# Research Papers — 2026-09 · bundle page 195 of 266 > Full-text markdown of papers #9701–9750 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/195 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/194.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/196.md) ### CEM-TUDASR: Computationally efficient multi-modality transformer based unsupervised domain adaptive super-resolution approach Authors: Anjali Sarvaiya, Jay Kadel, Kishor Upla, Kiran Raja Published: 2026-09-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.11201v1 Wireless Capsule Endoscopy (WCE) enables non-invasive visualization of the gastrointestinal tract, but its miniaturized optics, sensor limitations, and wireless transmission constraints result in low-resolution images with reduced visibility of diagnostically important structures. This paper proposes CEM-TUDASR, a computationally efficient unsupervised Transformer-based super-resolution framework for WCE image enhancement without paired low-resolution (LR) and high-resolution (HR) training data. A domain-adaptive degradation network synthesizes realistic WCE-like LR images from HR conventional endoscopy images, reducing the domain gap and enabling effective unpaired learning. The SR generator integrates Deep Attention Blocks (DABs) and a Fusion Attention Block (FAB) to capture long-range contextual dependencies and fine local structures while preserving perceptual and structural fidelity. The model is trained on a curated dataset derived from Kvasir Capsule and evaluated on KID and GIANA for cross-dataset generalization. No-reference quality metrics, including BRISQUE, PIQE, NIQE, and the domain-specific EndoQM, show that CEM-TUDASR consistently outperforms existing unsupervised SR methods. Qualitative results further demonstrate improved restoration of mucosal textures, vascular patterns, and clinically relevant anatomical details. Cross-domain experiments on retinal images additionally demonstrate the adaptability of the framework. With only 2.67 million parameters and 169.94 GFLOPs, CEM-TUDASR achieves high-quality reconstruction while maintaining computational efficiency, making it suitable for resource-constrained clinical and embedded endoscopic applications. --- ### An AI-Powered Culturally Aware Chatbot for Stress Detection and Wellness Support among Pakistani University Students Using NLP and Machine Learning Authors: Muhammad Fahad Bashir, Muhammad Afzal Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11199v1 With the existing digital mental health tools specifically developed for Western settings, Pakistani students are exposed to a uniquely compounded stress situation in their university that includes academic, financial, familial, and relational stressors, which have become a serious concern for academic and psychological development of students in Pakistani universities. This paper introduces a new, AI-driven and culturally sensitive stress detection and wellness support system that is tailored to the context of Pakistani university students. The system is based on a machine learning model called Random Forest which is trained using a validated student stress data set of 1100 responses on 20 features from psychological, physiological, academic, environmental and social aspects, with an accuracy of 89.09% and a macro F1-score of 0.89, in three stress severity levels. The classification outputs are passed on to an open-source large language model through OpenRouter API, where an appropriately crafted system prompt, culturally aware, gives the model a conversation about wellness, in English, Urdu and Roman Urdu. The second most predictive stress factor in this population identified by feature importance analysis was teacher-student relationship, which is a culturally important stress factor highlighting the need for region-aware mental health systems. Future research will involve primary data collection from students at various academic levels of Pakistani Universities with the validated DASS-21 instrument focusing on the students who are moving from FSc to undergraduate studies, which is a time of being psychologically vulnerable which is under-researched. --- ### (Whose defaults?) Is artificial intelligence reorienting archaeological methods? Authors: Lorenzo Cardarelli, Roberto Ragno Published: 2026-09-10 Categories: cs.CY, cs.AI, cs.CL, cs.HC arXiv: https://arxiv.org/abs/2609.11198v1 Generative AI and the practice of "vibe coding" are changing how archaeologists carry out computational research, but their effects on the discipline's range of methods is still understudied. In this paper, we evaluate whether large language models (LLMs) are narrowing the variety of methods archaeologists use. We first analysed approximately 119,000 archaeology abstracts from Scopus, covering publications from 2010 to 2025. Using a locally run LLM, we identified the computational methods reported in each abstract and organised them into 25 broad categories (L2) and 241 finer clusters (L3). A Bayesian Dirichlet-multinomial model of method composition within sub-disciplines found a small but credible shift in method use after 2023. However, this shift was smaller than the variation already present across the full study period. No individual technique showed a significant change, and overall methodological diversity increased rather than declined. We then ran a controlled experiment to see whether LLMs recommend a narrower set of methods than archaeologists have used in practice. Two different open-weight models were asked to suggest methods for 28 archaeological research problems, with prompts providing three levels of methodological guidance: novice, intermediate, and expert. Recommendation diversity was much lower than in the published literature, particularly without methodological guidance. The models also tended to favour methods that were widely used before 2023, and their recommendations more closely resembled the post-2023 literature. Taken together, these results are consistent with LLMs pushing methodological choice towards convergence, although our study cannot establish a causal effect. They raise a broader question: how can archaeology retain methodological diversity as LLMs become more involved in research? --- ### FlexComp: One Model for Every Ratio in Context Compression Authors: Kaiyan Zhao, Zhongtao Miao, Akiko Aizawa, Yoshimasa Tsuruoka Published: 2026-09-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.11192v1 Soft context compression condenses a context into a few memory tokens that a frozen LLM consumes in place of the raw text, but existing compressors fix the compression ratio at training and inference: each deployed ratio requires a separately trained model, and the chosen ratio is applied uniformly to all inputs, whose actual needs vary drastically. We propose FlexComp, a method-agnostic framework that decouples the ratio from both training and deployment: Matryoshka-style training samples the memory budget $K$ per instance, turning one model into an any-ratio compressor, and the budget is then chosen per input by: (1) confidence-based cascade routing or (2) a lightweight learned $K$ predictor. Across ICAE, 500xCompressor, and SAC on MRQA, a single FlexComp model matches separately trained fixed-ratio specialists with minimal degradation. Cascade routing preserves over 98% of the mildest ratio's accuracy at up to 266x average compression; the $K$ predictor, in a single compression-decoding pass, reaches 158-236x within 0.7 F1 of the mildest ratio. At serving-scale batch sizes, the $K$ predictor cuts context KV cache by 50% and improves decoding throughput by 47%. --- ### Agentic Share-of-Search: A Multi-Agent AI System for Competitive Decision-Making in LLM-Mediated E-Commerce Authors: Spandan Ghose Chowdhury Published: 2026-09-10 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2609.11190v1 AI shopping assistants increasingly redirect consumer discovery, creating an urgent need for tools that support seller-side competitive decision-making. We present a multi-agent AI system that automates competitive visibility measurement and root cause diagnosis in LLM-mediated ecommerce. The system introduces Agentic Share-of-Search (ASoS) as the decision target, deploys query agents across leading AI platforms, and uses a ReAct-based diagnostic agent to recommend prioritized merchandising interventions. A 100-trial ablation study, presented as a feasibility evaluation of this prototype, shows the agent recovers the ablated signal in 39% of trials (95% CI: 30.0% - 48.8%, 5.5x over chance), rising to 63.9% among high-correlation ablations. --- ### A Multi-View and Confusion-Guided Ensemble Framework for Robust Synthetic Image Attribution Authors: Zuomin Qu Published: 2026-09-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.11188v1 Synthetic image attribution (SIA) has become increasingly important with the rapid advancement of text-to-image generation models. However, accurately identifying the source model of a generated image remains challenging due to the growing similarity among modern diffusion-based generators and the presence of diverse post-processing operations. In this report, we present a multi-view and confusion-guided ensemble framework for the Synthetic Image Attribution Challenge of the DLMMDD Workshop at ICANN 2026. Our approach integrates multiple complementary architectures, including FFT-ConvNeXt, DINOv2, CLIP, and Xception, to capture diverse attribution cues from frequency, semantic, and forensic perspectives. To improve robustness against unknown degradations and image manipulations, extensive data augmentation strategies are employed during training, simulating realistic post-processing operations such as compression, resizing, grayscale conversion, and blur. Furthermore, we analyze the confusion patterns of the ensemble model and observe severe ambiguity between Stable Diffusion 3 and Stable Diffusion 3.5. To address this issue, we introduce a dedicated binary expert classifier that is selectively activated under low-confidence conditions. We additionally apply class-adaptive confidence calibration to improve the discrimination of challenging classes such as Tencent Hunyuan. The proposed framework achieved 99.53% on the public leaderboard and 99.20% on the private leaderboard. The source code and implementation details are publicly available at https://github.com/ZOMIN28/SIA. --- ### Can