# Research Papers — 2026-08 · bundle page 201 of 240 > Full-text markdown of papers #10001–10050 of 11995 published in 2026-08, newest first. AI-readable mirrors of the full 11995-paper 2026-08 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-08/page/201 [← Previous bundle](https://aipod.dev/papers/archive/2026-08/page/200.md) [Next bundle →](https://aipod.dev/papers/archive/2026-08/page/202.md) ### A GitOps-Driven Annotation Catalog for Fully Automatic Railway Operations Authors: Martin Köppel, Tobias Cronauer, Zekiye Ilknur-Öz, Sebastian Dubiel, Patrick Naumann, Philipp Neumaier Published: 2026-08-05 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2608.04724v2 Automatic train operation (ATO) at grade of automation 3 and above (GoA3-GoA4) requires robust AI-based perception systems capable of reliably detecting obstacles and railway-specific objects under real-world conditions. The effectiveness of these modern artificial intelligence approaches depends heavily on large-scale, high-quality, and highly dynamic annotated datasets. However, managing metadata, maintaining provenance, and tracking the iterative evolution of these annotations impose significant infrastructural and regulatory requirements. Existing monolithic data catalogs often suffer from massive operational overhead, poor integration into developer workflows, and severe documentation drift. This paper introduces an innovative, lightweight GitOps-based architecture for metadata management. By leveraging Data-as-Code principles, Continuous Integration/Continuous Deployment (CI/CD) pipelines, and Static Site Generation (SSG), the proposed approach establishes a seamless, developer-centric workflow. This ensures an traceability, enforces strict regulatory compliance, and automatically generates a highly performant dataset overview. --- ### A GitOps-Driven Annotation Catalog for Fully Automatic Railway Operations Authors: Martin Köppel, Tobias Cronauer, Zekiye Ilknur-Öz, Sebastian Dubiel, Patrick Naumann, Philipp Neumaier Published: 2026-08-05 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2608.04724v1 Automatic train operation (ATO) at grade of automation 3 and above (GoA3-GoA4) requires robust AI-based perception systems capable of reliably detecting obstacles and railway-specific objects under real-world conditions. The effectiveness of these modern artificial intelligence approaches depends heavily on large-scale, high-quality, and highly dynamic annotated datasets. However, managing metadata, maintaining provenance, and tracking the iterative evolution of these annotations impose significant infrastructural and regulatory requirements. Existing monolithic data catalogs often suffer from massive operational overhead, poor integration into developer workflows, and severe documentation drift. This paper introduces an innovative, lightweight GitOps-based architecture for metadata management. By leveraging Data-as-Code principles, Continuous Integration/Continuous Deployment (CI/CD) pipelines, and Static Site Generation (SSG), the proposed approach establishes a seamless, developer-centric workflow. This ensures an traceability, enforces strict regulatory compliance, and automatically generates a highly performant dataset overview. --- ### Multi-View Face and Gesture Animation with Dynamic Gaussians Authors: Alireza Javanmardi, Vippin Kumar Jeetmal, Christen Millerdurai, Alain Pagani, Didier Stricker Published: 2026-08-05 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2608.04722v1 Creating photorealistic 3D human avatars with realistic upper-body motion remains challenging. Existing approaches either focus on the head and overlook hand gestures, or reconstruct the full body but fail to preserve fine-grained facial fidelity and hand pose accuracy. As a result, current methods struggle to capture the subtle dynamics of facial expressions and hand gestures that are crucial for natural human communication. While methods based on full-body parametric models enable avatar reconstruction from monocular or multi-view inputs, they often lack accurate facial animation and detailed hand articulation. To address these limitations, we propose MVFGA, a novel multi-view-consistent pipeline for generating realistic upper-body avatars. Our approach models the face and hands separately and fuses them with a parametric upper-body mesh model, enabling the capture of fine-grained facial expressions and hand poses for accurate upper-body avatar reconstruction. We then splat 3D Gaussians onto the obtained mesh, enabling high-quality rendering of dynamic avatars from novel viewpoints. Furthermore, we introduce MVFGA-MoCap, a multi-view upper-body motion capture dataset featuring controlled facial expression sequences, diverse hand gestures, and free-form communication. Experiments show that MVFGA generates visually realistic avatars with high-fidelity facial expressions and hand motions, outperforming baselines for upper-body avatar animation. Project page: https://dfki-av.github.io/MVFGA/ --- ### YOLOv14: Adaptive Real-Time Object Detection for Diverse Imaging Conditions Authors: Jian Lu, Jinling Jia, Jone Yawl, Chenbin Zhang Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04720v4 Real-time object detectors achieve remarkable accuracy under controlled conditions, yet degrade sharply on non-ideal inputs-fisheye distortion, game-rendered content, aerial views, and 360°panoramas. We present YOLOv14, a unified adaptive detection framework that addresses these variations through four complementary mechanisms, formalized under a novel Adaptive Routing and Modulation (ARM) paradigm. Unlike conventional unsupervised domain adaptation, our approach employs Target-Prior Guided Source-Domain Augmentation(TP-SDA), using only 50 unlabeled target images offline to estimate style statistics, while adversarial alignment serves as a lightweight regularizer rather than the primary adaptation driver. Together, these components enable YOLOv14 to achieve 49.1 mAP on COCO val2017 at 2.91 ms (T4 GPU), with substantial gains of +4.1 (fisheye), +6.6 (panorama), +6.4 (drone), and +26.1 (gamestylized) mAP over YOLOv12s. Crucially, we validate generalization on real-world game screenshots (GTA-V, Unity), achieving +14.2 mAP, confirming practical transferability beyond synthetic benchmarks. Code and models are released at https://github.com/zhangcbb/yolov14. --- ### YOLOv14:Unified Cross-Domain Real-Time Object Detectionwith Adaptive Multi-View Representation Authors: Jinling Jia, Jian Lu, Jone Yawl, Chenbin Zhang Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04720v1 Real-time object detectors achieve remarkable accuracy under controlled conditions, yet degrade sharply on non-ideal inputs: fisheye distortion, game-rendered characters, aerial viewpoints, and 360° panoramas. We present YOLOv14, aunified detection framework addressing these challenges through four synergisticinnovations. (1) Deformable Area-Attention (D-AAttn) replaces rigid attentiongrids with learned 2D deformation fields, enabling adaptive sampling under geometric distortion. (2) Game2Real Domain Adaptation aligns rendered-game and photographic feature distributions via Adaptive Instance Normalization (AdaIN)and adversarial domain confusion, allowing game characters are detected as realhumans. (3) Multi-View Conditioning injects learned viewpoint embeddings intothe backbone with a cross-view contrastive loss that pulls same-class features fromdifferent perspectives closer. (4) An Adaptive Augmentation Policy automaticallyclassifies each input' scene type and routes to optimal augmentations, while a DynamicScaleRouter learns per-input feature pyramid weights. Together, YOLOv14achieves 49.1 mAP on COCO val2017 at 2.91 ms (T4 GPU), and delivers substantial gains on fisheye (+4.1 mAP), panorama (+6.6 mAP), drone (+6.4 mAP), andgame-character (+26.1 mAP) benchmarks --- ### Diagnosing Tool-Selection Reasoning in LLM Agents with Canary Tools Authors: Atul Anand, Sourav Chattaraj Published: 2026-08-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.04719v1 Agent evaluations tell us that a model picked the wrong tool, but rarely why. We introduce canary tools: diagnostic probe tools planted in an agent's Model Context Protocol (MCP) tool set, each engineered to probe one specific tool-selection weakness. A six-type taxonomy (semantic decoys, parameter traps, capability mirages, prerequisite blindness, temporal decoys, and granularity traps) turns a single "wrong tool" outcome into a multi-dimensional profile of how a model reasons about tools. We evaluate eight models -- six hosted and two 8B open-weight -- spanning three capability tiers, on 120 tasks across three canary-density conditions and three seeds (8,640 runs), plus a 2,880-run subtlety ablation. Task success is graded by a provider-independent judge, corroborated by a second independent judge (Cohen's kappa = 0.75). We report three findings. First, susceptibility drops sharply as models get more capable: the per-task canary susceptibility rate (CSR) ranges about 36x across models, lowest for Claude Opus 4.8 and highest for Llama 3.1 8B. Second, capability tier alone does not predict safety: the most susceptible hosted model is mid-tier, and within a provider the cheaper model can be the safer one. Third, the taxonomy is capability-stratified: capability mirages most reliably trap frontier models, while the other types are largely inert on strong models but fire on small open models, so they discriminate by capability rather than being weak. Softening each canary's give-away phrase leaves frontier CSR essentially unchanged, evidence that the probes measure reasoning, not phrase-spotting. Susceptibility also predicts task failure (Spearman rho = -0.34), while the most robust models are not significantly degraded by canary pressure. We release the framework, canary schemas, tasks, and logs. --- ### What We Observe as LLM Behavior Can Be a Side-effect of Inference Backend Authors: Shahed Masoudian, Passant Shafaei, Monorama Swain, Markus Schedl Published: 2026-08-05 Categories: cs.SE, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2608.04714v1 Benchmark scores are reported as properties of a model, yet the inference framework used to produce them, such as HuggingFace, vLLM, or Ollama, are considered non-influential and their names and versions are almost never disclosed. In this work we investigate how much this choice can influence the model output. In a fully-crossed study (three instruction-tuned models x five inference frameworks x six benchmarks x four generation modes) we investigate how different tools (wrappers/backend) influence benchmark scores and how their score