# Research Papers — 2026-10 · bundle page 63 of 156

> Full-text markdown of papers #3101–3150 of 7799 published in 2026-10, newest first. AI-readable mirrors of the full 7799-paper 2026-10 archive (50 papers per file).

Canonical page: https://aipod.dev/papers/archive/2026-10/page/63
[← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/62.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-10/page/64.md)

### Chunk-wise Model-based Policy Optimization

Authors: Weixi Song, Donglin Wang, Yong-Lu Li
Published: 2026-10-10
Categories: cs.RO, cs.LG
arXiv: https://arxiv.org/abs/2610.14564v1

Model-based reinforcement learning has shown superior sample efficiency by leveraging the learned latent dynamics either in the form of derivative-free planning/imagined rollout training or dynamics gradient-based policy updates. While these approaches achieve comparatively fast learning, they all suffer, to varying degrees, from issues of recurrent compounding errors, high inference-time latency, and instability in backward propagation through time (BPTT). In this work, we propose a model-based RL approach that uses the chunk world model as a differentiable training-time evaluator to optimize the chunk actor without recurrent model invocation. Through the collaborative chunk-wise prediction of the world model and policy, we can generate H-step hypothetical rollouts in a single forward pass and directly optimize value expansion, avoiding the costly and unstable BPTT. We evaluate our method on 19 challenging high-dimensional humanoid tasks from HumanoidBench and deploy our method to the real-world for quadruped bipedal walking. Project Website: https://song-wx.github.io/cmpo.github.io/

---

### PGSB: Pretrained-Guided Shared Basis for LoRA Model Merging

Authors: Muqing Liu, Chongjie Si, Zhuoya Liu, Yuheng Jia
Published: 2026-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.14558v1

Large pretrained models have achieved remarkable success across diverse domains, yet adapting them to multiple downstream tasks remains challenging: full fine-tuning is costly, while maintaining separate task-specific models is storage-inefficient and fails to produce a unified multi-task model. Low-Rank Adaptation (LoRA) offers an efficient alternative for Model Merging by representing each task adaptation as a low-rank update over frozen pretrained weights. However, existing LoRA merging methods often treat task-specific updates as directly comparable objects and merge them in their original low-rank parameter spaces. This overlooks two key factors: which update directions provide reliable shared support across tasks, and how these directions should be re-parameterized with respect to the pretrained weights before merging. To address this limitation, we propose Pretrained-Guided Shared Basis (PGSB), a two-stage framework for LoRA model merging. PGSB first identifies shared directions among task-specific LoRA updates, and then re-parameterizes these directions using their interaction with the pretrained weights, producing a more suitable basis for merging. Extensive experiments demonstrate that PGSB consistently outperforms state-of-the-art baselines across multiple benchmarks. Codes are available at https://github.com/Kitschnada/Pretrained-Basis .

---

### Adversarial Gradient Forcing for Robust Few-Step Autoregressive Video Generation

Authors: Chenxi Xia, Yixuan Xu, Shan Liu, Jiapeng Wang, Maohua Li, Hanlin Tang, Kan Liu, Tao Lan, Weiyi Lu, Xinggao Liu
Published: 2026-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.14553v1

Autoregressive video generation models generate videos sequentially, with each new block conditioned on previously generated blocks. By training on the model's own rollouts, Self Forcing and its follow-up methods substantially mitigate exposure bias caused by the mismatch between training- and inference-time history distributions. However, these models still suffer from error accumulation during long-horizon generation. We identify a key underlying limitation: the model remains highly sensitive to errors in its generated history, allowing small deviations to be amplified through subsequent generations. We refer to this underexplored problem as the local history robustness gap. To address this gap, we introduce Adversarial Gradient Forcing (AGF), a framework designed to improve the robustness of few-step autoregressive video models to perturbations and accumulated errors in their own generated histories. AGF uses distribution matching gradients from future block predictions to select adversarial perturbations of self-generated histories. It then jointly trains the generator on the original and perturbed histories to improve the robustness of future generation. Extensive experiments show that AGF achieves the highest VBench Total score on 60-second video generation and the best visual quality on four-minute video generation. Notably, AGF achieves such long-horizon robustness while requiring only 5-second rollouts for training.