LLMs Follow Medical Expert Logic? A Benchmark for Hierarchical Logical Consistency in Risk-of-Bias Assessment Authors: Jiayu Huang, Zichen Tang, Qianhui Ling, Zemin Kuang, Haihong E Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11185v1 Evidence-based medicine demands strict logical consistency, yet current evaluations of large language models (LLMs) prioritize superficial label matching over genuine reasoning. We introduce LogiMed-RoB, a benchmark grounded in Cochrane Risk of Bias (RoB) 2.0 expert logic, comprising 860 randomized controlled trials (RCTs) and 14,820 queries. It evaluates models under the Hierarchical Logical Consistency (HLC) framework across four dimensions: Atomic Consistency, Domain Consistency, Aggregation Consistency, and Evidential Faithfulness. Experiments on 10 state-of-the-art LLMs reveal a catastrophic Error Compounding Effect: despite the top model reaching 98.88% Atomic Consistency, its end-to-end consistency collapses to 45.13%, with several open-weight architectures plummeting to nearly 0%. We further uncover a systematic evidence-reasoning gap: even when models retrieve high-quality evidence, they fail to deduce correct outcomes in 18.63-40.05% of cases, while Blind Guess Rates reach 48.28%. LogiMed-RoB demonstrates that high outcome accuracy can conceal critical reasoning flaws, underscoring the necessity of white-box logical verification for clinical deployment. --- ### Exploring Second-Order Pattern Recognition in Speaker Recognition Authors: Yanze Xu, Wenwu Wang, Mark D. Plumbley Published: 2026-09-10 Categories: eess.AS, cs.AI arXiv: https://arxiv.org/abs/2609.11182v2 In traditional pattern recognition tasks, neural networks are trained to recognise human-defined patterns (e.g. audio categories) in model inputs (e.g. audio). Meanwhile, some Explainable AI (XAI) methods explain latent patterns characterising the network's recognition of inputs as human-defined patterns; this work calls these latent patterns second-order patterns and proposes to discover them. Accordingly, we apply a hierarchical clustering algorithm to analyse whether our speaker recognition network's representations learned from known utterances naturally form hierarchical clusters. Each resulting cluster is a second-order pattern that characterises a context in our network's recognition of the known utterances as speaker identities. All discovered second-order patterns are then interpreted using the Hierarchical Cluster-Class Matching (HCCM) method. Moreover, we propose a new task, second-order pattern recognition, to identify which of the discovered second-order patterns characterising the recognition of known utterances also apply to unseen utterances, thereby characterising the recognition of unseen utterances. Accordingly, we design the Hierarchical Cluster Navigation and Assignment (HCNA) method. HCNA recognises a second-order pattern as applying to an unseen utterance when the utterance's network representation lies within the extrapolation space of the cluster regarded as that second-order pattern. Experimental results demonstrate that the extrapolation space introduced in HCNA substantially improves task performance. --- ### SemVerBench: Benchmarking LLM Comprehension of Version-Constraint Resolution Semantics Authors: Qibai Chen, Zeming Liu Published: 2026-09-10 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.11180v2 Large language model (LLM) coding agents constantly decide whether a version satisfies a constraint such as ^1.2.3 or >=2.0,<3, yet their grasp of version-constraint semantics has never been measured directly. We introduce SemVerBench, the first benchmark of LLM version-constraint resolution semantics across three ecosystems (npm, PEP 440, Cargo): 240 machine-checkable items with unique answers, built author-neutrally from four balanced sources (each ecosystem's official test suite plus three frontier LLM proposers) and labeled by a non-circular two-implementation oracle. Evaluating six frontier models, we find systematic, predictable per-mechanism blind spots: a partial-comparator carry rule (>1.2 means >=1.3.0) traps every model on Cargo (near 60%), and although standard PEP 440 prefix matching is universal, on zero-pad/post-release corner cases GPT-5.1 collapses (0/26) while Claude stays at 97-100% (verified on a 67-item oracle-validated set). Opus significantly outperforms all other models, and Sonnet outperforms the OpenAI models (McNemar). The failures look more like an activation/application gap than a knowledge gap: injecting the rule or a light correct hint recovers most errors, whereas interval decomposition does not, and models are at ceiling on the basic forms of the same rules. An author-stratified analysis finds no statistically significant self-favoritism. Because the task is verifiable and a free, 100%-correct resolver exists, tool delegation reaches ~100%: coding agents should delegate version resolution to a resolver rather than reason about versions in-head. --- ### Debate-to-Skill: Capability-Bound Process Supervision for Industrial Query-to-Agent Annotation Authors: Shiyu Zhang, Leisheng Cheng, Huifu Li Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11176v1 Industrial query-to-agent matching fails when topical relevance is mistaken for executable capability, especially on long-tail and boundary-sensitive requests. We formulate annotation as \emph{capability-bound process supervision} and instantiate it with Debate-to-Skill, which uses reusable decision principles, structured deliberation, verifier-based verdict extraction, and disagreement-driven refinement. On an industrial Query2Agent benchmark, we compare Debate-to-Skill with direct-label supervision, reasoning-SFT, and structural ablations. The results test whether gains come from supervising the capability-critical decision process itself, especially on grey-zone cases where semantic relatedness and executable capability diverge. --- ### Hierarchical Clustering Can Jointly Satisfy Richness, Consistency, and Scale Invariance Authors: Daichi Kuroda, Maximilien Dreveton, Matthias Grossglauser, Patrick Thiran Published: 2026-09-10 Categories: cs.LG, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2609.11173v1 Despite its ubiquity, clustering lacks a universally accepted definition of what is a cluster. Kleinberg's Impossibility Theorem formalizes this difficulty by showing that no flat clustering method can simultaneously satisfy three natural axioms: scale invariance, richness, and consistency. In this paper, we ask whether this impossibility persists when the output is a hierarchy rather than a single partition. We show that, in contrast to the flat clustering setting, the hierarchical analog of these axioms are jointly satisfiable. In fact, there exist uncountably many hierarchical clustering methods satisfying these axioms, which we call admissible. We explicitly construct several admissible methods, including methods based on well-separated clusters and a non-binary version of single linkage. For certain pairs of admissible methods, the hierarchy produced by one always refines that produced by the other. This refinement relation defines a partial order on the class of admissible methods. This partially ordered set has no greatest element and contains uncountably many pairwise incompatible maximal elements, revealing substantial diversity among admissible methods. Nevertheless, this diversity is constrained: every admissible method contains a hierarchy of sufficiently well-separated clusters, and every finite collection of admissible methods shares such a nontrivial common backbone. --- ### Beyond Visual Quality: Evaluating Physical Consistency under Ego-Motion with EgoGenEval Authors: Yilin Long, Chenming Zhu, Zitang Gou, Jingli Lin, Tai Wang Published: 2026-09-10 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.11172v1 Recent visual generators produce high-fidelity images yet often violate physical consistency under ego-motion, limiting their use for spatial reasoning and embodied planning. Existing benchmarks largely focus on isolated images or single-step quality, leaving this challenge underexplored. We introduce EgoGenEval, a geometry-grounded, pose-free benchmark designed to evaluate the physical consistency of visual generators under ego-motion, and organize our study into two parts. (1) EgoGenEval contains 1,400 cases and 2,360 target views spanning single-step and multi-step ego-motion. It separately measures Camera Motion Grounding (CMG) and Scene State Preservation (SSP), with both metrics validated against blinded human judgments. Evaluating 16 pose-free generators together with two pose-conditioned references reveals that current models struggle to execute camera motion while maintaining scene state, and that no system performs well on both axes at once. (2) To examine whether benchmark-derived data can improve these capabilities, we build EgoGen-Train from the same geometry-grounded pipeline and run controlled SFT studies. These show that pairwise supervision does not reliably improve camera-motion grounding and scene-state preservation together: even at the full training pool and the longest budget, scene preservation gains a fraction of what camera motion does. This points to the pairwise teacher-forced objective itself as the binding constraint, motivating a trajectory-centric paradigm that couples self-conditioned rollouts with explicit pose and visibility supervision. --- ### From Explanations to Interventions: Execution-Guided Counterfactual Synthesis in Temporal Graphs Authors: Minwoo Yu, Young-guk Ha Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11170v2 Can a trace explaining model execution also compute the changes needed for a specified alternative prediction? We propose trace-guided intervention search, which uses executable reasoning traces as an intermediate representation for intervention synthesis. A Specified-Foil Counterfactual edits past events so that a frozen temporal predictor selects a designated foil. Our method constructs facts and replacement values from completed original and foil executions and recovered unmet conditions. Proposal generation constructs edits and selects candidates within a fixed cap; exact replay verifies foil top-1 outcomes among retained edits and compositions. Implemented in LiFTER for continuous-time dynamic graphs (CTDGs) and TLogic for temporal knowledge graphs (TKGs), the method improves success over coordinate-based proposal generation by 13.7-34.7 percentage points on four CTDG datasets and 60.0-83.3 points on two TKG datasets under matched downstream search and a proposal cap of 32. Separate shared-candidate comparisons retain 85.7-93.6% of black-box greedy's CTDG success rate with 75.0-80.0% fewer predictor evaluations. A Pulse case study confirms simulator-level survival for five of six interventions. Executable traces thus provide both explanatory evidence and a reusable computational representation for constructing and testing specified alternatives. --- ### Semi-Tensor Product-Based Multi-Term Randomized T-SVD and Its Visual Applications Authors: Xingchen Xiao, Feng Zhang, Wenjin Qin, Jianjun Wang Published: 2026-09-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.11168v1 Tensor singular value decomposition (T-SVD), which is built upon the tensor-tensor product (t-product), has emerged as a powerful tool for processing high-dimensional visual data such as color images and videos. However, the standard t-product imposes strict dimensional compatibility constraints. Although extensions based on the semi-tensor product (STP) relax this restriction, their single-term formulations still suffer from limited approximation accuracy. Moreover, these deterministic methods incur high computational costs when processing large-scale tensor data. To address these issues, this paper introduces a novel semi-tensor product for third-order tensors under the t-product framework induced by arbitrary invertible linear transforms. The resulting tensor semi-tensor product breaks the rigid dimension matching requirement of the standard t-product, while retaining the closed-form property of T-SVD. Based on this construction, we develop a multi-term semi-tensor product singular value decomposition (MSTP-SVD), which integrates multiple orthogonal decomposition terms to significantly improve low-rank approximation accuracy compared with single-term schemes. To reduce the computational cost of multi-term modeling, we incorporate randomized projection and power iteration techniques into the MSTP-SVD framework, yielding an accelerated multi-term randomized semi-tensor product SVD (MRSTP-SVD) algorithm that achieves a balance between reconstruction accuracy and computational efficiency. Experiments on image and video compression and completion tasks demonstrate the effectiveness of the proposed method. --- ### When does a spectral prior help graph learning? Connectivity-loss estimation under road-network disruptions Authors: Van-Truong Le Published: 2026-09-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.11166v1 Rapid evaluation of many simultaneous road-link disruptions requires a practical compromise between exact spectral recomputation and local approximation. We estimate relative algebraic-connectivity loss after multi-edge deletion using graph neural networks (GNNs) that learn a bounded correction to a first-order Fiedler sensitivity. The study considers independent, spatially clustered, and edge-betweenness-targeted failures, with graph-disjoint synthetic splits and zero-shot transfer to 13 OpenStreetMap (OSM) areas in six countries. GCN, GraphSAGE, and edge-aware MPNN backbones are compared with analytical baselines. In expanded OSM tests, residual GCN improves spatial-failure MAE by 0.0391 (95% hierarchical interval 0.0151-0.0662), while residual GraphSAGE improves targeted-failure MAE by 0.0257 (0.0095-0.0446). Second-order perturbation improves first-order MAE by only 0.0028-0.0053. Correction slopes decrease under targeted transfer, indicating residual shrinkage around systematic prior error. Leave-one-country-out OSM-to-OSM transfer is mixed: residual GCN improves targeted-failure MAE by 0.0622 (0.0169-0.1153) but worsens the spatial point estimate. Sparse scaling extends to 20,000 nodes and separates one-time spectral setup from amortized screening cost. These results characterize the spectral residual as a useful but domain-sensitive inductive bias for structural connectivity screening. Code, cached networks, and reproducibility artifacts are archived at doi:10.5281/zenodo.22307723. --- ### LILA: Calibration-Free Structured Pruning of Large Language Models via Latent Spectral Geometry Authors: Sankar Behera, Dhruv Singh, Anshika Agnihotri, Raj Kumar Choudhary, Satyadev Ahlawat, Yamuna Prasad Published: 2026-09-10 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.11163v1 Structured pruning of large language models (LLMs) offers hardware-efficient compression, yet existing methods require calibration data, gradient computation, or large auxiliary policy networks at pruning time. LILA (\emph{Latent-Informed Layer Analysis}) scores neuron importance via the Kolmogorov--Smirnov (KS) distance between empirical singular value distributions of the full and neuron-ablated feed-forward network (FFN) weight matrix, providing a closed-form spectral rule requiring no training, calibration data, or auxiliary network. Without any fine-tuning, LILA surpasses PruneNet (45M-parameter RL policy) by 1.57~pp in zero-shot accuracy on LLaMA-2-7B at 25\% sparsity, and outperforms WikiText-2-calibrated SliceGPT by up to 6.0~pp across all sparsity levels, while preserving the original architecture. After one epoch of LoRA recovery fine-tuning, LILA achieves highly competitive performance, matching the heavily calibrated SliceGPT baseline to within a 0.48~pp margin across LLaMA-2-7B and Phi-2, despite using zero calibration data. A Neural Tangent Kernel analysis confirms a 22$\times$ reduction in functional distortion versus random pruning, providing theoretical grounding for the spectral importance criterion. Finally, extending LILA to dynamically allocate sparsity budgets via KS-scores yields state-of-the-art generative preservation at moderate compression, while uncovering fundamental single-layer architectural bottlenecks at higher compression regimes. --- ### UniH$^3$: Unifying Hierarchical Homogeneity and Heterogeneity for All-in-One Medical Image Restoration Authors: Zhiwen Yang, Jiayin Li, Chengyu Liu, Hui Zhang, Bingzheng Wei, Yan Xu Published: 2026-09-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.11156v1 All-in-One medical image restoration (MedIR) aims to address diverse tasks across modalities and degradation types using a single universal model. Existing methods typically prioritize modeling inter-task heterogeneity (e.g., distinct data distributions and degradation types). However, they largely neglect the inherent homogeneity present in medical images, such as widely shared anatomical structures within and across modalities, which can be leveraged to ease model training and improve generalization. To this end, we propose UniH3, a novel framework that Unifies Hierarchical Homogeneity and Heterogeneity for all-in-one medical image restoration. Specifically, to comprehensively exploit homogeneity, we introduce a Hierarchical Homogeneity Memory (H2M) module that progressively distills intra- and inter-task homogeneity priors from high-quality images during training, and adaptively retrieves the most relevant priors tailored to the input for guided restoration. These retrieved priors are then injected into the restoration pipeline via an efficient Homogeneity-Guided Attention (HGA) mechanism. Furthermore, to comprehensively address heterogeneity, we design a Hierarchical Heterogeneity Balancer (H2B) that mitigates both inter- and intra-task conflicts during optimization, facilitating balanced and effective multi-task learning. Extensive experiments on two large-scale benchmarks, MedIR-2D-500K and MedIR-3D-3K, demonstrate that UniH3 achieves state-of-the-art performance on both all-in-one and single-task medical image restoration. We hope this work establishes a strong benchmark and advances the development of general-purpose medical image restoration models. Code is available at https://github.com/Yaziwel/UniH3. --- ### DRG-MAPPO: Hierarchical Dynamic Role-Graph Multi-Agent Reinforcement Learning for Cooperative Air Combat Authors: Junlin Liu, Chengwei Li, Yang Gao, Hui Chang, Xinchen Zhang, Zhijun Zhao, Hao Zhao Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11155v1 Multi-Agent Reinforcement Learning (MARL) has emerged as a pivotal paradigm for complex decision-making in autonomous systems and air combat. While MARL has demonstrated significant potential in air combat, achieving sophisticated tactical coordination remains a non-trivial challenge. This difficulty is largely attributed to two primary limitations: (1) the absence of structured relational modeling hinders agents from capturing complex, time-varying interactions among battlefield entities; and (2) conventional flat architectures often lack the capability to explicitly model tactical roles, leading to ambiguous task allocation in highly dynamic environments. To address these challenges, we propose Hierarchical Dynamic Role-Graph Multi-Agent Proximal Policy Optimization (DRG-MAPPO), a novel MARL framework that integrates graph-based relational modeling with dynamic role assignment. Specifically, DRG-MAPPO constructs a graph-based representation of battlefield interactions and leverages graph attention mechanisms to extract critical relational features among allies, enemies, and threats. Subsequently, a high-level policy employs a dynamic role assignment mechanism to determine tactical responsibilities (e.g., ``leader'' and ``supporter''). Conditioned on these roles and encoded graph-relational features, a low-level policy executes discrete maneuver actions, facilitating the joint optimization of tactical strategy and collaborative execution. Furthermore, a target-priority auxiliary task is designed to foster the emergence of behaviors such as focus-fire. Experimental results demonstrate that DRG-MAPPO achieves a state-of-the-art win rate of 87%, suggesting that our framework effectively balances relational modeling, interpretability, and optimization stability for cooperative air combat. --- ### terms.txt: A Consent and Compensation Protocol for Agentic Web Access Authors: Rajarshi Chowdhury Published: 2026-09-10 Categories: cs.NI, cs.AI, cs.CR, cs.CY arXiv: https://arxiv.org/abs/2609.11152v1 The open web ran on an unwritten bargain: sites admitted crawlers, and search engines sent visitors back. Public measurements show that bargain breaking under AI crawlers