changes is influenced by generation hyper-parameters. We find backend to be a non-negligible factor where even under greedy, sampling-noise-free decoding, changing the backend can significantly alter models performance and this effect is structural and strongly model-dependent. Decomposing the variance according to generation mode reveal that considerable portion of the variability (roughly 39\%) a practitioner sees out-of-the-box can stem from the backend, while the remaining stems from sampling noise and each framework's default generation parameters, both of which are avoidable by disclosing and matching the generation configuration. These divergences are more pronounced on factual than on social-bias benchmarks. Overall, benchmark numbers are not backend-agnostic therefore, we recommend disclosing the backend, its version, and the full generation configuration, also using deterministic decoding for cross-backend comparison. --- ### A 6G Integrated Sensing and Communication Framework for Railway Intrusion Detection and Collision Prediction Authors: Ajeet Kumar Yadav, Sankaran Balasubramaniam, Aritra Chatterjee, Vinod Aduru, Yogesh Simmhan, Pandarasamy Arjunan Published: 2026-08-05 Categories: cs.LG, cs.AI, eess.SP arXiv: https://arxiv.org/abs/2608.04710v1 Integrated Sensing and Communication (ISAC) combines sensing and communication to efficiently utilize wireless resources and is emerging as a key paradigm for next-generation wireless networks. By leveraging the wide bandwidth, high frequencies, and massive antenna arrays of 5G-Advanced and 6G systems, ISAC enables physical-layer sensing using Channel State Information (CSI). The 3rd Generation Partnership Project (3GPP) Release 19 identifies 32 potential ISAC use cases, with particular emphasis on detecting and tracking moving objects. In this work, we address the Sensing for Railway Intrusion Detection use case, where intruders, including wildlife, entering a railway track can pose serious collision risks. We generated 22,695 CSI matrices with corresponding ground truth using a 3D-rendered railway environment and the Sionna radio simulator. We developed a machine learning model combining a three-dimensional Convolutional Neural Network (3D CNN) and Bidirectional Long Short-Term Memory (BiLSTM) network to detect intruders in the track danger zone and estimate their real-time position relative to the train, velocity, and time to collision. On synthetic CSI data, the model achieves 99.57% intruder-detection accuracy on a balanced test set and a combined Mean Absolute Error (MAE) of 0.4240 for position, velocity, and time-to-collision prediction. These results demonstrate the potential of CSI-based ISAC sensing with machine learning for reliable railway intrusion detection. The complete codebase for CSI generation, preprocessing, and model development is publicly available at https://github.com/EdgeIntelligenceLab/6g-isac-railway-intrusion-detection. --- ### EmpaAva: An Open-source Agentic 3D-Avatar Empathetic Live Chatbot Authors: Jie Yang, Wenhao Xu, Shuhui Lin, Hao Fei Published: 2026-08-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.04709v1 This paper presents EmpaAva, to our knowledge the first open-source, agentic 3D-avatar empathetic chatbot, which carries empathetic response generation (ERG) from text-only exchanges into live, face-to-face interaction. Through a video-call-like interface, a user speaks to a 3D digital human that reads their affect from speech and optional vision, and replies with emotional speech, lip-synced facial motion, and photorealistic 3D Gaussian rendering. At its core, an LLM coordinates a Tri-Agent Architecture, in which perception, empathetic response planning, and embodied rendering form a closed loop, paired with a Response Planning layer that compiles each reply into an executable multimodal plan, keeping voice, expression, and rendering on one empathetic intent. Building on strong open-source modules, EmpaAva supplies the intelligence that binds them into one controllable, inspectable experience. In automatic and human evaluations, EmpaAva surpasses text-only, 2D talking-face, and multimodal avatar baselines in emotion understanding, response quality, and audio-visual consistency. We open-source EmpaAva with an online live demo. --- ### Benchmarking Deep Learning Models for Dense Event Classification of Offshore Wind Infrastructure in Sentinel-1 Time Series Authors: Thorsten Hoeser, Felix Bachofer, Claudia Kuenzer Published: 2026-08-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.04706v1 Monitoring of offshore wind energy infrastructure life cycles, especially during the deployment phase, is an important contribution for stakeholders to make informed decisions in a phase of increasing deployment activities. ESA's Sentinel-1 Synthetic Aperture Radar (SAR) mission produces large data archives that enable the global monitoring of offshore wind infrastructure. Turning these high-volume archives into information requires algorithms that automatically extract single event labels from dense time series at a global scale. In this study, we present a structured comparison of ten deep learning model-training variants for the dense classification of Sentinel-1 based offshore wind infrastructure time series, aiming to advance rule-based event classification of this task. We trained LSTM, Transformer, and fully connected model variants with monotemporal, unidirectional, and bidirectional context awareness, each with and without self-supervised pretraining. Among these, the supervised BiLSTM performs best, raising the target AUC score from 0.7853 for the rule-based baseline to 0.8509, and the perfect match rate from 0.3508 to 0.5063. Combining the BiLSTM predictions with the existing baseline labels in a label-transition-minimising ensemble further improves agreement with the test data. Using these improved labels, we isolate the deployment phase of individual turbines at a global scale and conduct a regional and subregional analysis covering 2016-01-01 to 2025-03-31, reporting median deployment durations of 84 d (China), 242 d (EU), and 258 d (UK). Deployment-related drivers, including legal regulations such as subsidies, and environmental conditions, emerge clearly from the analysed results across multiple spatial scales. --- ### A Multi-Sensor Dataset for Monitoring the Operational Environment of Rail Vehicles Authors: Claudio Diotallevi, Rodrigo Gudiño, Zaharia Pachalieva, Philipp Neumaier, Patrick Naumann, Erik Bochinski, Volker Eiselein, Martin Köppel Published: 2026-08-05 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2608.04704v1 Reliable environment monitoring is essential for the safe and efficient operation of automated railway systems, covering all Grades of Automation (GoA), from partially automated (GoA2) to fully automated operation (GoA4). Artificial Intelligence (AI) plays a central role in enabling these systems to detect, classify, and react to potential hazards in real time. The development of such AI-based perception systems requires large volumes of accurately annotated data for training and validation. Within the Digitale Schiene Deutschland (DSD) program, DB InfraGO AG and understandAI GmbH have developed a comprehensive multi- sensor dataset tailored to the needs of railway environment perception. This dataset contains over 7 million high-quality annotations of both railway-specific and general perception objects, captured under varying operational scenarios. The finalized dataset can now be requested at the DB InfraGO AG and serve as a valuable resource for advancing AI-driven environment monitoring in the railway domain. --- ### IslamicTurathBench: A Multi-Task, Multi-Discipline Benchmark for Evaluating Large Language Models on the Islamic Scholarly Tradition (turath) Authors: Shahd Gaben, Heba Sbahi, Samer Rashwani, Abdessalam Bouchekif, Mutaz Al-Khatib, Emad Mohamed, Somaya Eltanbouly, Mohammed Ghaly Published: 2026-08-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.04703v1 Large language models (LLMs) are increasingly used for question answering, education, and research, including in religious and cultural domains where answers depend on specialised source traditions. Yet in Islamic Studies, key concepts, methods, and debates preserved in the authoritative scholarly tradition, known as turath, lack high-quality annotated resources. We introduce IslamicTurathBench (ISTB), a multi-task, multi-discipline dataset for evaluating LLMs on classical Islamic scholarship. Developed and reviewed by domain experts, ISTB contains 3,465 question-answer items drawn from 35 recognised source works spanning more than 12 centuries of scholarship across seven key fields of Islamic Studies. To enable comprehensive profiling of model capabilities, ISTB is structured along two axes: scholarly demand (Beginner, Intermediate, and Advanced) and task format (multiple-choice questions, passage-based comprehension, and open-ended knowledge questions). ISTB includes aggregated scores from a scholarly human reference panel and zero-shot baselines from ten systems. The dataset supports reproducible evaluation of language-model behaviour across source works, disciplines, scholarly-demand levels, and question formats in a historically layered scholarly domain. --- ### Design Choices That Matter: A Functional ANOVA Analysis for Remote Sensing Multi-Label Classification Authors: Maryam Gholami Shiri, Eva Tuba, Sašo Džeroski, Tome Eftimov, Ana Nikolikj Published: 2026-08-05 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2608.04702v1 Benchmarking deep learning (DL) models for multi-label classification (MLC) of remote sensing images (RSI) typically yields rankings that do not generalize beyond the evaluated datasets. In this work, we move beyond rankings by employing functional analysis of variance (fANOVA) to systematically quantify the contributions of individual design choices and their interactions to performance variability. We conduct two empirical analyses covering 48 and 20 DL models, respectively, spanning design choices such as network architecture, fine-tuning strategy, learning strategy, and initialization. By applying fANOVA across seven MLC RSI datasets, we construct dataset meta-representations that capture design-choice sensitivity profiles. Hierarchical clustering of these meta-representations reveals that datasets naturally group according to how they respond to design decisions, with patterns strongly linked to intrinsic dataset properties such as scale, spatial resolution, and label space complexity. Our findings show that for large-scale datasets, fine-tuning strategy and architecture are dominant factors, while