy. We find that teacher confidence in role-playing dialogue exhibits a bimodal structure---sharply peaked at character-critical tokens yet diffuse at generic utterances---and introduce role-aware divergence switching to match this structure. A progressive trait masking curriculum further forces staged internalization of character knowledge along semantic dimensions. Experiments on CharacterBench, CharacterEval, and SocialBench show that OSPD substantially improves persona consistency over supervised fine-tuning and multi-turn RL baselines, without requiring any external teacher or reward model. --- ### PROACT-Agent: Progressive Runtime Oversight and Active Circuit-breaking for Real-Time Safety Authors: Ding Jia, Wei Liu, Xianglong Du, Yingjie Li, Yingqing Yang, Huili Yu, Zhangsong Zhan, Chu Zhou Published: 2026-09-28 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.34415v1 The transition from Large Language Models (LLMs) to agents shifts safety stakes from toxic text to irreversible environmental harm. While current defenses remain largely retrospective, proactive runtime intervention is bottlenecked by the lack of large-scale, causally-consistent data. We propose PROACT-Agent, a framework for synthesizing high-fidelity trajectories to enable real-time guardrails. We identify a critical "safety drift" in prior benchmarks, where lenient annotation paradigms fail to enforce temporal consistency. PROACT-Agent addresses this through: (1) Progressive Trajectory Unrolling to reveal risks hidden in long-context interactions; (2) Reasoning-Augmented Causal Rectification to enforce monotonic causal consistency; and (3) Culturally-Aware Data Localization for cross-border robustness. We introduce PROACT-Bench, a bilingual safety benchmark with 155,780 states labeled through multi-model adjudication. Evaluating updated context before the next LLM inference, the trained guard achieves 91.46% unsafe-class F1 and 90.63% exact-boundary detection under complete source holdout. In AgentDojo, it reduces non-DoS targeted attack success from 20.82% to 0.40%. --- ### Mathematics for and by human cognition: A resource-rational search for bottlenecks in problem-solving Authors: Sneha Aenugu Published: 2026-09-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.34410v1 Human cognitive constraints are generally viewed as limiting factors in problem-solving. We argue that these constraints can instead play a critical role in driving advances in mathematics and beyond. We propose a theory of mathematical abstraction as a resource-rational search for bottlenecks in problem-solving. Bottlenecks arising from cognitive constraints create pressure to restructure existing knowledge, potentially giving rise to novel formalisms with applications beyond the problems that originally motivated them. Drawing on episodes from the history of mathematics, we illustrate how such bottlenecks can drive the development of novel abstractions and examine how cognitive constraints and affective responses shape this process. Finally, we discuss the implications of this account for machine mathematical discovery and argue that incorporating human-like constraints may facilitate the discovery of useful mathematical abstractions. ---