and agents. Automated clients now make up most requests, training dominates Cloudflare-classified crawling, and the largest AI platforms fetch thousands of pages for each visitor they return. The web's common control, robots.txt, cannot express identity, purpose, terms, or price, can be circumvented, and newer alternatives are largely proprietary CDN features. We specify terms.txt, a robots.txt-style file for per-path, per-purpose machine-access terms, plus an origin-enforced exchange using Web Bot Auth signatures, signed intent, delegation tokens, HTTP 402 negotiation, and signed receipts. We define what the exchange can enforce, audit, and leave to contract. A dependency-free implementation adds 0.20 to 0.65 ms per request on one vCPU. --- ### Break Step: Recursive Training Resonates with Replayed Sampling Noise Authors: Yangze Liu, Zhongyi Han Published: 2026-09-10 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.11149v3 How fast does a language model degrade when trained on its own outputs? Theory traces it to gradually accumulating errors, while experiments report repeated phrases within ten generations. Under a fixed sampling seed in vLLM, the fast loss of lexical diversity comes from the sampler. When vLLM serves a batch from one seeded sampling configuration, every request receives the same random draws, and a fixed seed replays them every generation. Fine-tuning raises the tokens that won, and the replayed draws let them win by more. Sharing across requests and replay across generations matter only together. Remove either one, by changing the shared seed every generation or by giving each request its own seed that repeats every generation, and the unique-4-gram fraction of two StableLM checkpoints stays near its starting value of about 0.98 through generation 3. Keep both, and the replayed shared seed takes seven checkpoints from five families to between 0.045 and 0.38 by then. Three generations of replay write the favoured phrases into the weights: decoded with one seed per request, the generation-3 weights of the replayed StableLM-2-1.6B chain recover most of their diversity, yet the phrase that filled every sample under the shared seed still opens 46% of them. Without replay, five checkpoints drift slowly, consistent with the gradual accumulation that theory describes, and three turn incoherent though their diversity scores stay high. One peer-reviewed model-collapse pipeline that fine-tunes Gemma-2-27B samples identical prompts under one seeded configuration, and three quarters of the rows it released for one iteration repeat nearly as often as one such batch copies them. A seed per request restores the fresh sample that stability analyses assume. --- ### Autonomous Chemical Mechanistic Discovery through Agentic Reasoning and Validation Authors: Dong Li, Sixuan Mi, Zihao Ye, Huan Xiong, Tao XU, Tong Zhu, Aijia Zhang, Junqi Gao, Kaiyan Zhang, Shijie Wang, Bowen Zhou, Yuqiang Li et al. Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11147v1 Unraveling reaction mechanisms is central to modern chemistry, yet automating these investigations remains challenging because computational workflows still rely heavily on expert intervention. Here we introduce ARCHE, an autonomous agentic system that integrates a general-purpose reasoning model, a domain-specialized computational chemistry model, and a structured tool registry to transform mechanistic inquiry into a scalable, self-validating process. ARCHE interprets scientific questions, generates and prioritizes mechanistic hypotheses, orchestrates computational workflows, and iteratively refines conclusions based on computed evidence within a closed loop. We validate its capabilities across three increasingly demanding scenarios: reconstructing stereocontrolling transition states and validating the corresponding reaction mechanism in a previously reported asymmetric catalytic reaction; proposing and validating a plausible radical pathway through iterative hypothesis refinement for a recently discovered but unpublished $α$-iodoboronate C-I cleavage reaction; and identifying a chemically interpretable descriptor that governs selectivity in nickel-catalysed migratory cross-coupling reactions. By coupling agentic reasoning with rigorous computational validation, ARCHE advances autonomous mechanistic discovery and establishes a foundation for broader machine-assisted chemical research. The code for ARCHE is publicly available at https://github.com/JetAstra/Arche-Harness. --- ### A Dominant Supplier Slows Recursive Drift More Than It Steers It Authors: Yangze Liu, Zhongyi Han Published: 2026-09-10 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.11146v3 More and more of the text future language models learn from is written by a few of today's models. If one supplier writes most of a shared corpus, does it pull the models trained on it toward its own writing, or change how fast they drift? We retrain eight open models from their base weights on a shared pool of each other's text for five generations, varying the part written by one model, Phi-2, from an equal share to 90%. The models drift together toward a style with fewer function words, and none starts repeating itself. No share of Phi-2 brings the other models closer to its text than the equal share does. We split each ecosystem's separation from the equal-share one into a delay along its route and a departure from that route, both counted beyond the difference between two equal-share runs. With Phi-2 at 90%, delay outweighs departure 72 to 28 and 64 to 36 in two runs, and the ecosystem falls 2.7 and 2.5 generations behind. With Phi-2 at half the pool the two parts are about equal. When SmolLM2 or Qwen3-1.7B writes half instead, the ecosystem slows less or not at all. The departure leans toward Phi-2 more as its share grows, but more than toward every other model only at 90%. Human text filling a quarter or half of the pool slows the models along the same route. --- ### Human Agreement and Return Association Are Not Interchangeable Criteria Authors: AS Aravinthakshan, Laven Srivastava, Harsh Nandwani Published: 2026-09-10 Categories: cs.AI, cs.CL, cs.SI arXiv: https://arxiv.org/abs/2609.11144v2 Financial NLP has a standard workflow: validate a sentiment tool against human labels, then trust it to extract market signal. This assumes the two evaluations measure the same thing. We test that assumption in a setting where both can be measured at once: a corpus of securities class actions (2002-2025) linking 70,500 X messages to abnormal stock returns, with a single-annotator human labelled gold sample. Running five instruments (VADER, Loughran-McDonald, FinBERT, Twitter-RoBERTa, and an LLM annotator) through one identical pipeline, we find that the relationship between construct and predictive validity depends on the sampling convention and score representation. Under conventional method-specific sampling, human agreement aligns more closely with graded same-day associations than with one-day leads. On a fixed-n panel, however, agreement has similar graded rank correlations at both horizons, while the coarse ordering remains weak. Benchmark agreement therefore establishes semantic validity but does not by itself determine predictive rankings. In a conversation that is 17.6% spam, message volume predicts neither market damage nor settlement size. --- ### Can LLMs Normalize Databases? A Benchmark and Multi-Agent Framework for Schema Normalization Authors: Dong-Jae Koh, Huisu Kim, SeongHwan Yoon, Lasse M. Jantsch, Chun-Hee Lee, Seonghyeon Lee, Young-Kyoon Suh Published: 2026-09-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.11141v1 Large Language Models (LLMs) are increasingly used to generate structured outputs, but their reliability remains unclear when those outputs must satisfy database-level constraints. We study this issue through database normalization, involving reasoning about functional dependencies, lossless join decompositions, and inter-table constraints. We introduce a Database Normalization Benchmark (DNBENCH), comprising 3,275 samples for evaluating LLM-driven database normalization from 1NF to BCNF. DNBENCH uses a three-axis protocol to measure semantic equivalence, structural accuracy, and logical validity. Across Single, Complex, and Real World levels, DNBENCH uncovers recurring failures in dependency inference, schema decomposition, and inter-table constraint reconstruction. We further propose Multi-Agent Reasoning for Schemas (MARS), which separates evidence extraction, violation diagnosis, and decomposition planning from schema generation and verification. MARS improves the DNB-SCORE by 82.0% over the single-prompt baseline. All artifacts will be released upon acceptance. --- ### Bidirectional Multimodal Fusion of Sky Images and Time-Series for Solar Forecasting with Large Language Models Authors: Ken Chen, Maneesha Perera, Wei Wang, Sachith Seneviratne, Hansani Weeratunge, Saman Halgamuge Published: 2026-09-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.11135v1 Short-term photovoltaic (PV) power and global horizontal irradiance (GHI) forecasts are essential for effective dispatch, reserve scheduling, and grid operations. At these forecasting horizons, errors are predominantly driven by cloud induced ramps: relying solely on historical numerical data may struggle to anticipate an incoming cloud, making ground-based sky images a crucial complementary physical signal. Furthermore, forecast performance is highly sensitive to location and local observing conditions, creating a strong need for site-specific data that are often scarce. Recently, large language models (LLMs) have demonstrated competitive performance and high data efficiency in time-series forecasting. Despite their success, existing LLM-based forecasting methods remain predominantly unimodal, relying primarily on historical numerical time-series data. Effectively incorporating sky imagery into an LLM-based forecasting framework remains under-explored and an open challenge. In this paper, we propose SolCloudLLM, an LLM-based multimodal forecasting framework. SolCloudLLM aligns sky-image patches with time-series patches and fuses their corresponding representations through bidirectional multimodal fusion, yielding a unified representation that is subsequently mapped into the embedding space of an LLM. Extensive experiments on the SIRTA and SKIPP'D datasets demonstrate