---

### LASER: Defeasible Certificate Search for Long-Range Protocol Dependencies

Authors: Zihan Tang, Huibiao Zhu
Published: 2026-10-10
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2610.14552v1

Protocol specifications rarely justify executable behavior in one place: a transition may inherit its source state from a remote control block, name its events only by number, and remain subject to a later correction. Recovering the nearby clause is therefore insufficient; the system must assemble a source-linked certificate and determine whether any stronger rule applies. We introduce LASER, a proof-carrying search method for sparse, defeasible protocol evidence. LASER represents missing support and unresolved corrections as explicit obligations, searches typed evidence routes, and lets an occurrence-aware symbolic checker alone decide whether a certificate warrants, refutes, or leaves a behavior unresolved. On 91 BehaviorCheck families under a matched 2K budget, LASER improves Joint Edge+Proof over the strongest local reader on two model families ($Δ=+.132$, $p=.017$; $Δ=+.198$, $p<.001$) while cutting false discoveries by 2--3$\times$. Controlled replays attribute the gain and its limits: neither reference routing nor deterministic citation expansion alone exceeds the local baseline; whole-document reading (29--65K tokens), two-round iterative retrieval, and even oracle minimal-certificate evidence all fail to reach LASER, localizing the bottleneck in typed obligation closure rather than evidence access. An authority-corruption replay shows source-link errors only cause abstention, while the only unsafe-warrant mechanism is mis-recording a Held erratum as Verified, confined to that erratum's family. LASER thus improves long-range dependency recovery while converting incomplete evidence and uncertain authority into explicit residual obligations rather than unsupported behavior.

---

### LLM-based User-Product Relevance Evaluation for Dynamic Product Advertising

Authors: Wen Zhang, Xiao Bai, Hsiang-wei Chao, Mengfan Zhu, Tong Zhao, Mingxuan Ju, Xiaodong Liu, Congfei Zhang, Yunzhi Zhou, Yajun Wang, Jinchao Li, Yu Zhang
Published: 2026-10-10
Categories: cs.IR, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2610.14550v1

Dynamic Product Advertising (DPA) retrieval systems are a core component of modern e-commerce advertising; however, their evaluation remains challenging. Conventional offline engagement metrics are limited by sparse, biased feedback, while online evaluation is costly and time-consuming. In this work, we propose a user--product relevance metric for offline evaluation of DPA retrieval systems. The metric first compresses the engagement history into a compact set of intent representatives via Semantic-ID-grouped, event-stratified segmentation, then assesses the relevance of a recommended product to each individual intent and aggregates them as a user-level relevance score, yielding a stable and interpretable measure of retrieval quality. To compute this metric at scale, we leverage large language models (LLMs) as semantic relevance judges, capturing nuanced functional and semantic relationships that heuristic methods often miss. We propose an agentic, human-aligned prompting framework that generates multi-category relevance guidelines and distills teacher-model judgments into an efficient student model for scalable evaluation. The resulting metric strongly correlates with offline and online business metrics, enabling faster and more reliable DPA retrieval iteration before A/B testing.

---

### Active Neural Matter: A Finite-Window Response Framework for Physical Systems and Distributed AI Workflows

Authors: Tianchi Chen, Shihua Yu
Published: 2026-10-10
Categories: cs.LG, cond-mat.stat-mech, cs.AI, cs.DC, physics.app-ph
arXiv: https://arxiv.org/abs/2610.14549v1

Physical systems and distributed AI workflows can retain past information, receive delayed feedback and combine conflicting signals. These features pose a common modeling problem: determining what information about past interactions a state must retain to account for future responses. Active Neural Matter (ANM) addresses this problem through a finite-window response framework. Forward analysis uses each system's dynamics to predict state and output responses to perturbations, separating propagation from output sensitivity. Inverse analysis tests whether a candidate state retains enough information for these responses. With the system and future inputs or interaction rules fixed, distinguishable responses from histories assigned the same state identify distinctions the representation must preserve. Their distributional separation gives a lower bound on the worst-case error of approximating behavior from that state. Uncertainty bounds distinguish rejection, support within a fixed tolerance on tested comparisons, and unresolved cases. Delayed-consensus simulations check analytical predictions of response amplitude and phase over a narrow range of delays; flocking illustrates dependence on the chosen observable. An exact delayed-return construction establishes a sufficient state for a specified program. The AI studies identify distinctions in interaction history and evidence order that a representation must preserve to reproduce observed response distributions. Tests on new histories in a controlled feedback task provide bounded support for a state specified by the task instructions; the instructed rule predicts every observed action. ANM connects response prediction to measurable information requirements, separating behavioral fidelity from task correctness and initial-state adequacy from a closed evolution law for that state.

---

### A Self-Defining Operator for Distributed Applications

Authors: Shihang Li, Thomas Anderson, Ratul Mahajan, Simon Peter, Luke Zettlemoyer
Published: 2026-10-10
Categories: cs.DC, cs.AI
arXiv: https://arxiv.org/abs/2610.14547v2

To operate distributed applications with high availability, operators must rapidly diagnose and mitigate any incident. LLM-based agents can help, but unlike experienced human operators they do not improve as they handle more incidents. Each investigation starts from scratch, re-deriving likely causes and re-testing hypotheses even when a similar fault was resolved in the past. We present a self-defining operator (SDO) that improves with time by updating its operational behavior and memory. When SDO resolves an incident, it distills that experience into playbooks, which describe reusable diagnostic instructions for future incidents with similar faults, and detectors, scripts that take application telemetry as input and output relevant playbook(s). SDO continuously runs the detectors and invokes LLMs only when one fires, so its token cost scales with incident count rather than with operational time. Across 11 open-source microservice applications, SDO cuts resolution time by 52% on recurring faults while keeping per-incident cost low. SDO is open source at https://github.com/self-defining-systems/sdo.