in data-limited regimes, initialization becomes decisive. For intermediate regimes, the interaction between architecture and learning strategy governs performance. --- ### UniWorld-View: Large-Baseline View Synthesis via Video Diffusion Models Authors: Haiyang Zhou, Wangbo Yu, Chaoran Feng, Xunyu Zhou, Yonghong Tian, Li Yuan Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04701v1 The abundance of casually captured monocular videos and images on social media provides a valuable source for immersive content creation, where generating novel views from such sparse observations can greatly enhance user experiences. However, producing photorealistic and geometrically consistent views with precise camera control remains challenging when input coverage is extremely limited. Reconstruction-based approaches such as NeRF and 3D Gaussian Splatting (3DGS) deteriorate severely under sparse inputs and fail to explicitly handle occlusions. Generative methods ease data requirements but still struggle with large-baseline view synthesis due to inaccurate or implicit geometric guidance. To overcome these limitations, we introduce UniWorld-View, a unified framework for controllable large-baseline novel view synthesis from monocular inputs. UniWorld-View integrates explicit 3D guidance with generative diffusion modeling to enable precise camera control and geometrically consistent view generation. The geometric guidance is obtained through an occlusion-aware point cloud rendering strategy that resolves visibility ambiguities and provides accurate priors for diffusion-based synthesis. By coupling this rendering strategy with powerful video diffusion backbones, UniWorld-View achieves high-fidelity novel view generation even under extreme camera motions and wide-baseline changes, and can further provide multi-view videos for downstream dynamic 3DGS reconstruction. Experiments on the WorldScore benchmark and zero-shot NVS benchmarks demonstrate the effectiveness of UniWorld-View in controllability, geometric consistency, and visual fidelity. --- ### Teaching MLLMs to Say No: Generalized Referring Expression Comprehension via Refusal Calibrated GRPO Authors: Xuzheng Yang, Jun Ling, Tao Huang, Caiyan Qin, Peng Wang Published: 2026-08-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.04698v1 We tackle the challenging yet underexplored task of Generalized Referring Expression Comprehension (GREC), which requires a model to localize the object described by a textual expression when it exists (positive sample) and to refuse output when it does not (negative sample). Although Multimodal Large Language Models (MLLMs) excel at localizing existing objects, they often fail to reject nonexistent ones due to the absence of negative samples during training, producing hallucinated bounding boxes. Existing post-training approaches such as supervised fine-tuning (SFT) and reinforcement learning (RL) enhance refusal behavior but usually degrade localization accuracy on positive samples, undermining the model's core competence. To address this, we propose Refusal-Calibrated Group Relative Policy Optimization (RC-GRPO), a calibrated RL strategy that strengthens the refusal ability of MLLMs while preserving localization performance. It enforces "None" outputs in rollouts for valid advantage estimation on negative samples and applies a penalty to prevent over-refusal on positives, achieving a balanced trade-off between accuracy and reliability. A second-stage reasoning reinforcement further consolidates causal understanding and interpretability. Experiments on three GREC benchmarks demonstrate that RC-GRPO attains superior localization accuracy while maintaining strong refusal capability. --- ### Traceable LLM-Generated Hazard Scenarios for Operational Safety Analysis of Aviation Systems Using ASRS Reports Authors: Cristian Mascia, Roberto Pietrantuono, Daniel Rodriguez, Stefano Russo Published: 2026-08-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.04697v1 Operational hazard analysis of aviation system operations must consider interactions among weather, ATC actions, airspace constraints, aircraft operations, and human factors - distinct from the functional hazard assessment applied at the aircraft-system level. We present an AI-assisted approach that generates candidate hazard scenarios from NASA's Aviation Safety Reporting System (ASRS). Given a target adverse outcome, it produces a structured hypothesis as categorical factors and a narrative scenario describing an operational event sequence consistent with the structure. Each scenario includes by a plausibility score from historical co-occurrence evidence and traceability to the most similar held-out ASRS reports. We then propose a hybrid variant, conditioning narrative generation on a structured hypothesis produced via evolutionary abduction, improving correctness and reducing variability. We evaluate multiple large language models, zero-shot versus few-shot prompting, and optional fine-tuning, measuring how prompting and model choice affect the validity and realism of the generated structures and narratives. --- ### Personalized Federated Sparse Adaptation of Time-Series Foundation Models Authors: Priyanka Nihalchandani, Naman Srivastava, Varun Ojha, Pandarasamy Arjunan Published: 2026-08-05 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2608.04695v1 Federated adaptation of time-series foundation models (TSFMs) is attractive for building energy forecasting because meter data are private, distributed, and highly non-IID. However, a single parameter-sharing strategy is unlikely to serve all pretrained TSFMs or building clients: fully shared adapters can suppress building-specific temporal behavior, while fully local adaptation discards cross-building transfer. We propose a personalized federated sparse adaptation framework with a heterogeneous temporal mixture-of-experts (MoE) adapter placed after the pretrained TSFM representation. A sequence-level router maps each 168-hour context window to a top-$k$ subset of experts specialized for periodicity, long-range interactions, local variation, trend-residual structure, and multi-resolution behavior. We compare global FL, local training, and personalized FL variants with globally shared or client-private expert banks. Across 50 buildings and three TSFM backbones, personalization consistently outperforms Global FL-MoE and Local MoE, while the best sparse-adaptation strategy varies by backbone and metric. Routing behavior further reveals client-level expert specialization, expert concentration, and near-uniform routing across backbones, showing that federated TSFM adaptation should be both client-aware and backbone-aware. --- ### Suppression Sticks, Locality Is Fragile: A Closed-Loop Target-and-Control Audit of Task-Vector Negation in VLA Policies Authors: Shaoguang Wang, Weiyu Guo, Rushi Dai, Yiren Zhao, Yandong Guo, Hui Xiong Published: 2026-08-05 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2608.04692v1 Task-vector arithmetic offers a closed-form way to modify a model, yet its behavioral locality remains unclear in closed-loop robot control. We present a target-and-control audit of per-skill task-vector subtraction from multitask vision-language-action (VLA) policies. Across all ten LIBERO-Goal skills, subtraction produces three qualitatively different regimes: target-control separation for five skills, resistance for three, and global collapse for two. On held-out initial states, the five suppressible targets remain at 0% success; however, mean baseline-normalized control retention is only 52%, and each target-suppressing edit materially harms at least one nominally unrelated control. Additional Goal panels show separation across tested policies with continuous-regression, discrete-token, and flow-matching action heads, whereas we observe no clean separation on Spatial and control collapse on the tested Object and Long-horizon panels. Mean task-vector cosine does not account for this variation. A matched-norm control identifies a local sign asymmetry around one Goal anchor, while multi-vector outcomes vary with anchor and scale. Retain-aware gradient baselines provide data-dependent comparators but require removal-time data and optimization; subtraction is data- and gradient-free only at edit time, assuming precomputed expert deltas. Finally, a single-skill relearning probe is consistent with behavioral masking, not certified unlearning. These results characterize task-vector subtraction as a fast but brittle intervention and underscore the need for closed-loop target-and-control evaluation when assessing locality in embodied model editing. --- ### The Sample Complexity of Distributionally Robust PAC Learning under Cressie--Read Divergences Authors: Elad Aigner-Horev, Daniel Rosenberg, Roi Weiss Published: 2026-08-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.04686v1 We study distributionally robust PAC learning for the $0$--$1$-loss, where adversarial perturbations of the data distribution are constrained by a Cressie--Read divergence of order $k>1$ and radius $ρ\geq 0$. For hypothesis classes with VC dimension $d$, we establish realizable and agnostic sample-complexity bounds tight up to constant and logarithmic factors, respectively; ordinary empirical risk minimization attains both rates up to logarithmic factors. For target accuracy $\varepsilon\in(0,1)$ and confidence $δ\in(0,1)$, their respective orders are \( \max\!\left\{\frac{1}{\varepsilon}, \frac{ρ^{\frac 1{k-1}}}{\varepsilon^{k_\star}} \right\}\cdot(d+\log δ^{-1}) \qquad\text{and}\qquad \max\!