that SolCloudLLM consistently outperforms the best baseline methods in MSE across all forecasting horizons, achieving a maximum relative MSE reduction of 25.4%. Stratified analysis further indicates that the benefits of multimodal fusion are concentrated primarily under cloudy conditions. Notably, SolCloudLLM achieves the best performance in nearly all few-shot settings, whereas other deep learning baselines experience substantial performance degradation and are frequently outperformed by the non-learning physical method. --- ### LAION-Mobile: Evaluating Deepfake Detectors On One Million Smartphone Photos Authors: Achim von Stryk, Janis Keuper Published: 2026-09-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.11134v1 Most Deepfake detectors report near-perfect AUC scores on their reference benchmarks. However, a recent ICML position paper argues that these evaluations collectively neglect the impact of modern smartphone photography: the widely used on-device neural image-signal processing pipelines (like multi-sensor fusion or noise and motion-blur suppression) increasingly shift the imaging paradigm from simple lens projections towards computational photography. Hence, devices actually generate, rather than record photos. This increases the risk that deepfake detectors may flag ordinary phone photos as fake. Due to the lack of large-scale datasets containing images from modern smartphones, this hypothesis has so far only been tested in small proof-of-concept studies. The aim of this paper is to close this gap. We introduce LAION-Mobile, an open dataset containing about 1 million smartphone images with EXIF metadata distilled from re-LAION-5B. Evaluating twelve state-of-the-art deepfake detectors with their original paper checkpoints on a 9,115-image evaluation sample of this pool (DIRE on 738), we report three key findings: (i) On modern AI content no detector exceeds AUC 0.624, and five of twelve fall below chance. (ii) Real-photo false-alarm rates are an artefact of threshold calibration: thresholds fitted on legacy GAN data make several detectors look deployable (less than 11 percent FPR), yet the same detectors flag 17-91 percent of real photos once the identical criterion is refit on modern content. (iii) Consequently, no detector both beats chance on modern AI content and keeps a deployable real-photo false-alarm rate. Mirroring the device mix of web collections, the corpus probes the first neural-ISP generation (2018-2020); current flagships are essentially absent, leaving the modern-ISP regime as the open gap. --- ### Phase-Decoupled, Model-Calibrated Power Control for Disaggregated LLM Serving Authors: Jae Gon Kim, Donghoon Yoo, Hanyul Ryu, Sungho Ha, Juyeon Lee, Soojung Ryu Published: 2026-09-10 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2609.11133v1 Datacenter GPU power is the binding constraint on LLM serving capacity, and production serving has shifted to prefill/decode (PD) disaggregation. Deploying NVIDIA's Max-Q inference profile on a disaggregated B200 system, we found its realized gain modest (+8.6% tokens/J), model-dependent, and carrying a mean end-to-end latency cost (+5.2%) that throughput-only evaluation does not surface; the profile also applies one setting to prefill and decode GPUs that operate in opposite hardware regimes. We hypothesize that the optimal power setting is a property of the deployed (model, quantization, engine, hardware) combination rather than of the GPU class, that each lane warrants its own profile, and that converting SLO headroom into energy safely requires latency-gated calibration under a runtime SLO guard rather than a fixed recipe. We present a phase-decoupled, model-calibrated controller: the prefill lane runs under an SM-clock window whose floor is a latency guarantee by construction, and the decode lane under a power cap placed by automatic calibration just above a measured throughput/latency cliff. Because a disaggregated decode lane draws flat, memory-bound power, the cap binds continuously, the reactive-overshoot weakness that led POLCA to reject capping is absent, and the GPU's own power manager retains throughput under the cap. On an 8x B200 node serving Qwen3-Coder-480B (FP8) under agentic load, our balanced mode delivers +20.4% tokens/J at +3.5% mean e2e versus +8.6% at +5.2% for Max-Q, a Pareto improvement on both axes. On Qwen3-235B-A22B (NVFP4) every operating mode meets the ITL-p99 SLO in every repetition; both vendor profiles miss it. A decode-actuator A/B shows the calibrated cap beats static clock locks, and a three-day sustained run saves 32.3% of a lane pair's electricity. Both models are MoE; a dense model recovers roughly 5x less, so we scope our claims to MoE serving. --- ### How Wrong Can a Good Predictor Be? Diverging Updates with Vanishing Predictive KL Authors: Qifu Wen, Shuaijun Liu, Zihan Zhou, Xi Zeng, Ningxin Su Published: 2026-09-10 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.11132v1 Accurate posterior prediction need not require accurate approximation of Bayesian updates. We prove that an unbounded gap between the update maps can coexist with vanishing predictive KL for every fixed finite $K\ge2$ in a stationary symmetric Gaussian HMM. Exact Bayesian mixing and an explicit deterministic radial filter act on the same $K-1$ belief coordinates. As $q\to0^+$, their separation in centered logits in the worst case grows at least linearly in the natural confidence scale $L_K(q)$, while their categorical $D_{\mathrm{KL}}(\mathrm{exact}\|\mathrm{radial})$ vanishes at the same explicit witness. Along stationary HMM trajectories, the expected terminal KL between filtered posteriors also converges to zero at $H(q)=\lceil-\log(q)/c\rceil+1$. Typical blocks without switches drive both filters into a common confidence cone, where softmax curvature suppresses their disagreement; a single Gaussian maximal event controls adaptive noise. A sweep with equally spaced Gaussians over $K\in\{2,4,8\}$ illustrates the opposing trends, and binary controls at long horizons compare saturating and nonsaturating recurrences. The result isolates two missing links between internal update gaps and predictive cost: the contribution of separating states to expected loss and decoder sensitivity. Thus even an unbounded internal update gap does not by itself certify predictive failure. The construction is fixed in $K$ and does not provide a universal criterion for when compression is harmless or characterize when internal gaps must incur task loss. --- ### Rubric-Aligned Disentangled Evaluation of Human Simultaneous Interpreting Authors: Ziyu Zhang, Satoshi Nakamura Published: 2026-09-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.11131v1 Human simultaneous interpreting (SI) is commonly assessed with analytic rubrics separating meaning transfer, delivery quality, and temporal synchrony, yet no automatic metric is designed for rubric-aligned segment-level SI evaluation. We construct a professionally annotated corpus of 1,101 SI segments with scores for meaning transfer (LQ), delivery quality (EXP), and perceived latency (LAT). We show that structured LLM prompting and scalar supervision collapse rubric dimensions, yielding near-zero correlation with human ratings and strong cross-dimension coupling. To isolate supervision structure under identical backbone capacity, we introduce dual regression heads on a LoRA-adapted COMET-KIWI encoder. On a held-out talk-level test set, the model achieves Pearson correlations of 0.388 (LQ) and 0.301 (EXP), improving over frozen COMET-KIWI. Given low absolute rater agreement, we interpret results relative to human consistency and target stable ranking signals for formative assessment. --- ### ReconPlusGen: Injecting Reconstruction Prior into Multi-view 3D Generation through Noise Inversion and Modulation Authors: Jiarui Liu, Heng Li, Weiyu Li, Keng Deng, Junyuan Deng, Zheng Zhongxing, Junyu Huang, Jiahao Chang, Xiaoguang Han, Ping Tan Published: 2026-09-10 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.11129v1 Qualitative results and an illustration of our core idea. Top left: reconstruction results on benchmark images. Top right: reconstruction results on real-world images. Bottom: illustration of reconstruction-guided noise initialization and modulation. Given multiple input images, we predict a point cloud in canonical space, deterministically inject the predicted geometry into the diffusion process through noise inversion, and modulate the resulting noise to preserve the generative flexibility required to complete unobserved regions and refine visible geometry. --- ### From Repetition to Recognition: Inductive Discovery of Disinformation Narratives Authors: Max Upravitelev, Veronika Solopova, Jing Yang, Charlott Jakob, Alexandra Tsiakalou, Neda Foroutan, Vera Schmitt Published: 2026-09-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.11128v1 In disinformation datasets, narratives are often understood as recurring interpretive patterns that group texts under narrative labels. Recent work formalized narrative mining as inductively inferring narrative labels from corpora, but its evaluation stays tied to predefined taxonomies, a closed-world setting that cannot capture narratives absent from the reference labels. We introduce a three-tier evaluation framework for unsupervised narrative label generation: recovery (against a corpus's own taxonomy), mining (against external label sets), and discovery (without predefined labels). Applying it, we compare clustering-based and graph-community-based pipelines across seven disinformation datasets, with human validation of discovery on two. The two families are complementary under automated metrics, but in a corpus with two prominent topics, clustering can reduce one topic to 2% of generated labels while graph-based pipelines stay balanced. Discovery validation also reveals many singletons (narrative labels derived from single claims, 30-62% of graph outputs), which clustering cannot produce. Annotators confirm many as recognizable disinformation narratives, suggesting that in open-world discovery the repetition assumed by narrative mining may be recognized outside the corpus, not within it. We release human-validated narrative candidate labels for the Climate Obstruction and PolyNarrative datasets to support taxonomy development and dataset extension. --- ### KuaiRP Series Role-playing Models