---

### InfraWorld: Motion-Aware World Modeling for Roadside 4D Occupancy Forecasting

Authors: Songkai Wang, Xiaokai Bai, Lei Yang, Hui-liang Shen
Published: 2026-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.14544v1

Roadside 4D occupancy forecasting predicts future semantic occupancy in a fixed infrastructure coordinate system. Roadside scenes are globally structured yet locally evolving, requiring comprehensive scene-level context while precisely modeling sparse localized motion. However, existing forecasting paradigms struggle to tackle both requirements. To address this limitation, we propose InfraWorld, a world model formulating roadside occupancy forecasting as selective state evolution. InfraWorld maintains a dense BEV state for the complete scene and persistent motion queries for moving regions, assigning complementary states to global context and local dynamics. Specifically, we propose Dynamic Query Initiation (DQI) to initialize queries in regions supported by motion in the observed history. During rollout, the dense state is updated through Long--Short Temporal Fusion (LST), which combines short-term adjacent-frame continuity with long-term motion trends to preserve stable context while capturing scene evolution. Motion-Aware Query Injection (MAQI) then refines the queries and injects confidence-weighted query information into corresponding BEV tokens, enabling targeted dynamic updates without perturbing global context. Together, InfraWorld concentrates predictive capacity on evolving traffic regions. Extensive experiments on the InfraOcc and Occ3D--nus datasets demonstrate that InfraWorld achieves state-of-the-art performance for roadside 4D occupancy forecasting while retaining its effectiveness under ego-centric settings. Code will be released.

---

### Are Multimodal Benchmarks Really Useful? Item-Level Multimodal Benchmark Diagnosis via Structure-Response Co-Calibration

Authors: Shiqi Zhang, Weixin Zeng, Ziheng Zhang, Jiuyang Tang, Xiang Zhao
Published: 2026-10-10
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.14542v1

Multimodal benchmarks are central to evaluating whether models can reason over information from multiple modalities. However, existing literature has revealed that some benchmarks may contain shortcut items that can be answered from a single modality or from answer-option regularities alone, thus failing to reflect models' true cross-modal inference ability. Existing methods for assessing benchmarks typically rely either on intrinsic item structure or extrinsic model response behavior, which are not adequate for a comprehensive diagnosis. To fill in this gap, we propose SRCoD, a structure-response co-calibrated diagnosis framework for item-level diagnosis of modality dependence in multimodal benchmarks. SRCoD estimates sample-wise multimodal information structure from benchmark content and target answers, and uses this signal as a soft structural anchor for modality-conditioned item response modeling. The learned model produces a calibrated item-level diagnostic profile that reflects how strongly an item depends on different forms of modality evidence, enabling fine-grained benchmark analysis beyond aggregate accuracy. SRCoD also provides interpretable attribution of the diagnosis results. Extensive experiments under two complementary protocols, i.e., benchmark-refinement oriented validation and human-labeled item validation, demonstrate that SRCoD outperforms state-of-the-art benchmark diagnosis methods.

---

### GG-Nav: Goal-oriented Grounding Chain-of-Thought Elicits Reasoning in Navigation Foundation Models

Authors: Shaohao Zhu, Wensi Huang, Siqi Zhang, Yuqiang Yang, Meng Wei, Chenyang Wan, Tianle Liu, Tai Wang, Jinming Xu
Published: 2026-10-10
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2610.14541v1

Emerging strong Vision-Language Models (VLMs) have advanced navigation foundation models. However, these models still have unsatisfactory performance on goal-oriented navigation settings. In this paper, we first conduct a motivating study revealing that the key issue lies in the insufficient utilization of streaming observations and imprecise target localization. To tackle these challenges, we propose GG-Nav, a simple yet effective navigation framework that learns a structured chain-of-thought (CoT) mechanism via supervised fine-tuning to explicitly localize goal-relevant instances, accumulate grounding evidence, and support more reliable decision-making. Specifically, the structured CoT is designed with two key stages: (i) Observation-to-Evidence, which reasons over all intermediate frames between adjacent decision steps to retain useful goal-related evidence; and (ii) Evidence-to-Decision, which integrates the current observation with accumulated evidence to ground the target and support the generation of navigation actions or clarification questions. Extensive experiments show that GG-Nav outperforms the state-of-the-art methods in terms of success rates on the HM3D ObjectNav, VL-LN, and CoIN benchmarks by 3.1%, 16.6%, and 5.3%, respectively. Ablation studies further show that accumulating evidence from intermediate observations reduces evidence-missing failures by 11.6%, while explicit grounding reduces stopping errors by 49.4%. We also deploy GG-Nav on a Unitree Go2 robot, demonstrating its zero-shot generalization capability in unseen real-world environments.