\left\{\frac{1}{\varepsilon^2}, \frac{ρ^{\frac1{k-1}}}{\varepsilon^{k_\star\vee 2}} \right\}\cdot(d+\log δ^{-1}), \) where $k_\star={k}/{(k-1)}$. For every fixed $ρ>0$, robustness changes the realizable $\varepsilon$-dependence from $\varepsilon^{-1}$ to $\varepsilon^{-k_\star}$ as $\varepsilon\downarrow0$. In the agnostic case, for $1<k<2$, robustness changes the $\varepsilon$-dependence from $\varepsilon^{-2}$ to $\varepsilon^{-k_\star}$, whereas for $k\geq2$ the exponent remains the classical $2$, with nontrivial $ρ$-dependence. Building on the known scalar reduction of robust $0$--$1$ risk to ordinary classification error, our analysis reveals a scale-sensitive interaction between the statistical estimation of classification error and its amplification by robustness, sharply explaining the transition in the agnostic rate. We extend the previously studied $χ^2$-divergence case to every Cressie--Read order $k>1$, close its upper--lower gaps, and recover standard PAC learning rates as $ρ\to0$, unlike previous bounds that fail to interpolate correctly in this limit. --- ### Active-SWE: Benchmarking Coding Agents for Proactive Bug Fixing without Issue Reports Authors: Haobin Li, Ping Deng, Weizhong Qian, Liang Jiang, Zhenyu Huang, Mouxing Yang, Xi Peng Published: 2026-08-05 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.04682v2 Coding agents powered by large language models (LLMs) are increasingly adopted in software engineering (SWE) scenarios, capable of fixing a specific bug in large-scale codebase. However, existing SWE benchmarks typically assume that high-quality issue reports with detailed information are always available, which is easily violated in practice due to the complexity of report acquisition and curation. To address this, we introduce Active-SWE, a benchmark for evaluating coding agents on proactively discovering and fixing multiple bugs without report guidance, covering 1,663 tasks across six bug categories and eight languages. Beyond shifting the focus from existing reactive bug fixing to proactive bug fixing, Active-SWE enables a more in-depth evaluation by expanding the scope from fixing a specific recorded bug to multiple-bug fixing and potential bug discovery scenarios. To construct Active-SWE, we propose a novel difficulty-aware task formulation pipeline with a dual-track evaluation framework, facilitating comprehensive evaluation of proactive bug-fixing capability. Extensive experiments reveal that most state-of-the-art coding agents struggle with proactive bug-fixing tasks, demonstrating limited performance in locating and resolving recorded bugs, handling multiple bug fixing scenarios, and discovering valid potential bugs. --- ### Active-SWE: Benchmarking Coding Agents for Proactive Bug Fixing without Issue Reports Authors: Haobin Li, Ping Deng, Weizhong Qian, Liang Jiang, Zhenyu Huang, Mouxing Yang, Xi Peng Published: 2026-08-05 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.04682v1 Coding agents powered by large language models (LLMs) are increasingly adopted in software engineering (SWE) scenarios, capable of fixing a specific bug in large-scale codebase. However, existing SWE benchmarks typically assume that high-quality issue reports with detailed information are always available, which is easily violated in practice due to the complexity of report acquisition and curation. To address this, we introduce Active-SWE, a benchmark for evaluating coding agents on proactively discovering and fixing multiple bugs without report guidance, covering 1,663 tasks across six bug categories and eight languages. Beyond shifting the focus from existing reactive bug fixing to proactive bug fixing, Active-SWE enables a more in-depth evaluation by expanding the scope from fixing a specific recorded bug to multiple-bug fixing and potential bug discovery scenarios. To construct Active-SWE, we propose a novel difficulty-aware task formulation pipeline with a dual-track evaluation framework, facilitating comprehensive evaluation of proactive bug-fixing capability. Extensive experiments reveal that most state-of-the-art coding agents struggle with proactive bug-fixing tasks, demonstrating limited performance in locating and resolving recorded bugs, handling multiple bug fixing scenarios, and discovering valid potential bugs. --- ### MOAT: Model-Agnostic Randomized Transformations for preventing Efficiency Degradation Attacks on ViTs Authors: Anadi Goyal, Nandish Chattopadhyay, Chandan Karfa, Anupam Chattopadhyay, Norrathep Rattanavipanon Published: 2026-08-05 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2608.04680v1 To adopt the Vision Transformers (ViTs) in resource-constrained environment, token pruning is widely used to reduce computational cost without impacting accuracy. However, adversaries have developed targeted attacks against said token pruning techniques to undermine such attempts to make ViTs efficient. In this paper, we propose MOAT, a model-agnostic pre-processing defense pipeline that applies a combination of input transformations to protect efficient ViT implementations against adversarial efficiency attacks. MOAT operates directly on the input without requiring modifications to the model architecture or token pruning mechanism. Experimental results demonstrate that, across all evaluated ViT models, MOAT limits GFLOPs degradation under adversarial attacks to within 3.4% of the original unattacked model. --- ### Kathleen Writes: Autoregressive Generation and Data Scaling Without Attention Authors: George Fountzoulas Published: 2026-08-05 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2608.04678v1 Papers 1-2 of the Kathleen series showed that a byte-level, attention-free architecture built from a wavetable encoder and multi-scale reverberant state can match strong baselines on classification at ~450-700K parameters, without pretraining. We ask whether the same ingredients can generate. (1) Scaling: on byte-level language modeling (WikiText-103, raw UTF-8, no tokenizer), the reverberant model beats a parameter-matched transformer at every dataset scale measured (2-512 MB), e.g. 1.84 vs 2.04 bits/byte at 512 MB with ~0.5M parameters; the transformer needs more than 512 MB to match what the attention-free model learns from 32 MB. (2) Measurement: we introduce FORM DISTANCE, a non-parametric, gaming-resistant instrument for "reads like text": nine statistical axes of human text define a reference cloud, and five constructed fakes are all rejected. (3) Generation: decoding policy dominates architecture -- widening the sampler halves the same model's distance (3.17 to 1.52), and a retrieval-augmented decoding scheme takes the frozen model further (1.52 to 1.14) with no training step involved; the ablation attributes the gain to the sparse phrase dose itself, not the selection gate. The gain has a sharp boundary condition: the phrases must come from the model's own training corpus -- a 40x larger foreign library helps not at all, an effect the attention twin shares, consistent with in-context integration being a capability of scale. We also report four architectural additions that did not help, and a computed lexicon reaching 94% of a learned table's top-1 accuracy at one fifth of the parameters. Everything runs offline; all experiments are reproducible on a free Kaggle T4. --- ### Diverse and Plausible Algorithmic Recourse via Tractable Recourse Distributions Authors: Anagha Sabu, Hrithik Suresh, Narayanan C. Krishnan Published: 2026-08-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.04677v1 Algorithmic recourse seeks to help individuals reverse unfavorable automated decisions by recommending actionable changes that achieve a desired outcome. As an individual usually has several distinct routes to a favorable decision, and different people can act on different ones, a recourse system should offer multiple realistic alternatives rather than one. Existing approaches formulate recourse as an optimization problem that constructs one or a small set of counterfactuals rather than modeling the underlying space of feasible solutions, and in practice each sacrifices diversity, plausibility, or feasibility to secure the others. We propose Tractable Recourse Distributions, a probabilistic framework that represents the space of feasible alternatives for a given factual instance as a probability distribution over favorable outcomes. For commonly used cost functions based on proximity and the number of feature changes, we show that this distribution admits an exact representation as a probabilistic circuit, obtained by exponentially tilting the circuit; each individual's distribution is therefore available in closed form, without retraining the model. Sampling from these distributions naturally produces diverse and plausible recourses, while the tilting parameters provide explicit control over their proximity and sparsity. Experiments on standard algorithmic recourse benchmark datasets demonstrate that the proposed framework attains diversity, plausibility, and feasibility simultaneously, while retaining sufficient probability mass over feasible counterfactuals for rejection sampling to be practical. A visual study on MNIST illustrates how the tilt strength trades proximity against validity. --- ### SurgNarrator: A Generative Retrieval Framework for Surgical Video Understanding Authors: Yuqing Feng, Jiawei Ma, Kevin Qinghong Lin, Kun Yuan, Nicolas Padoy, Daniel S. Elson, Anh Nguyen, Stamatia Giannarou, Baoru Huang Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04676v1 Surgical procedures unfold as structured and recurring clinical events, whose real-time understanding via intraoperative surgical videos is critical for intraoperative decision-making and support. However, existing video understanding methods force a trade-off: autoregressive video-language models support comprehensive reasoning but are not practical for time-sensitive clinical applications, whereas contrastive models offer low latency but struggle with complex scene understanding. Recently, generative retrieval has been explored for general-domain video understanding, but transferring it to surgery is not trivial because near-identical visual appearances may indicate semantically distinct events, and the terminology involved is highly surgery-specific. To this end, we propose SurgNarrator, a new generative retrieval framework tailored for surgical video understanding. We construct a well-curated surgery-centric vocabulary from surgical captions to define a clinically meaningful retrieval space. We then adapt the pre-trained Qwen3-VL-Embedding-8B to learn discriminative clinical representations with a temporally-aware contrastive objective. During inference, a hierarchical, procedure-aware retrieval strategy narrows the search space to the relevant procedure type, delivering fast and effective responses. Our method is comprehensively evaluated on twelve benchmarks in a zero-shot setting and achieves consistent performance gains over state-of-the-art baselines, while reducing output-stage latency by more than two orders of magnitude compared with the generative baseline. --- ### Differential 6-DOF Pose Estimation with Provable First-Order Immunity to Camera Calibration Errors Authors: Yueqiang Zhang, Liang Deng, Yi Zhang, Baoqiong Wang, Wenjun Chen, Shuixin Pan, Yulan Guo, Qifeng Yu Published: 2026-08-05 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2608.04673v1 Accurate six-degree-of-freedom (6-DOF) motion estimation is essential for robotic manipulation, autonomous systems, and structural displacement monitoring. Conventional 3D-2D methods estimate absolute camera poses independently at each time and recover platform motion through camera-to-platform extrinsics, making them sensitive to extrinsic calibration