Technical Report Authors: Yipeng Wang, Ziwei Zhang, Jiahui Zhang, Qi Gan, Kai Sheng Published: 2026-09-10 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.11127v1 This paper introduces the complete technical solution for the KuaiRP series of role-playing models. We aim to achieve four core objectives for a dedicated role-playing model: simplified prompt engineering, highly stable output quality, built-in domain world knowledge, and high-efficiency deployment with a small parameter size. However, effectively injecting deep domain knowledge often leads to a severe catastrophic forgetting of the model's general agent capabilities. To overcome this trade-off, we propose a multi-stage training pipeline. First, we design a standardized character template and construct an SFT data pipeline based on user behavior simulation and reverse profile filtering. Next, we utilize a rule-based composite reward function during the Reinforcement Learning (RL) phase to eliminate common degradation phenomena like length expansion and repetitive generation. Finally, to recover the general capabilities compromised during SFT and RL, we propose a novel self-distillation paradigm using Two-stage On-Policy Distillation (OPD) equipped with Cumulative-Divergence Decay (CDD). By using the domain-adapted model as the teacher and the original base model as the student, we effectively balance deep domain knowledge injection with the preservation of general agent capabilities. Experimental results demonstrate that the KuaiRP models not only match the current state-of-the-art proprietary models in role-playing fidelity within our target domains, but also successfully recover general agent capabilities, maintaining extremely low deployment costs. --- ### Beyond Benchmarks: Using VLMs to Reveal Systematic Classification Failures Under Real World Conditions Authors: Dieuwertje Alblas, Alma M. Liezenga, Jan Erik van Woerden, Fedor Taggenbrock, Dalia Aljawaheri, Klamer Schutte Published: 2026-09-10 Categories: cs.CV, cs.AI, cs.ET arXiv: https://arxiv.org/abs/2609.11126v1 Verification and validation (V&V) of classification models is crucial to enable a wide range of sensor processing applications. Currently, the V&V process relies on time-consuming manual inspection of erroneous samples to find meaningful patterns. This work explores the use of Vision Language Models (VLMs) to speed up this laborious process. VLMs are trained to embed images into a semantically meaningful vector representation, from which human-interpretable systematic errors can be distilled. Deploying such VLM-based methods in a defence context introduces two major challenges: (1) the defence domain is underrepresented in the training data of VLMs, and (2) surroundings and context are less diverse than for other domains. This study provides an initial assessment of the suitability of VLM-based methods for V&V of defence applications. We propose a VLM-based error slice detection (ESD) method that independently groups and labels systematic errors made by a classification model. We demonstrate that this method is able to identify operationally-relevant artificially added perturbations in a non-military dataset. In a military context, our method clusters and describes images based on their surroundings, but also exhibits overlap between cluster descriptions. We further investigate the difference in embedding variation between our military and non-military dataset, which remains a topic of interest. Although the results do not yet warrant fully automated V&V through VLM-based ESD, they show that VLMs could be used to accelerate V&V processes in the future. --- ### HERALD: High-Fidelity Exemplar Retrieval with Adaptive Landmark Distillation for Heterophily-Aware Graph Condensation Authors: Sujan Chakraborty, Priyanka Saha, Saptarshi Bej Published: 2026-09-10 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.11123v1 Graph condensation aims to produce a small surrogate graph that preserves the downstream node-classification performance of a much larger original graph. Existing methods rely on Weisfeiler-Lehman neighbourhood aggregation or gradient-based distribution matching, both of which assume that adjacent nodes share the same label, an assumption that breaks down under heterophily. We propose HERALD (High-fidelity Exemplar Retrieval with Adaptive Landmark Distillation), a gradient-free graph condensation framework that adapts the node scoring and feature selection in the condensation pipeline to the graph's measured heterophily. HERALD selects features via a joint Fisher-discriminability and activation-density criterion that down-weights aggregated representations on heterophilic graphs, and scores nodes by a weighted combination of prototype representativeness, decision-boundary proximity, and Local Intrinsic Dimensionality (LID), where the weights are driven by a smooth sigmoid function of the heterophily ratio. Nodes are then assembled into a condensed subgraph through score-ordered BFS expansion, Personalised PageRank pruning, and class rebalancing, all at an identical storage budget to BONSAI, enabling direct comparison. Experiments on eight benchmark datasets spanning homophilic and heterophilic settings show that HERALD matches or outperforms state-of-the-art condensers on heterophilic graphs and remains competitive on homophilic ones across four GNN architectures. --- ### Overview of the NLPCC 2026 Shared Task 11: Agent-Based Experiment Reproduction from Scientific Papers Authors: Hanhua Hong, Yizhi Li, Luu Gia Huy, Jian Yang, Ming Zhou, Chenghua Lin Published: 2026-09-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.11117v1 Reproducibility is essential to scientific progress, yet the growing volume and complexity of scientific publications make exhaustive manual verification increasingly impractical. Although recent advances in large language model (LLM) agents enable automated experiment reproduction, existing evaluations largely focus on final repositories and are typically limited to machine learning (ML). We introduce AgentActionBench, a process-oriented benchmark for evaluating agent-based experiment reproduction across ML and AI4Science domains. Our framework uses an MCP-based Action Recorder to capture agents' behaviour throughout the reproduction process and evaluates the resulting traces with paper-specific rubrics. AgentActionBench contains 150 papers, including 120 ML papers and 30 AI4Science papers. A human-annotated subset covering 10% of the benchmark provides validation data, while model-assisted augmentation expands the full benchmark to more than 10,000 rubric items. Experimental results show that current systems remain limited, with execution as the primary bottleneck. Meanwhile, the strong Pearson and Spearman correlations between model-generated and human-annotated rubrics validate the reliability of our scalable rubric-generation approach. --- ### Benchmark Radar: A Living Database and Search Engine for AI Benchmarks and Evaluation Authors: Koutian Wu, Junjie Zhou, Ergan Shang, Jiayu Wang, Pengqian Han, Junkai Wang, Wanghan Xu, Songyuanyi Lu, Lin Shi Published: 2026-09-10 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2609.11115v3 Benchmark researchers and developers of large language models (LLMs) and other AI systems need to find relevant evaluations, locate their benchmark datasets and code, and understand the settings behind reported scores. We present Benchmark Radar, a living database and search engine for retrieval and discovery of AI benchmarks, covering LLM evaluation, agentic and tool-use benchmarks, coding, reasoning, safety, and domain-specific evaluations. The system combines daily discovery of benchmark papers, repositories, datasets, and releases with a searchable benchmark catalog, mentions in model cards and technical reports, and score histories. It retains source identities and citations so readers can inspect candidate benchmarks and their evaluation evidence. Daily discovery draws on 37 sources: 13 direct connectors and 24 first-party research and engineering feeds. The catalog contains 1,283 source records drawn from 4 benchmark catalogs and 12,916 numeric observations on 790 records. We describe collection and retrieval, audit the full catalog, and examine benchmark saturation, adoption trends, and the limits of score comparisons. A worked example walks through a complete prior-art search, showing how to query the catalog and inspect benchmark evidence when designing a new evaluation. We release the web dashboard with a benchmark leaderboard, a Pareto frontier view of score against measured use, saturation and trend views, daily feeds, downloadable evidence, a command-line interface (CLI) for offline queries, and reproducible analysis. --- ### How AI Coders Discuss, Disagree, and Reach Consensus: Challenges and Opportunities for LLM-Based Qualitative Coding Authors: Jeongyeon Kim, John Mitchell Published: 2026-09-10 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2609.11109v1 The utility of AI in multi-coder qualitative coding has been widely discussed, yet little empirical evidence exists to delineate the contexts in which it performs reliably. We address this gap by quantifying the effectiveness of multi-agent LLM coding across varied qualitative datasets, revealing key contextual and structural factors that mediate coding outcomes. We developed a literature-informed baseline pipeline that enables AI agents to independently code, debate, and reconcile disagreements. Results revealed that coding accuracy depends on factors such as codebook length, qualitative data similarity, and agent disagreement. Notably, intense and unresolved debates between agents led to higher accuracy. Our analysis showed that while LLMs emulate many human discussion behaviors, they lack adaptive responsiveness to context. From these findings, we offer design recommendations for building automated coding systems. Our open-source AI discussion dataset and methodological framework lay the groundwork for advancing the design of AI-mediated automated thematic analysis. --- ### ProMediConv: Benchmarking Proactive Conversational Agents in Legal Dispute Mediation Authors: Zesheng Wei, Mengfan Li, Wenhao Liu, Yixin Zhang, Zilei Wang, Yang Deng Published: 2026-09-10 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.11101v1 Dispute mediation is essential for maintaining social harmony and