---

### FDE-Bench: Evaluating End-to-End Delivery from Underspecified Real-World Business Requests

Authors: Huaiming Li, Can Huang, Chufan Wu, Danyu Liu, Xiaomin Yuan, Qinze Li, Yui Lo, Wenhui Bai, Qianlong Wang, Zengle Ge, Xiyu Yan, Yuheng Cheng et al.
Published: 2026-10-10
Categories: cs.AI, cs.MA
arXiv: https://arxiv.org/abs/2610.14539v1

Real-world business requests often leave key requirements unstated because frontline practitioners struggle to articulate rules they know from experience. Forward-deployed engineers (FDEs) must elicit these requirements through dialogue and iteratively optimize solutions until they meet customer acceptance criteria. Existing optimization benchmarks typically start from explicit objectives, while clarification benchmarks often introduce information gaps into existing tasks. We introduce FDE-Bench, the first benchmark built from real customer projects to evaluate agents on end-to-end delivery tasks that require interactive requirement clarification and iterative optimization. Its 49 cases span combinatorial optimization and machine learning. Clarification targets are distilled from discussions between frontline practitioners and FDEs. Agents elicit missing requirements from a simulated customer, and case-specific evaluators assess final artifacts against acceptance criteria using customer-accepted FDE solutions as quality references. Across 18 agent configurations, the highest mean delivery score is only 0.5510, compared with an FDE reference of 1.0. The highest delivery success rate (DSR), defined as the fraction of planned runs reaching this reference, is only 6.12%. These results show that reliable FDE-level end-to-end delivery remains out of reach for the evaluated agents on underspecified real-world business tasks. The benchmark project page is available at https://baidubce.github.io/FM-Agent/FDE-Bench/.

---

### Learn the Dynamics, Enforce the Physics: Generative Latent Operators for High-Dimensional Beam Evolution

Authors: Haohan Zou, Alexander Scheinker, Yuanyuan Shi
Published: 2026-10-10
Categories: physics.acc-ph, cs.LG
arXiv: https://arxiv.org/abs/2610.14537v1

Particle-in-cell (PIC) simulation is indispensable for modeling beam dynamics in particle accelerators, but its computational cost limits many-query workflows, such as configuration exploration, virtual diagnostics, and controller design. We introduce GLOBE, a physics-constrained Generative Latent Operator for Beam Evolution, modeling the nonlinear, control-dependent, multiscale evolution of 3D beam--field dynamics covering $10^7$ spatial cells. GLOBE compresses beam--field states into a compact latent space and factors their evolution into two components. A control-conditioned neural operator learns the dominant dynamics, while an amplitude-normalized generative model captures the residual distribution, recovering fine-scale structures and conditional variations. GLOBE enforces field constraints by regime-specific, parameter-free operators: a Poisson solve reconstructs the electric field in the electrostatic regime, and a spectral projection imposes Gauss's law and a divergence-free magnetic field in the electromagnetic regime. On two high-resolution 3D particle-accelerator benchmarks, bunched-beam transport and plasma-wakefield acceleration, GLOBE lowers relative-$L_2$ error by 9\% and 13\%, low-frequency spectral error by 69\% and 51\%, and mid-frequency spectral error by 54\% and 20\% relative to the strongest baseline (DPOT), satisfies field constraints, and reduces hours of PIC simulation to a minute per rollout.

---

### LongHisDoc: A Comprehensive Benchmark for Chinese Long Historical Document Understanding

Authors: Zeyu Shan, Yongxin Shi, Shihang Wu, Peirong Zhang, Yuanyuan Fang, Zhaopeng Wang, Lianwen Jin
Published: 2026-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2610.14531v1

Chinese historical documents are invaluable cultural treasures, but their complexity poses significant comprehension challenges for non-specialists. Despite the success of Large Language Models (LLMs) and Large Vision-Language Models (LVLMs) in modern document tasks, their capabilities in Long (full-book) Historical Document understanding remain largely unexplored. To bridge this gap, we introduce LongHisDoc, the first benchmark tailored for long historical document understanding. LongHisDoc comprises 1,012 expert-annotated QA pairs derived from 101 PDF documents, with an average of 85.1 pages and 42,776.4 tokens per document. It spans 10 categories and features four QA types: single-page, cross-page, multi-hop, and unanswerable. Answering these questions requires integrating multimodal evidence from text, layout, and visual elements. Our evaluation reveals that the top-performing LVLM and LLM achieve only 62.9% and 69.9% accuracy, respectively, highlighting a substantial performance gap. Furthermore, we assess various modal retrievers and the end-to-end RAG performance to provide references for Retrieval-Augmented Generation (RAG) systems. We believe that LongHisDoc serves as a comprehensive benchmark to catalyze future research in full-book historical document understanding. The benchmark and evaluation code are available at https://huggingface.co/datasets/qweq12433454/LongHisDoc and https://github.com/SCUT-DLVCLab/LongHisDoc.