errors, especially for micromotion. We present a differential pose estimation method that directly recovers platform motion from inter-frame image displacements and known 3D control points. By differencing perspective projection equations, using a depth-invariance approximation, and modeling motion on SE(3), the method avoids independent absolute-pose estimation and supports both monocular and multi-camera systems. We prove that translational extrinsic errors cancel exactly, while rotational errors induce a bounded perturbation determined by calibration error, motion magnitude, and observation geometry. We also derive generic observability conditions, a Cramer-Rao lower bound, and a bias-eliminated consistent estimator, and characterize the validity limits of the approximations. Extensive synthetic and real-world experiments establish a new state of the art for 6-DOF platform micromotion estimation, outperforming representative PnP and generalized-PnP methods in accuracy, calibration robustness, and computational efficiency. With five control points and 0.5-pixel image noise, the monocular solver obtains a combined pitch-yaw rotation RMSE of 10.09 arcsec, a translation RMSE of 3.70 mm, and a runtime of 0.34 ms. The binocular solver achieves a rotation RMSE of 10.58 arcsec, a translation RMSE of 3.91 mm, and a runtime of 0.27 ms. Code will be released upon publication at https://github.com/zyoungszu/pami2026. --- ### Easy to Complete, Hard to Choose: Investigating LLM Performance on the ProverbIT Benchmark Authors: Enrico Mensa, Lorenzo Zane, Calogero Jerik Scozzaro, Matteo Delsanto, Tommaso Milani, Daniele Paolo Radicioni Published: 2026-08-05 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2608.04670v1 Large Language Models (LLMs) have transformed computational linguistics and achieved remarkable performance across numerous natural language processing tasks, yet significant gaps persist in understanding how these systems process culturally embedded linguistic expressions. This paper introduces ProverbIT, a novel Italian benchmark comprising 100 multiple-choice questions designed to evaluate LLMs' ability to complete Italian proverbs. We assess 13 frontier models, including Large Reasoning Models (LRMs) and traditional LLMs, across three tasks: proverb completion, multiple-choice selection with correct answers, and multiple-choice selection without correct answers. Our evaluation reveals surprising results: while nearly all models demonstrate knowledge of the proverbs through successful completion tasks, performance drops dramatically when transitioning to multiple-choice formats without correct answers, with even state-of-the-art reasoning models showing substantial degradation. Through detailed Chain-of-Thought analysis of two LRMs, we uncover that models exhibit a strong bias toward selecting literal synonyms and frequently mention correct proverb endings during reasoning without successfully identifying their absence from the given options. These findings suggest that current LLMs rely heavily on memorized patterns rather than deeper semantic understanding of culturally grounded expressions, highlighting important limitations in their reasoning capabilities for figurative language comprehension. --- ### Differentiating Through Dual Prices: End-to-End Policy Learning Under Capacity Constraints Authors: Mohammadsaeed Haghi, Mahdi Salmani, Nima Kelidari Published: 2026-08-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.04669v1 Many social services assign scarce resources, such as housing assistance or hospital interventions, to people who arrive one at a time: each arrival must receive a decision immediately, and the long-run usage of every resource must stay within its capacity. We study how to learn such an assignment policy from logged observational data. The standard pipeline is decision-blind: fit one outcome model per arm by regression, price each capacitated resource from the fitted models, and assign each arrival the arm whose predicted outcome minus price is largest. We instead train the outcome models end-to-end, differentiating an off-policy estimate of the deployed policy's value through the dual prices themselves. We study two formulations: an exact nonconvex one, and a convex relaxation whose optimum always satisfies the capacity constraints in expectation and which is suboptimal by at most a term linear in the smoothing temperature and logarithmic in the number of arms. Every method is evaluated in a queueing simulation with resources replenished at their capacity rates. Across six datasets, the two end-to-end variants take the top slots on a deployment-adjusted value index at every delay cost, including zero; when capacities are binding, decision-blind baselines frequently violate them and incur much longer queueing delays. On the largest dataset, a hospital cohort of seventy thousand patients, end-to-end training also achieves significantly higher policy value, a margin that survives a capacity-matched neural baseline. Flexible decision-blind regression remains the stronger pure predictor where ground truth is measurable; end-to-end training is best suited to settings where resources are genuinely scarce and feasibility matters. --- ### Automatic Statistical Test for Rationally Expressible Algorithms by Selective Inference, with Applications to Feature Selection Authors: Teruyuki Katsuoka, Tomohiro Shiraishi, Shuichi Nishino, Ichiro Takeuchi Published: 2026-08-05 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2608.04667v1 Selective inference (SI) provides statistically valid $p$-values for hypotheses selected by applying an algorithm to the data, correcting for the bias that arises when the same data are used both to select and to test a hypothesis. Developing an SI procedure for a new algorithm, however, has required an expert to derive, and then implement, the selection event, i.e., the conditions under which the hypothesis is selected. Repeating this specialized effort for every new algorithm is why exact SI has so far been available for only a narrow class. We propose AutoSI, a framework that removes this barrier in two ways. First, AutoSI constructs the selection event automatically from the algorithm's individual operations, so the user only writes the algorithm as ordinary NumPy-like code and derives nothing by hand. Second, AutoSI broadens the class of selection events SI can handle: existing exact methods are limited to selection events characterized by linear or quadratic inequalities in the data, whereas AutoSI covers any algorithm expressible through rational functions of the data (ratios of polynomials). We prove that the $p$-values computed by AutoSI are exactly valid in finite samples. We demonstrate AutoSI on three feature-selection methods, each written in a few dozen lines of code. One of these methods, the lasso with its tuning parameter selected by cross-validated $R^2$, cannot be handled within existing exact SI frameworks and is made possible by AutoSI. Experiments on synthetic and real datasets show that the resulting $p$-values control the type I error rate (i.e., the false positive rate) at the nominal level while retaining high power. --- ### Calibrating Artificial Guilt: Neurally Grounded Reward Shaping for Prosocial Multi-Agent Reinforcement Learning Authors: Aaditya Mehta, Arya Shah Published: 2026-08-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.04663v1 Cooperative multi-agent reinforcement learning often adds social terms to individual rewards, yet the scale of those terms is usually chosen by hand. We ask whether a guilt signal can instead be calibrated from human neural and behavioural data and transferred to artificial agents. Using the public SoDec responsibility fMRI dataset (40 participants), we fit a subject-fixed-effects regression of momentary-happiness changes on outcome-type counts and recover a guilt weight as the Partner-negative minus Social-negative contrast ($\hat{w}=1.118$, Cohen's $d=0.214$). We embed this weight in a two-agent Social Lottery environment and train independent Proximal Policy Optimization actor-critics under four shaping regimes: neurally calibrated, uniform constant, zero (selfish), and a unit-coefficient oracle. Across 1{,}000 evaluation episodes per condition, the calibrated agents track the human Social safe-choice rate most closely ($0.459$ vs.\ human $0.484$; $\mathrm{KL}=0.0012$), while the other three conditions deviate by one to three orders of magnitude in KL. Human neurobehavioural priors can therefore act as quantitative constraints on prosocial reward shaping. --- ### MobileWAM: Bridging World Action Models to Mobile Manipulation with Chain-of-Foresight Authors: Zehua Fan, Junjie He, Wenxuan Song, Xi Wang, Wenqi Lyu, Linge Zhao, Fuhao Li, Zihan You, Yifei Yang, Kaiming Xu, Qi Jiang, Yue Jiang et al. Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04657v2 World action models (WAMs) built on video generation backbones are a rising recipe for robot learning, yet remain confined to tabletop manipulation. Mobile manipulation demands simultaneous locomotion and whole-body manipulation amid scene-scale dynamics, yet is still dominated by dynamics-blind visual encoders with hand-crafted coordination. We bridge this gap with MobileWAM, a mixture-of-transformers architecture that fuses a pretrained video diffusion transformer with a lightweight action expert through layerwise joint attention, translating internet-scale motion priors into whole-body control. To reconcile the heterogeneous dynamics of moving and manipulating, each feed-forward layer of the action expert becomes a three-expert mixture of shared, locomotion, and manipulation experts, softly routed by the motion intent in the action tokens. To densify supervision, we further propose Chain-of-Foresight (CoF): intermediate representations sequentially predict a chain of future latent chunks, each step conditioned on its predecessor. CoF pairs naturally with our decoupled video--action denoising scheme. At deployment, the WAM serves as a pure current-frame encoder; foresight acts only through gradients, so at inference the foresight chain and video generation are discarded, leaving only policy-level cost. MobileWAM surpasses state-of-the-art mobile manipulation policies on ManiSkill-HAB and fine-tunes to a real ARX Lift2 mobile manipulator across diverse tasks with strong generalization. Code will be released soon. --- ### CSGen: A Multi-Domain Curvilinear Structure Generation Model via Hierarchical Multimodal Diffusion Authors: Zhe Shan, Ziming Yang, Lei Zhou, Wenwen Zhang, Cong Lin, Xia Xie Published: 2026-08-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.04655v1 Curvilinear structure analysis is an important and fundamental task in multimedia. However, the controllable generation of images with precise curvilinear structure objects remains an open challenge. To address this, we propose CSGen, a hierarchical multimodal diffusion model that synthesizes high-fidelity images precisely aligned with multiple control conditions. The CSGen is built upon three key innovations: 1) We construct a multi-domain and multimodal dataset, including over 24K samples from 5 domains and 7 different types of annotations, to train the unified generation model. 