resilience, yet developing skilled mediators is costly and time-consuming. Existing LLM-based mediation research remains limited by unrealistic task formulations, low-fidelity datasets, and coarse evaluation metrics that obscure turn-by-turn dynamics. To address these gaps, we introduce ProMediConv, a novel benchmarking framework that models mediation as a proactive, multi-stage, and party-aware dialogue process incorporating 11 mediation strategies and four party behavior pattern (BP) states. Using 972 complete real-world cases, we construct a high-fidelity mediation dataset with utterance-level annotations of strategies and BP states. Furthermore, to better assess agent impact, we propose MAD (Mean Attribute Difference), a fine-grained metric that captures BP shifts throughout the dialogue. Leveraging this framework, we establish a comprehensive benchmark by evaluating diverse models alongside our tailored baseline ProMediAgent. Extensive empirical analyses reveal critical behavioral phenomena and underscore the persistent challenges current models face in dynamic, multi-party mediation. Ultimately, ProMediConv provides a rigorous foundation and a vital quantitative standard for advancing AI-assisted conflict resolution. Our dataset and codebase are accessible at https://github.com/ZsWei66/ProMediConv_repo. --- ### Beyond Solver Verdicts: Generative Reward Models for Autoformalization Authors: Vikash Singh, Debargha Ganguly, Aman Goel, Ali Torkamani, Xiaoxue Han, Joseph Lilien, Ferhat Erata, Vipin Chaudhary Published: 2026-09-10 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.11085v2 Neurosymbolic systems rely on mathematical solvers to guarantee reasoning correctness, yet solvers are fundamentally blind to whether a formal translation maintains strict reference-equivalence to a designated formalization. We formalize this vulnerability as Verdict-Preserving-Unfaithfulness (VPU): a failure mode where an incorrect encoding executes successfully and matches the expected verdict. We theoretically prove that structural, verdict-only verification heuristics are mathematically bounded to chance-level detection on these deceptively valid traces. To resolve this, we introduce Generative Verification (GenV), which distills an offline Z3-equivalence oracle into a reference-free, continuous reference-equivalence score by repurposing the language model's native vocabulary space. Mechanistic analysis via decision-projected logit lenses and sparse autoencoders shows this generative readout natively extracts precise spatial error coordinates without explicit localization training. Empirically, our oracle-mined verifier (GenV+HN) achieves 0.961 AUROC in reference-equivalence verification, generalizes zero-shot across unseen translators and divergent formal styles, and yields an 11.3-point downstream accuracy gain in agentic test-time compute allocation. --- ### TailProp: content-adaptive light- and heavy-tailed propagation for vision Authors: Jiahao Kong, Zihan Li Published: 2026-09-10 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.11081v1 Science-inspired vision models show that explicit propagation dynamics can provide structured and interpretable alternatives to conventional token mixing. Existing formulations, however, typically construct and adapt visual propagation within a particular dynamical family, while visual representations can require substantially different spatial interactions across samples, channels, and network stages. We explore cross-regime adaptive propagation and introduce TailProp, a hierarchical vision backbone built upon the Tail Propagation Operator (TPO). TPO uses Gaussian and Cauchy stable-process propagators as complementary bases with rapidly decaying and heavy-tailed spatial influence, and predicts a content-conditioned channel-wise coefficient to adaptively combine them. Because this coefficient is spatially shared, the two responses are fused directly in the DCT domain with a single DCT/IDCT pair, yielding $O(N^{1.5})$ spatial mixing for square feature maps with $N=HW$ and fixed channel width. Across image classification, object detection, semantic segmentation, robustness, and cross-backbone restoration, TailProp consistently outperforms matched propagation baselines; TailProp-B reaches 84.4% Top-1 accuracy on ImageNet-1K, 50.3/44.8 box/mask AP under the 3x Mask R-CNN schedule, and 50.8% mIoU on ADE20K. Controlled ablations further show that these gains are not explained by single-basis propagation, an additional same-family branch, or within-family adaptive order alone, supporting complementary two-basis propagation as an effective design principle for visual representation learning. --- ### Conformal-DRO: Distributionally Robust Optimization with Conformalized Ambiguity Set Authors: Luhao Zhang, Shixiang Zhu Published: 2026-09-10 Categories: math.OC, stat.ML arXiv: https://arxiv.org/abs/2609.11073v1 Data-driven distributionally robust optimization (DRO) typically treats the conditional outcome law as fixed and uses ambiguity sets to capture estimation error. This paper studies latent distributional heterogeneity, where each instance has an unobserved law but contributes only one observation, so uncertainty persists even if the mixture law is known. We propose Conformal-DRO, which uses nested conformal regions to construct an ambiguity set for the future latent law. Under exchangeability, the set covers this law with probability at least $1-α$ in finite samples, without estimating underlying latent laws or their mixing mechanism. The conformal path induces a data-driven transport geometry, while $α$ determines the radius. The worst-case problem reduces to a finite linear program over conformal shells and admits sparse adversarial solutions. The resulting robust value provides a finite-sample certificate for the selected decision's expected cost. --- ### Coherent Floquet quantum reservoirs for molecular property prediction Authors: Luofei Wang, Da Zhang, Congren Wang, Yiming Li, Yuxiao Yang, Xuan Zhang, Xuefeng Cui, Zhang-Qi Yin Published: 2026-09-10 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2609.11071v1 Quantum reservoir computing (QRC) uses quantum dynamics to represent input histories for prediction through a trained classical readout. Discrete time crystals (DTCs) exhibit robust subharmonic responses under periodic driving, and previous work has used their dynamics to construct DTC-QRC. Here we construct a DTC-based reservoir architecture to predict molecular properties from structural and dynamical observations. Coherent Floquet evolution processes local molecular graph events and surface-hopping frames, while controlled reset regulates the contribution of earlier inputs. Measurements at the end of each input sequence yield a feature vector of fixed dimension. Trained classical decoders use this vector for inhibitor-activity and blood--brain-barrier permeability classification and electronic-gap forecasting, while the reservoir parameters remain fixed during training. With matched input lengths and output widths, DTC-QRC outperforms echo-state networks on long-prefix graph classification and the studied ethene gap forecasting tasks. Dephasing lowers performance in both applications, consistent with a role for coherent propagation. Experiments on the Quafu superconducting quantum cloud platform show that pair observables retain task information under device noise. The architecture provides a common framework for molecular screening and time-resolved property prediction using quantum reservoir computing. --- ### When Noise Fabricates Bias: The Fragility of LLM-as-a-Judge Bias Measurement under Noisy Text Authors: DongHyun Ryu, Jaehyeok Lee, YeongJun Hwang, JinYeong Bak Published: 2026-09-10 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.11067v1 Large language models are increasingly used as judges to measure social bias in text, yet the passages they judge are often noisy, containing typos, informal spelling, and broken punctuation. The consequences of such surface noise for social bias measurement remain unclear. To investigate this question, we apply five realistic noise conditions at multiple intensity levels to 3,822 stereotype-related responses and compare the resulting bias judgments with those on the original text. We find that such surface noise does not degrade bias measurement symmetrically: it is far more likely to turn neutral judgments into biased ones than biased judgments into neutral ones, by up to a 120x margin. We further observe two non-obvious effects across four LLM judges: in the most fragile judge the distortion is at its purest at mild, realistic noise levels, where erasure is scarcest, and as judges grow robust it attenuates toward parity rather than reversing. Bias measured on noisy text is therefore systematically overestimated, most in the categories that matter most for fairness. --- ### Less can be More: What Aspects of Speech Drive End-of-Turn Detection Authors: Rini Sharon, Manickavela A, Kadri Hacioglu, Andreas Stolcke Published: 2026-09-10 Categories: eess.AS, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2609.11066v1 In conversational AI, detecting when a speaker has finished talking is crucial for natural turn taking. While recent work incorporates semantics, the relative contribution of different modalities remains unclear. We present a controlled ablation of acoustic, prosodic, and semantic signals for streaming end of turn detection using a lightweight trimodal classifier. Under identical training conditions, the acoustic prosodic combination achieves the best balance of accuracy and latency, achieving utterance F1 of 0.93 with 7.8% false alarms at 400ms median latency. Adding text increases premature detections without improving performance. Feature space analysis confirms that prosodic features have the strongest class separability, while text representations overlap substantially. These findings suggest that turn-taking is primarily conveyed through intonation and silence patterns rather than semantic completeness, enabling faster and more reliable systems without expensive text inference. --- ### MOSAIC: Query-Aware Exploration Policy Adaptation for GraphRAG Authors: EunKyeong Lee, Kyeong-Jin Oh, Jinwon Kim, Hye Woo Lee, Minsang Song, Hyeongjun Jang, Junyoung Youn Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11065v1 Graph Retrieval-Augmented Generation (GraphRAG) can connect evidence distributed across a corpus graph, but most systems use largely shared exploration procedures across queries. This creates a structural mismatch: direct facts may need compact local neighborhoods, comparisons need balanced coverage of multiple targets, and mediated questions may require deeper paths through weakly related connectors. We present Mosaic, a training-free framework that formulates GraphRAG retrieval as a per-query control problem. An LLM analyzer converts query-specific evidence requirements into a bounded policy over seed selection, graph traversal, stopping, and evidence selection, while the corpus graph, indexes, scoring functions, grounding procedure, and answer generator remain shared. On GraphRAG-Bench, Mosaic achieves query-weighted Answer Correctness of 76.97 on Medical and 64.33 on Novel, improving over the strongest previously reported overall results by 5.13 and 4.43 points. On Medical, it reaches 95.1 Evidence Recall and 86.1 Context Relevancy. Controlled comparisons on an identical graph and generator show that no fixed narrow, medium, or wide policy is consistently optimal; Mosaic improves by 9.96 points over the strongest canonical fixed policy. Relative to Fixed Wide, it evaluates 81.9% fewer paths and retains 47.2% fewer evidence items. Transfer experiments on HotpotQA, MuSiQue, and 2WikiMultiHopQA further show that the policy interface can be applied without benchmark-specific retriever training. --- ### The information geometry of large language models is shared, learned, and controllable Authors: Dario Picozzi Published: 2026-09-10 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2609.11063v1 Large language models learn similar behaviours, yet it remains unclear what structure they share or how to change one behaviour without disturbing others. The Fisher-Rao geometry of next-token probabilities connects these questions: behaviour determines this geometry up to output-preserving symmetries, whereas activation geometry depends on coordinates. Across transformer, state-space and recurrent models, output geometries agree more strongly than activation geometries, and shared geometry supports semantic-category transfer. Agreement with human word choices increases with predictive accuracy, scale and training, and improves further after model-only calibration. Token probabilities and read-out geometry jointly predict the spectrum and its effective dimension. Controlled language assignments show that geometry follows the language law across architectures. Pretraining corpus statistics predict held-out fact acquisition without recalibration, while randomised experiments show that deeper evidence substantially delays acquisition across every tested architecture and evidence construction. Finally, the geometry prescribes minimum-disturbance local interventions, predicts their relative cost, and supports reusable control: updates learned on donor prompts transfer to unseen prompts while better preserving behaviour on reference prompts than Euclidean control. The same geometric correction improves steering, editing, attribution, dictionary learning and fine-tuning. --- ### Fork Where the Model Changes Its Mind: Belief-Shift Branching for Tree-Structured Reinforcement Learning Authors: Bin Lei, Yu Li, Prafulla Kumar Choubey, Jiaxin Zhang, Becky Xiangyu Peng, Qinyuan Ye, Kartik Narayan, Caiwen Ding, Silvio Savarese, Chien-Sheng Wu Published: 2026-09-10 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.11061v1 Tree-structured rollouts give critic-free reinforcement learning with verifiable rewards (RLVR) step-level credit: fork a chain at an intermediate point, and sibling outcome differences estimate step value. Each fork adds sampling cost, so realistic budgets typically allow only a few forks per chain. A fork placed where the outcome is already largely settled yields siblings that mostly agree and provide almost no credit signal; hence, for a given tree size, where forks are placed largely determines how much step-level RL can gain. Most existing mainstream methods place forks by structure, such as fixed lengths, midpoints, and delimiters, or by next-token entropy. We formalize fork placement as locating the \emph{pivots} of the chain's value curve, where the expected outcome turns. We propose \emph{belief-shift branching}: read the model's answer belief at candidate boundaries and fork just before the step where consecutive beliefs diverge most. Three instantiations, none needing step-level supervision, span access levels: a black-box probe, a logit-lens depth profile, and a learned activation direction, which is fit offline and therefore used only in the validation before RL training. The signal only \emph{places} forks, and the probe costs about $1\%$ of step compute on mathematics and under $5\%$ on code when it runs inside the rollout engine. In that validation, against Monte-Carlo value curves, a belief-shift signal ranks first in each of the eight model$\times$benchmark panels, ahead of entropy, structural, and LLM-judge baselines. In RL across three model families and two domains, belief-shift forking leads every mathematics aggregate, on OLMo-3-7B by $+2.6$ aggregate and $+2.9$ on AIME 2026 over the strongest baseline, and sweeps every OLMo code column, by $+6.5$ on LiveCodeBench-medium. --- ### Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents Authors: Susheel Suresh, Hazel Mak, Sahil Bhatnagar, Chhaya Methani, Alejandro Gutierrez Munoz Published: 2026-09-10 Categories: cs.AI, cs.SE arXiv: https://arxiv.org/abs/2609.11060v1 Persistent memory is entering production-oriented agent platforms to help long-horizon agents accumulate experience across sessions. Yet a post-task curator agent restricted to completed trajectories can preserve errors, overgeneralize partial evidence, or retain stale knowledge. We introduce environment-probing curation, a deployment-compatible extension that gives an existing asynchronous curator agent least-privilege, read-only world tools to check, scope, and refresh candidate memories. It requires no model retraining and leaves the task agent, retriever, memory representation, and production write authority unchanged. In a production-like GitHub Copilot (GHCP) harness built on its SDK, we compare stateless execution, full in-context learning, GHCP + Mem, and GHCP + Mem (w/ Env Probing) on CLBench database exploration and 90 adapted APEX management-consulting tasks. On CLBench, probing raises pass rate from 39% to 73% and pass-discounted reward from 8.60 to 22.60 while reducing queries from 8.8 to 4.7 per question and task-agent cost from \$3.38 to \$1.68. Across six APEX worlds, all 18 memory-versus-baseline mean reward comparisons are positive and task-agent tool calls fall by 16--75%; probing gives the best task-agent reward gain per dollar in five worlds. Probing also attains higher mean reward than GHCP + Mem on both Sonnet 4.6 and Opus 4.7 without schema drift. Environment probing therefore turns existing agent-memory curation into an environment-informed, auditable process while preserving a compact task-time interface. --- ### EMMI: Edge Multi-Modal Intelligence for Communication-Efficient MLLM Inference via Fused Representation Compression Authors: Motahare Mounesan, Irfan Khan Published: 2026-09-10 Categories: cs.LG, cs.DC arXiv: https://arxiv.org/abs/2609.11058v1 Recent advances in multimodal large language mod- els (MLLMs) have opened new opportunities for edge intelligence by enabling reasoning across heterogeneous sensor modalities, such as vision, text, and telemetry data. However, deploying these capabilities on resource-constrained edge platforms remains challenging due to the substantial computational, memory, and communication demands of modern MLLMs. Rather than transmitting raw sensor observations or partitioning neural networks at intermediate layers, Edge Multi-Modal Intelligence (EMMI) communicates a compact representation between edge devices and server resources, enabling communication-efficient edge MLLM inference. To achieve this, EMMI performs modality-specific encoding, cross-modal representation fusion, and learned compression at the edge, transmitting only a compact latent representation to server-side resources for high-capacity MLLM reasoning. This representation-centric design reduces communication overhead, preserves local data privacy, and provides a fixed-size interface between heterogeneous edge devices and server-side MLLMs. Evaluation on a representative multimodal benchmark demonstrates that EMMI can reduce the communication payload by 32x while maintaining comparable downstream accuracy, resulting in up to a 3.4x reduction in estimated end-to-end inference latency under bandwidth-constrained edge conditions. --- ### T1: Terminal Agent Reinforcement Learning for Long-Horizon Tasks Authors: Junyao Yang, Yucheng Shi, Zhongzhi Li, Ruhan Wang, Zongxia Li, Haitao Mi, Leowei Liang Published: 2026-09-10 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.11042v1 Agent usage is shifting toward long-horizon tasks such as coding and scientific discovery, among which terminal tasks are especially important. We introduce T1, a Mixture-of-Experts model of 122B total trained with reinforcement learning, operating a real shell in a cloud sandbox for up to 300+ tool-call turns per task, rewarded by executing each task's own verifier. We provide a comprehensive recipe: First, an aggressively warm-started to stabilize actor-critic training, with a dense process reward scoring trajectories by the absolute number of passing verifiers. Second, stable optimization through TITO construction, training on the exact sampled token identifiers with drift repair at turn boundaries, and rollout routing replay, recording the sampler's per-token expert choices at every MoE layer and replaying them during training. Third, fully out-of-distribution training corpus: isolated seeds and synthesized tasks disjoint from Terminal-Bench 2.1 ensures gains reflect genuine capability transfer over benchmark overfitting. Together, TITO and R3 cut the training-to-inference log-probability difference from 0.021 to 0.013, with exactly aligned zero token drift in the loss region. On Terminal-Bench 2.1, our post-train pipeline raises initial base model from 43.8% to T1 with 64.0% resolved. On Long-Horizon Terminal Bench, T1 reaches 27.9% and surpasses GPT-5.4 and GLM-5.1. ---