---

### Who to Trust in Multi-Robot Systems: Joint Robot and Object Hypothesis Trust Inference via Neural-Symbolic Cross-Validation

Authors: Ruiyang Wang, Hao-Lun Hsu, Haocheng Meng, Songqiao Xie, Miroslav Pajic
Published: 2026-10-10
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2610.14530v1

Multi-robot systems (MRS) offer significant advantages over single-robot platforms by enabling information sharing and coordination for different tasks in complex environments. However, reliable cooperation critically depends on inferring trust over both information sources and their reported hypotheses, particularly under partial observability and potential adversarial interference. Existing approaches frequently depend on pre-established trust relationships or side-channel information, which may not be available or reliable in realistic MRS deployments. This work introduces NeST-Bayes, a neural-symbolic framework that jointly infers robot and hypothesis trust using only inter-robot observation consistency, without requiring preassigned robot trust, reputation information, or side-channel identity signals. NeST-Bayes infers instantaneous trust evidence scores for robots and object hypotheses by reasoning over ego-centric relational graphs constructed from cross-robot observation consistency. These evidence scores are accumulated over time through Bayesian updates of Beta-distributed trust beliefs to maintain temporally consistent trust estimates. Extensive evaluations show that NeST-Bayes achieves the highest mean object precision and robot recall across all benchmark settings, and the highest robot precision and object recall in most settings, compared to trust-based methods. Under aggressive optimized attacks, it achieves 98.8\% robot precision with 93.8\% recall and 98.3\% object precision with 95.4\% recall. In an occlusion-rich Webots warehouse, it maintains 92.5\% robot precision with 92.0\% recall and 97.3\% object precision with 85.0\% recall. These results demonstrate joint improvements in the reliability and coverage of retained information.

---

### Energy Ring: Stabilizing Feature Energy for Reliable Test-Time Discovery

Authors: Tianle Liu, Fan Lyu, Tingliang Feng, Chenggong Ni, Tianqi Wang, Zhang Zhang, Fuyuan Hu, Rui Yao, Tao Zhou
Published: 2026-10-10
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.14525v1

Test-Time Discovery (TTD) aims to recognize known classes and detect novel ones while the model is already deployed, operating directly on a continual unlabeled test-time stream. Existing TTD methods are overly conservative, making known and unknown features collapse together and difficult to separate during test-time adaptation. In this paper, we revisit TTD from an energy perspective and find that the natural energy gap between known and unknown classes is unstable and easily collapses. We therefore propose Energy Ring to stabilize this gap by anchoring the two types of classes on separate energy rings. A lightweight adapter enlarges the norm-based gap while preserving known-class directions, and a perturbation consistency module forms compact clusters for unknown classes. During testing, energy-guided confidence weighting improves prediction, and a stability-based rectification mechanism corrects unreliable updates. Experimental results show that Energy Ring significantly improves TTD performance against existing TTD methods.

---

### ScienceBuddy-PV: A Verifier-Governed Recursive Improvement Protocol for Panvascular Intervention Research

Authors: Lingsen You, Yujun Guo, Xinyu Zhong, Zijie Feng, Wentong Wang, Li Shen, Junbo Ge
Published: 2026-10-10
Categories: cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2610.14523v1

ScienceBuddy-PV adapts the ScienceBuddy recursive-in-recursive design to panvascular intervention research through a structured evidence ledger, device-vessel ontology, abstention contract and retained failure bank. Recursive harness optimization and outer-loop model learning, including reinforcement learning, are specified but not executed. We audit a fixed component, rather than the complete agent system, using separate ARCADE syntax and stenosis tasks (1,000 training, 200 validation and 300 test images per task). A shared task-masked convolutional model was trained once per representation setting; its parameters and the heuristic support gate were then held fixed, with the stenosis threshold selected on validation data. On the locked test sets, syntax Dice was 0.475 (95% CI 0.459-0.490) and stenosis Dice was 0.231 (95% CI 0.214-0.249). Stenosis pixel AUROC was 0.957 and AUPRC was 0.180 at prevalence 0.00975. The gate accepted 76.7% of stenosis-frame outputs versus 100% for a score-only comparator. In a constructed mask audit, it accepted 44.3% of reference candidates and rejected 55.7%, exposing the cost of abstention. Acceptance does not establish correctness or clinical safety. This public-data component audit makes support and abstention measurable and provides a reproducible protocol for future collaborative agent studies. It does not demonstrate recursive improvement, patient-level validity or clinical device recommendations.