2) We propose a novel hierarchical progressive control strategy that decouples topology clues from visual context by a phased signal injection, mitigating semantic drift while ensuring the topological integrity of sparse structures. 3) We design a sparsity-aware loss re-weighting mechanism to address the extreme sparsity of curvilinear structures, significantly enhancing the attention on thin and fragile structures during optimization. Extensive experiments demonstrate that CSGen generates images with superior structure accuracy and visual realism, significantly improving downstream segmentation performance while maintaining robustness across diverse prompts. Our results confirm CSGen as a scalable, data-centric paradigm for the analysis of complex curvilinear structures in diverse multimedia applications. Code and dataset are available at https://github.com/ShanZard/CSGen. --- ### Overcoming Statistical Bias in Action-Controllable World Models Authors: Yuhong Shi, Zhenhao Chu, Jie Wei, Jun Hao, Jianyi Liu, Jingwen Fu Published: 2026-08-05 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2608.04653v1 Action-conditioned world models aim to predict how visual environments evolve under an agent's actions. Yet future frames are often highly predictable from visual inertia and recurring motion patterns alone. This creates a shortcut: models can fit the data by exploiting statistical biases without making their visible dynamics meaningfully depend on the action. As a result, different actions may produce similar futures, while motion may persist even under zero action. The key question is how to reduce reliance on statistical shortcuts from dominating action-conditioned prediction. We argue that action control requires more than injecting action features; it requires enforcing consistency under counterfactual changes to actions and observations. Based on this insight, we introduce CoCo, a Counterfactual Consistency framework to enhance action controllability through two complementary constraints. Multi-step counterfactual consistency constrains reference, inverse-action, and zero-action rollouts, while action-spatial counterfactual consistency enforces consistent predictions under mirrored scenes and transformed actions. Together, they reduce reliance on statistical shortcuts from substituting for action-dependent dynamics. We further introduce Action Response Consistency (ARC) and Drift Energy (DE) to assess action controllability, together with Mini-SSMB for same-state, multi-action counterfactual evaluation. On Mini-SSMB, our full model achieved ARC_inv of 0.412 and ARC_ref of 0.483, while reducing DE by 17.07% relative to the baseline. On VP2 visual planning, it achieves the highest average success rate among SOTA models, at 73.1%. Experiments on BAIR and RoboNet further show that these gains preserve video prediction quality and transfer across model settings. --- ### DisMix: Order-Aware Mixup for Medical Imaging via Disentangling Ordinal and Non-Ordinal Features Authors: Dileepa Pitawela, Gustavo Carneiro, Hsiang-Ting Chen Published: 2026-08-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2608.04652v2 Image mixup is a widely adopted data augmentation strategy, yet it is ill-suited for ordinal classification tasks such as medical disease grading, where labels encode a progression of severity. By indiscriminately blending disease-severity cues (ordinal) with appearance-level variation (non-ordinal), standard mixup produces samples that distort the very ordinal structure that underpins clinical severity grading. We introduce DisMix, an order-aware mixup framework for ordinal classification. DisMix disentangles ordinal and non-ordinal features via a dual-codebook VQ-VAE, allowing each subspace to be mixed independently: ordinal codes are interpolated to produce meaningful intermediate ranks, while non-ordinal codes are varied to introduce appearance diversity without corrupting the ordinal signal. Across four medical imaging datasets, DisMix shows the best aggregate performance among six image mixup baselines paired with six ordinal classifiers and remains effective under data scarcity and clinical grading variability. --- ### Active Learning Guided Design Space Refinement for Scalable Multi-Objective Bayesian Optimization in Materials Discovery Authors: Alexandros Ntagiantas, Panagiotis Tsilimidos, George Giannakopoulos, Christoforos Rekatsinas, Panagiotis Krokidas Published: 2026-08-05 Categories: cs.LG, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2608.04651v1 Advanced materials discovery increasingly relies on machine learning and Bayesian optimization to explore large discrete design spaces under limited evaluation budgets. However, conventional Bayesian optimization (BO) can become inefficient as candidate spaces grow, often evaluating low-value regions before reaching informative areas. We propose an active-learning (AL)-guided adaptive search-space refinement framework combined with multi-objective BO to accelerate materials optimization while preserving Pareto-relevant regions. We evaluate the approach on CH4/N2 separation in covalent-organic frameworks and pressure-vessel design with material-direction stress components and thickness objectives. Results show that the AL-guided refinement reduces the candidate space by approximately half while preserving more than 99 percent of the original hypervolume. The reduced-space strategy improves early convergence and cumulative Pareto-front discovery from the BO, demonstrating efficient large-scale materials optimization across constrained autonomous materials discovery settings. --- ### Evaluating Theory of Mind in Reasoning Models: Robustness over Reasoning Authors: Ian B. de Haan, Peter van der Putten, Max van Duijn Published: 2026-08-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2608.04646v1 Large language models (LLMs) have recently shown strong performance on Theory of Mind (ToM) tests, prompting debate about the nature and validity of the underlying capabilities. At the same time, reasoning-oriented LLMs trained via reinforcement learning with verifiable rewards have demonstrated notable improvements across a range of benchmarks. In this work, we examine the behavior of such reasoning models in ToM tasks using novel adaptations of machine psychological experiments together with results from established benchmarks. We observe that reasoning models consistently exhibit increased robustness to prompt variations and task perturbations. Our analysis suggests these gains come at least partly from models being more robust at reaching the correct answer under prompt and task variation. We read this as evidence for a robustness-based account rather than for a new ToM-specific ability. --- ### YOLO-PVC: 2D-to-3D Consolidation of Slice-wise Detections for Volumetric Liver Tumor Localization in MRI Authors: Talha Waqas, Mounir Lahlouh, Kawther Taibouni, Mahnoor Waqas, Salar Ahmed, Sébastien Mulé, Yasmina Leroul-Chenoune Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04642v1 Slice-wise 2D object detectors are increasingly applied to volumetric data due to their computational efficiency and scalability, yet they often yield fragmented and unstable predictions along the depth axis. We propose YOLO-PVC, a lightweight and model-agnostic framework for 2D-to-3D consolidation of slice-wise detections. The method enforces depth continuity, aggregates bounding box coordinates using robust percentile statistics, and further refines axial extent through a lightweight MLP-based calibration module. Unlike naïve stacking or averaging strategies, YOLO-PVC explicitly addresses missing detections and outlier slices along the depth dimension. Experiments on 3D liver MRI volumes across three tumor categories demonstrate consistent improvements over multiple aggregation baselines. The heuristic PVC achieves an overall $\mathrm{IoU}_{3D}$ of $0.665$, while the calibrated variant further improves performance to $0.710$, with high planar overlap ($\mathrm{BEV\ IoU} \approx 0.78$). These results demonstrate that structured geometric consolidation provides an effective and practical solution for volumetric liver tumor localization in clinical MRI. --- ### AI Literacy for Legal Translation: Developing Digital Resilience Authors: Łucja Biel Published: 2026-08-05 Categories: cs.AI, cs.CL, cs.CY, cs.HC arXiv: https://arxiv.org/abs/2608.04641v1 Generative AI is transforming legal translation by introducing opportunities alongside linguistic, technical, legal, ethical and cognitive risks. This chapter examines the implications of AI for professional legal translation and proposes an AI literacy framework tailored to the profession. It argues that AI does not change the fundamental objectives of legal translation but requires an extension of professional competence through AI literacy. The proposed framework comprises four mutually reinforcing dimensions, foundational, procedural, critical and strategic, and conceptualises AI literacy as a transversal component of legal translation competence that fosters digital resilience. It further discusses the pedagogical implications of this framework by proposing classroom activities designed to develop AI literacy in legal translator education, enabling future translators to integrate AI critically, responsibly and in accordance with professional standards. --- ### A/B Agent: A Self-Evolving Agent for Strategy Iteration in Industrial A/B Testing Authors: Zhuohang Jiang, Yuxin Chen, Yongsen Pan, Zheng Hu, Wenqi Fan, Qing Li, Hongyang Wang, Jun Wang, Wenwu Ou Published: 2026-08-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.04625v1 Industrial recommendation strategy iteration heavily relies on large-scale A/B experimentation. Traditional tuning requires experts to repeatedly design strategies, configure experiments, analyze results, and adjust parameters, making the process labor-intensive and time-consuming. Meanwhile, valuable knowledge from historical experiments is often fragmented, making systematic reuse difficult through manual expert effort alone. Existing