---

### Identication and Estimation of Nonsymmetric Determinantal Point Processes

Authors: Christian Gouriéroux, Yang Lu
Published: 2026-10-10
Categories: stat.ML, cs.LG
arXiv: https://arxiv.org/abs/2610.14520v1

Determinantal point processes (DPPs) provide tractable distributions for nite random sets, but the usual symmetric-kernel specication imposes nonpositive pairwise dependence. Statistical research on nonsymmetric DPPs, which can accommodate positive dependence, remains limited. We study identication and estimation for this class. First, we establish the equivalence of two sucient conditions for a nonsymmetric kernel to dene a valid DPP. Second, we develop the identication theory for nonsymmetric DPPs. Third, we develop cardinality-truncated maximum-likelihood estimation and estimation by averaged stochastic gradient descent. Under suitable identication and regularity conditions, we establish consistency and asymptotic normality for both estimators. Simulations examine identication, the eects of truncation, and the nite-sample performance of the estimators. We illustrate the methods with two cyber applications at markedly dierent sampling frequencies: weekly sector-by-breach data and second-by-second enterprise authentication data with more than ve million set observations.

---

### Query-Adaptive Neuron Steering for LLMs

Authors: Yanxi Pan, Jingwen Wu, Yan Ma, Xiang Chen, Xinyi Zeng, Xuetong Wang, Yunyan Zhang, Yue Zhang, Yu Tian, Jianfeng Pan, Hongyi Zhou
Published: 2026-10-10
Categories: cs.LG, cs.CL
arXiv: https://arxiv.org/abs/2610.14519v1

Effectively eliciting the capabilities of large language models (LLMs) at inference time remains an important challenge. Existing studies address this challenge through context engineering, which shapes user queries, or fixed steering, which directly intervenes in model computation. However, predefined contexts and fixed intervention strategies may not adequately accommodate the diverse knowledge and reasoning requirements of individual queries. This raises a natural question: Can internal steering in LLMs be made query-adaptive? In this work, we investigate whether query-induced internal activations can guide adaptive steering in LLMs. Our pilot studies reveal subject- and layer-dependent neuron activation patterns and show that amplifying subject-relevant neurons can improve model performance. Based on these findings, we propose Q-NeuS, a query-adaptive neuron steering framework that dynamically determines which neurons to modulate and by how much for each input. Specifically, Q-NeuS constructs a compact subject-guided candidate neuron space and employs a lightweight Query-Adaptive Gate to generate neuron-wise modulation coefficients. To learn effective intervention strategies, we introduce Neu-GRPO, which adapts group-relative optimization to train the Query-Adaptive Gate using the relative performance of multiple neuron intervention configurations. The base LLM remains frozen throughout training and inference, and no task or subject labels are required at inference time. Extensive experiments across multiple LLMs and benchmarks demonstrate that Q-NeuS consistently improves model performance and exhibits broad applicability across model scales and strong cross-benchmark transferability.

---

### We Have Been Interpreting Diffusion Transformer Backbone Behaviors Wrong

Authors: Benyuan Meng, Longtao Huang, Yang Liu, Liang Lin
Published: 2026-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.14517v1

During a network forward call, diffusion backbones naturally produce many intermediate activations. If we evaluate each activation against an identical downstream benchmarking protocol, we can plot an inverted-U-shaped curve as the depth profile: the activation's performance goes up first and then down as we go into deeper layers. This is common knowledge in the diffusion model community, which inspires many applications including the widely-adopted REPA. However, the common interpretation of this depth profile may be inaccurate and incomplete: (i) The cause of this curve has long been treated as signal granularity changing from coarse to fine, but there has been little evidence in DiT backbones. (ii) The community has been mainly studying this phenomenon in a static way, without investigating how it changes across different training setups. In this work, we show a different mechanism picture for the profile, where the prediction target (e.g., $ε$, $u$, $x_0$) sits at the center. The semantically meaningful signal is only an intermediate by-product for predicting the target, while the prediction target is the real goal for the backbone. Furthermore, the choice of prediction targets as part of the training setup can predictably affect the curve's shape. With this updated interpretation, many applications of this depth profile can be better grounded and predicted. As an example, we show how to predict the near-optimal placement of REPA based on the prediction target of the training setup. The code is at https://github.com/Darkbblue/dit-compare.

---

### Compositional Concepts Erasing: Disentangling Object-Style via Domain Manifold Unlearning

Authors: Yang Xiao, Kaiyuan Deng, Gen Li, Zinan Ling, Zhiwen Fan, Xiaolong Ma, Bo Hui
Published: 2026-10-10
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2610.14514v1

Diffusion models pose significant risks regarding the generation of illicit content and copyright violations. This has catalyzed the development of machine unlearning to erase harmful concepts or copyrighted styles from a model's generative repertoire without compromising its general performance. However, previous unlearning approaches on diffusion models have largely focused on either object-level or style-level removal and failed to address the challenge of object-style combination unlearning. At the same time, multiple unlearning requests often lead to catastrophic forgetting. To address these challenges, we propose Domain Manifold Unlearning (DMU), which uses an EMA-updated feature proxy extracted from intermediate diffusion features to represent the retained target-object direction. As a lightweight estimating representation of an object, DMU enables selective unlearning of domain-specific components in a specific object while preserving other semantic representations, thereby significantly enhancing robustness in multiple combination unlearning scenarios. Extensive experiments show that DMU significantly mitigates catastrophic forgetting, outperforming state-of-the-art methods of batch unlearning and sequential unlearning.

---

### TartanMatch: Towards Universal Dense Correspondence Across Modalities

Authors: Hyeokjoon Kwon, Jiting Cai, Ruogu Li, Kritan Bhandari, Geethika Hemkumar, Parv Maheshwari, Yuheng Qiu, Yuchen Zhang, Sebastian Scherer, Wenshan Wang
Published: 2026-10-10
Categories: cs.CV, cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2610.14513v1

Robotic platforms combine multi-modal sensors to operate across diverse conditions, but most dense correspondence methods are limited to RGB-RGB or a small set of modality pairs. Training a model for each modality pair scales poorly with the number of sensors and forgoes potential positive transfer across modality pairs. We introduce TartanMatch, a unified model for dense correspondence across all 25 ordered pairs among RGB, event, thermal, depth, and projected LiDAR. TartanMatch adapts an RGB-pretrained correspondence model to all modality pairs within one shared network using lightweight modality adaptation and a mixture of real and synthetic data. On zero-shot real-world benchmarks, TartanMatch reduces Endpoint Error (EPE) over the strongest baselines by 61.9% for cross-modal matching and 49.6% for same-modal matching. We further show that joint training across all modality pairs improves average matching accuracy over pair-specific training, especially on challenging cross-modal pairs. These results demonstrate that a single model achieves broad multi-modal coverage and state-of-the-art matching accuracy.

---

### Spatiotemporal Recognition of Abdominal Wall Layers During Optical Trocar Entry

Authors: Shreeram Athreya, Andrew Melehy, Carter Prentice, Michael Robinson, Corey Arnold, James Wu, William Speier
Published: 2026-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.14512v1

Objective. To evaluate whether spatial context and temporal modeling improve tissue recognition during optical trocar entry, where multiple abdominal-wall tissues may be visible simultaneously. Methods. We retrospectively analyzed 71 recordings with 1,516 annotated keyframes and 29,151 clinician-annotated sectors. Each frame was divided into 28 polar-grid sectors assigned to nine tissue classes. Color, neighboring-sector, and positional features were used by a gradient-boosted emission classifier. The full pipeline added learned temporal smoothing, spatial regularization, an anatomically ordered hidden Markov model, and posterior blending. Five-fold cross-validation was grouped by recording, with cross-fitting to prevent within-recording leakage. Primary evaluation used clinician-derived targets; ablations assessed classification, probability quality, spatial consistency, and peritoneum-or-deeper onset timing. Results. The complete pipeline achieved accuracy 0.646, macro-F1 0.507, and macro-AUROC 0.882, versus 0.600, 0.440, and 0.861 for emissions alone. Macro-F1 improved by 0.066 across all five folds. Label changes decreased from 66.2 to 19.5 per sector-minute, and median onset error decreased from 15.58 to 2.45 seconds. Frames with a tissue-depth-code span of at least three decreased from 53% to 26%, versus 22% for clinician annotations. Performance remained limited for preperitoneal fat and bowel, and the full model retained a 44% premature-onset proportion. Conclusions. Combining appearance, spatial context, and temporal constraints improved retrospective tissue recognition and onset localization. Tissue-specific limitations, residual timing errors, and reliance on future observations warrant external validation and prospective evaluation of causal models before clinical use.

---

### MontageFlow: Physically Valid High Fidelity EEG Generation

Authors: Sudarshan Regmi, Nabaraj Subedi, Arvind Pillai, Yu Yvonne Wu, Yuliang Chen, Michael V. Heinz, Lisa Marsch, Nicholas C. Jacobson, Andrew Campbell
Published: 2026-10-10
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.14507v1

Generative modeling offers a promising solution to the scarcity of labeled clinical electroencephalography (EEG) data under strict privacy constraints. Although recent generative models report substantial improvements in generation fidelity, we show that these gains conceal a categorical failure. EEG is not an arbitrary multichannel signal: each channel records the voltage difference between two scalp electrodes. So, real recordings satisfy exact linear identities among their channels. Current generators invariably violate these identities, yet this has gone unnoticed. We therefore propose MontageFlow, a flow-matching model that enforces these identities by restricting its output to signals that satisfy them. We find that ensuring validity improves fidelity too. Furthermore, to retain the EEG characteristics that oversmoothing washes out, we also condition the model on a physiological state vector encoded from the whole EEG window. For generation, this vector is drawn from a prior fitted in latent space. Across 3 large-scale benchmarks, our model improves TS-FID by 1.6-3.6$\times$ over the closest rival while ensuring validity.