RAG agents partially alleviate this burden by retrieving prior strategies, but typically organize experience in a flat manner, overlooking the hierarchical relationships among business scenarios, recommendation stages, optimization objectives, and experimental contexts. This often results in mismatched retrieval and limited cross-scenario transfer, while preventing agents from continuously refining strategies and parameters through sequential A/B feedback. % To address these limitations, we propose A/B Agent, a closed-loop A/B agent for industrial recommendation strategy optimization. The framework comprises three tightly coupled core components: Historical Strategy Knowledge Organization, Autonomous Target-Aware Strategy Generation, and Experiment-Guided Strategy Self-Evolution. It organizes historical strategies into a hierarchical experience tree, retrieves transferable evidence through multi-path Tree-RAG to generate executable strategies, and continuously analyzes online A/B feedback to guide autonomous tuning and update the experience tree for self-evolution. Extensive offline and online evaluations demonstrate its effectiveness, including a 4.829% improvement in GMV in a real-world short-video e-commerce recommendation system while maintaining positive gains across all guardrail metrics. --- ### Masked diffusion enables coherent beat tracking Authors: Francesco Foscarin, Filip Korzeniowski, Richard Vogl Published: 2026-08-05 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2608.04624v1 Current neural networks for beat tracking generate invalid outputs, such as consecutive downbeats and erratic tempo changes, even when these are not present in the training data. Heavy post-processing techniques can alleviate these problems, but the original cause of this inconsistent behaviour remains unknown. We hypothesise that it stems from inadequate modelling of multiple plausible output beat grids, resulting in an invalid mixture of competing interpretations. We propose a masked diffusion approach that properly models multiple outputs and enables the model to build coherent predictions through iterative inference. We devise three modifications to standard masked diffusion that enable its application to beat tracking: independent masking of beats and downbeats during training and inference, a balanced masking scheduler for inference, and peak-picking across inference steps. Our approach reduces erratic behaviours and improves beat-tracking performance. --- ### Visual Anchoring in Diffusion: Multimodal Zero-Shot Skeleton Action Recognition Authors: Zehao Bao, Shujun Guo, Bruce X. B. Yu Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04623v1 Zero-shot Skeleton Action Recognition (ZSAR) remains ambiguous when unseen actions share similar skeleton joint dynamics but differ in objects or scene context. RGB provides these missing cues, yet existing multimodal methods typically maintain independent skeleton and RGB scoring branches and fuse their outputs. Without using unlabeled test data for adaptation or fusion calibration, a fixed fusion weight cannot capture class-pair-dependent modality reliability, while an adaptive rule lacks target-side feedback for deciding which branch should dominate. We bypass this weight-selection problem via the classify-by-generation paradigm, where each class is scored by how accurately a text-conditioned denoiser predicts the noise added to the skeleton feature. This formulation separates the progressively corrupted skeleton from fixed conditioning, allowing RGB and text to jointly condition a single class-scoring function rather than produce independent scores. We instantiate this idea as Multimodal Triplet Diffusion for Skeleton-Text Matching (TDSM-MM), augmenting a text-conditioned denoising Transformer with a non-diffused RGB condition token that serves as a stable visual anchor during skeleton data reconstruction. Our proposed TDSM-MM has been ablated via extensive experiments and achieved the best inductive accuracy on three of four NTU-60/120 splits and surpasses the transductive state-of-the-art on NTU-120 96/24 (i.e., 71.3% vs. 69.1%), without test-time adaptation, suggesting that diffusion-based methods can be a promising direction for zero-shot learning. --- ### DAC-Pose: Dual-Agent Collaborative Framework for Pose-Guided Human Generation Authors: Haotian Yang, Zhile Yang, Huiyu Zhou, Xin Sun Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04622v1 AI agents have emerged as a powerful new paradigm in generative image synthesis, enabling systems to perform complex semantic reasoning rather than passive pixel-level mapping. In pose-guided human generation, conventional methods inevitably produce severe visual artifacts under drastic viewpoint shifts, fundamentally because they lack the cognitive capacity to logically deduce unseen regions and model complex spatial deformations. To bridge this gap, we propose DAC-Pose, a novel agent-driven multimodal framework that reformulates single-view human generation as a collaborative dual-agent system. DAC-Pose integrates two complementary components, namely, the Prior Semantic Reasoning (PSR) agent and the Discrepancy-Aware Visual Encoding (DAVE) agent. Functioning as a cognitive engine, PSR utilizes collaborative reasoning to deduce the fine-grained attributes of unseen regions. Concurrently, acting as a specialized visual perception agent, DAVE quantifies and encodes viewpoint-induced spatial misalignments, continuously feeding robust spatial constraints back into the generative process. This autonomous feedback loop between semantic deduction and visual perception ensures high-fidelity detail synthesis. Extensive experiments on the DeepFashion and Market-1501 benchmarks validate the superiority of our agent-driven paradigm. Notably, DAC-Pose excels in preserving texture alignment and identity consistency under drastic viewpoint changes. The code is available at https://github.com/AIVRC/DAC-Pose. --- ### Agreement Before Diversity: Verification-First Complementarity for Heterogeneous Language-Model Coordination Authors: Ruitong Li, Binjie Guo, Aisheng Mo, Guowei Su, Jie Li, Ru Zhang Published: 2026-08-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2608.04618v1 Heterogeneous language-model ensembles expand the space of candidate responses, yet they lack a principled criterion for when a newly generated answer should supersede an already supported one. We decouple candidate headroom from replacement authority, rendering the latter as an explicit, auditable object. Our proposed method, Agreement-Before-Diversity (ABD), is a frozen, label-free decision rule: an anchor answer is retained if two additional trusted samples corroborate it under a fixed equivalence relation; otherwise, it is replaced by a heterogeneous synthesis. For this gating mechanism, we prove two exact identities. The first shows that the accuracy gap relative to unconditional synthesis is determined jointly by the agreement coverage and the anchor's advantage on the protected subset. The second shows that the gap relative to never synthesizing reflects a contrast between authorized recovery and authorized destruction. Neither identity assumes independence or calibrated confidence, and the expected inference cost is approximately eight minus five times the coverage in number of calls. Under blind, exact-ID evaluation, ABD achieves 59.43% on the complete LiveCodeBench-v6 (vs. 52.57% for Single9 and 52.00% for HAC; n = 175) and 75.00% on an untouched GPQA-Diamond split (both controls at 72.78%; n = 180). Furthermore, these identities localize every aggregate difference to an enumerable protected stratum: no discordant items occur among the 3 protected cases on LiveCodeBench, where coverage bounds the gate's contribution to 1.71 points a priori; 13 versus 8 discordant cases among 132 on GPQA-Diamond; and 12 versus 0 among 71 under a frozen anchor perturbation. Diversity supplies potential; verification structure supplies authority. --- ### An entropic explanation of insistence on sameness in autism Authors: Przemysław Śliwiński Published: 2026-08-05 Categories: cs.LG, q-bio.NC, stat.ML arXiv: https://arxiv.org/abs/2608.04616v1 An information theory-based framework is proposed in attempt to explain insistence on sameness in autism as an instance of a general behavior pattern in which an individual tries to reduce surprise and uncertainty. It offers a new definition of autism as an impairment in which cognitive functions are restricted to discrimination, memorization and prediction of tangible properties of the environment. An analogy between insistence on sameness and constrained minimization of the entropy metric is observed and examined for a set of assumptions that describe cognitive limitations of a person with autism. The metric is given by the formula $D_H(R, M) = H(R|M) + H(M|R)$, where $R$ represents sequences of random stimuli, $M$ is a memory that stores and retrieves them, and where $H(.|.)$ denotes their conditional entropies interpreted as surprise and uncertainty, respectively. It is first inferred that to minimize the metric an individual can learn about $R$ (and store that knowledge in $M$) or can restrict $R$ to the already known $M$. Then, it is concluded that insistence on sameness is a manifestation of the latter. Moreover, it is shown that the proposed framework: (1) Helps to quantify the concepts of surprise, uncertainty, sensory overload and deprivation, anxiety, comfort zone, disappointment, disorientation, pedantry, rigidness, observance or aberrant precision. (2) Leads to a list of guidelines for learning therapies and daily care routines, and allows them to be defined as optimization algorithms and implemented as programs for robotic live-in caregivers. (3) Can be validated with the help of a Turing test-like approach that requires no experiments involving individuals with autism. The framework-if positively validated-will provide formal foundations and design guidelines for therapies aimed at improving self-reliance of individuals with autism in basic activities of daily living. --- ### Why Ranking Anomaly Detection Algorithms Isn't as Reliable as You May Think Authors: Simon Klüttermann, Jérôme Rutinowski, Frederik Polachowski, Alice Kirchheim Published: 2026-08-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2608.04613v1 Anomaly detection is a safety-critical machine learning problem with applications ranging from fraud detection to network intrusion prevention and industrial monitoring. Despite the large number of proposed anomaly detection algorithms, many novel methods claim state-of-the-art performance. However, many authors do so under benchmark settings that are not aligned with one another. This lack of comparability raises concerns regarding the reproducibility and reliability of anomaly detection benchmarks. In this work, we study the impact of common benchmarking choices on the stability of algorithm rankings. Using seven representative anomaly detection algorithms and 690 datasets from the OddBench benchmark suite, we analyze how rankings change under varying dataset selections, evaluation metrics, hyperparameter configurations, and random seeds. To quantify this effect, we introduce a rank instability metric measuring the variability of algorithm rankings across benchmark settings. Our results show that algorithm rankings in anomaly detection are highly unstable. In many cases, almost every competitive algorithm can appear as the best-performing method under some benchmark configuration. Among the studied factors, dataset selection and hyperparameter choice contribute most strongly to ranking uncertainty, while random seeds and evaluation metrics have a comparatively limited impact. We also observe that reliable benchmarking requires substantially larger and more diverse dataset collections than the ones commonly used in prior work. --- ### The Order Is the Guarantee: Verifier-Budgeted Code Deletion with Static-First Learned Proposals Authors: Ruitong Li, Binjie Guo, Aisheng Mo, Guowei Su, Han Wang, Jie Li, Ru Zhang Published: 2026-08-05 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2608.04611v1 Frontier coding models now match or exceed strong human reference points on programming benchmarks, yet benchmark success does not imply maintainable software. Prompt-driven "vibe coding" is additive: new branches, guards, and fallbacks accumulate faster than obsolete logic is removed. We study the inverse problem-how an Al system should remove code when execution-verification capacity is finite. We formulate redundant-code reduction as proposal scheduling: a ranker orders single-statement deletion candidates, an execution suite accepts the first candidate that passes, and a budget bounds how many candidates may be tested. Our central observation is that candidate order, not model confidence, is the control surface a deployment can reason about. DELSCOUT instantiates two schedules. Given representative target-domain validation, a five-slot budget spends three slots on deterministic shortest-first candidates and two on complementary learned candidates; across nine MBPP replications with 0.5B, 0.6B, and 8B rankers this raises verified-deletion coverage by 9.5% relative (+6.7 accepted tasks) while consuming slightly fewer verifier calls than the matched static baseline. Without such validation the same rankers can lose coverage under shift, so we instead evaluate the complete static prefix first and append learned candidates only afterwards; for a deterministic verifier this makes coverage and character reduction non-decreasing by construction, at a measured 4.8-62.5% increase in verifier calls. MBPP+ then erases the in-domain advantage, showing that scheduling governs search while the test suite alone governs what "preserving behavior" means. The result is an auditable division of labor: models widen the search for removable code, order bounds the damage a mis-ranked proposal can do, and execution retains authority over every committed deletion. --- ### HiSC: Hierarchical Spatial Clustering Token Compression for Efficient 3D Scene Understanding Authors: Jiuhe Qu, Yingping Liang, Ying Fu Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04610v1 3D vision-language models (3D VLMs) enable spatial reasoning over multi-view scenes but suffer from substantial token redundancy due to duplicated observations and large uninformative regions, leading to high computational cost. Although visual token compression has shown promise in accelerating 2D VLMs, it fails to capture the structured nature of 3D scenes and leads to incomplete spatial coverage and loss of fine-grained details. In this paper, we propose \textbf{HiSC}, a training-free framework for hierarchical spatial clustering token compression in 3D VLMs. HiSC lifts token compression from token-level selection to cluster-level processing by organizing tokens into spatially grounded clusters using joint geometric and semantic cues. Specifically, we first introduce a \textbf{spatial graph-based merging (SGraM) strategy} that models cross-view redundancy as spatial connectivity and consolidates physically consistent regions, effectively merging extremely similar redundant tokens prior to LLM inference. We then propose a \textbf{spatial clustering-based pruning (SCluP) paradigm} within LLM inference, which performs hierarchical compression across clusters and within clusters, preserving object instance completeness while retaining fine-grained details for important regions. Extensive experiments on diverse 3D reasoning benchmarks show validate the effectiveness of HiSC, particularly under high visual token pruning ratios. Besides, HiSC achieves over 90\% token reduction with minimal performance degradation. Code is accessible at https://github.com/elecreak/HiSC. --- ### On MUON optimization: From non-convergence to an error analysis with Polar Express and the Newton-Schulz polynomial from implementations Authors: Thang Do, Steffen Dereich, Arnulf Jentzen Published: 2026-08-05 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2608.04607v1 Stochastic gradient descent (SGD) optimization methods are the standard instruments for the training of deep neural networks (DNNs). In many relevant artificial intelligence (AI) systems - such as popular large language models (LLMs)-not the standard SGD scheme is used as the optimization method but instead suitable accelerated variants of SGD are employed. One of the most popular methods of such accelerated SGD variants is the momentum orthogonalized by Newton-Schulz (MUON) optimizer proposed by Jordan et al. in 2024. The MUON optimizer exploits the special matrix structure of the weight parameters in the training of the DNNs and, in its original form, employs five Newton-Schultz (NS) matrix steps in each MUON iteration. In this work we propose and study a generalized variant of the MUON optimizer involving an arbitrary number of generalized NS steps with polynomials of possibly arbitrary high degree. The considered optimizer covers MUON with the original NS polynomial as well as MUON combined with the recently proposed Polar Express method as special cases. For a simple class of stochastic optimization problems (SOPs) we show for almost every mini-batch size that MUON fails to converge to the solution of the SOP as the number of gradient steps converges to infinity. We also establish an error analysis for MUON with the generalized NS steps that provides convergence rates in terms of the number of gradient steps and in terms of the size of the mini-batch. We illustrate our general error analysis for MUON in the case of several concrete examples including quadratic stochastic optimization problems (SOPs) as well as $\ell_2$ regularized logistic regression for binary classification. --- ### TRCoRSurg: Temporal-Relational Co-Reasoning for Surgical Video Triplet Recognition Authors: Fang Li, Shihao Zou, Weixin Si, Yang Gao, Shuai Li, Aimin Hao Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04606v1 Understanding complex surgical scenes requires recognizing multiple interdependent entities, such as instruments, actions, and targets, while maintaining their relational consistency across time. Existing surgical triplet recognition methods struggle to jointly model intra-frame label dependencies and inter-frame temporal semantics in a unified manner. To address these limitations, we propose a unified framework that integrates spatial, relational, and temporal cues for robust surgical triplet recognition. Specifically, class-specific spatial priors are first extracted through a multi-scale encoder. These priors are then refined by a Label Correlation Modeling module with multi-scale class activation map-guided relational extraction (MS-CAMRE), enabling the model to capture both static co-occurrence patterns and dynamic contextual dependencies among triplet components. Furthermore, a Bidirectional Temporal-Relational Fusion Attention (BTRFA) module harmonizes temporal and relational representations to achieve coherent temporal reasoning. We also introduce a new evaluation metric, the Triplet Consistency Error Rate (TCER), which quantitatively measures the model's ability to preserve causal and semantic consistency across triplets. Extensive experiments on the CholecT45 and ProstaTD datasets show that our method achieves state-of-the-art performance, improving AP_IVT by 5.1 percent and 7.8 percent, respectively. Moreover, according to TCER, our approach achieves relative reductions of more than 36 percent and 25 percent on the two datasets, respectively, demonstrating the effectiveness of our framework in temporal-relational co-reasoning. --- ### COSMO: Consensus-Driven Shift Modulation for Source-Free Domain Adaptation Authors: Bo Li, Junjie Peng, Xiaohua Xie, Jianhuang Lai Published: 2026-08-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2608.04604v1 Source-free domain adaptation (SFDA) adapts a source-trained model to an unlabeled target domain without source data, a practical setting under privacy or storage constraints. Yet its self-generated supervision can reinforce source bias under substantial domain shifts. Pretrained vision-language models (VLMs) offer complementary semantic knowledge, but the relative reliability of the source model and VLM varies across target samples. Existing cross-model guidance does not explicitly account for this variation and may overwrite valid source-derived evidence under conflict, a failure we term source-derived evidence forgetting. We formulate VLM-guided SFDA as a sample-wise reliability-allocation problem and propose Consensus-Driven Shift Modulation (COSMO). COSMO replaces expert-to-expert guidance with co-adaptation through an anchored shared consensus. It first forms a sample-specific initial consensus that favors the more concentrated prediction. During adaptation, COSMO re-aggregates both branches' evolving evidence and regulates how far the resulting consensus moves from its initial anchor based on consensus uncertainty and training progress. This keeps the shared supervision anchored yet adaptive. Across four benchmarks, COSMO achieves state-of-the-art performance under matched VLM backbones. Further analyses indicate that it better balances the retention of valid source-derived evidence with the absorption of complementary VLM evidence. ---