---

### Rethinking Activation Functions: Reducing Bias Can Improve Learning

Authors: Zirui Peng, Heyi Zheng, Ziming Liu
Published: 2026-10-10
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.14500v1

Many widely used activation functions, including ReLU, GELU, Swish, and Sigmoid, produce predominantly positive outputs, resulting in a nonzero post-activation mean that we term activation bias. Despite its ubiquity, the effect of activation bias on learning dynamics remains poorly understood. Using controlled models, we show that activation bias accelerates initial loss reduction but is accompanied by overshooting, non-smooth trajectories, and a steeper loss landscape. This early advantage also coincides with the model's use of task-relevant information encoded in the activation bias component. In contrast, reducing activation bias slows initial learning but ultimately yields a lower final loss. Experiments on larger vision and language models exhibit a similar pattern: smaller activation bias is associated with better late-stage performance. Thus, this seemingly minor property can substantially redirect learning and warrants explicit consideration in future activation function design.

---

### DSR-Net: Learning Dynamic Skeletal Relations for 3D Human Pose Estimation

Authors: Fan Li, Jian Xiang, LinLin Xue, Yuwei Gao
Published: 2026-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.14499v1

3D human pose estimation requires accurate modeling of motion-dependent skeletal structures and inter-joint dependencies. Existing methods typically rely on fixed skeletal topologies in Graph Convolutional Networks (GCNs) or implicitly capture joint relations through attention mechanisms, limiting their adaptability to diverse motion patterns. To address this issue,we proposeDSR-Net, a dynamic structural representation learning framework for 3D human pose estimation. Specifically, a Dynamic Skeletal Relation (DSR) module is introduced to learn adaptive inter-joint dependencies beyond predefined skeletal connections. Based on the learned relations, a hierarchical relation-aware fusion strategy is developed to facilitate effective interaction between structural and temporal representations. In addition, a lightweight Residual Skeletal Recalibration Module (RSRM) is designed to enhance the stability of structure-aware features with minimal computational overhead. Extensive experiments on Human3.6M and MPI-INF-3DHP demonstrate that DSR-Net achieves MPJPEs of 37.9 mm and 14.9 mm, respectively, consistently improving over strong baselines. The results validate the effectiveness of dynamic skeletal relation modeling for robust 3D human pose estimation.

---

### Copula Active Subspaces II: Error Decomposition, A Posteriori Estimation, and Sharpness of the Bounds

Authors: Joshua Chen, Peter Jan van Leeuwen
Published: 2026-10-10
Categories: stat.ME, math.ST, stat.CO, stat.ML
arXiv: https://arxiv.org/abs/2610.14498v1

The Copula Active Subspaces (CAS) method of Part I estimates a non-Gaussian observation-noise density $π_{\boldsymbol{n}}$ from residual samples. Its KL divergence $\mathcal{D}_{\mathrm{KL}}(π_{\boldsymbol{n}}\,\|\,\widehatπ_{\boldsymbol{n}})$ admits an exact decomposition into marginal, projection and reduced-density terms and bounds the posterior KL divergence averaged over the true predictive distribution. We derive a priori bounds for these terms, together with estimates or bounds from residual samples. Weighted nearest-neighbour entropy estimates are used for the marginal and reduced-density terms, with asymptotic standard errors in exact rank coordinates. For the projection error, the score is fitted on one subsample, and its mean squared magnitude outside the retained subspace is estimated on the other. Corrections for score and rank-transform errors then give a theoretical upper bound with probability at least $1-δ$. Tests on three known noise laws in dimension $d=20$ use the estimated rank transform. In $11$ of $12$ combinations of example and sample size, the sum of the two estimated terms and the trace bound exceeds the reference divergence after averaging over the estimated models. In these tests, score-estimation error is the largest contribution to looseness. The overestimate from the Gaussian log-Sobolev inequality alone is at most a factor of $1.98$.

---

### DEFT: Disentanglement-Enhanced Fine-Tuning for EEG Foundation Models

Authors: Yuqiu Deng, Muyun Jiang, Zongpeng Zhang, Mingqing Xiao, Yansen Wang, Peiliang Gong, Xinliang Zhou, Yi Ding, Ziyu Jia, Cuntai Guan, Chenyu Liu
Published: 2026-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.14497v1

EEG foundation models (EFMs) have improved EEG representation learning by transferring knowledge from large-scale multi-subject pretraining to downstream tasks. However, their pretrained embeddings often retain substantial subject-related variation, placing an additional burden on the downstream classifier, especially in cross-subject settings. To address this issue, we propose Disentangled EFM Fine-Tuning (DEFT), a plug-and-play adapter for EFM fine-tuning. DEFT encourages branch disentanglement of pretrained tokens into task-oriented content and subject-consistent style, then performs downstream prediction from the content branch. It instantiates this factorize-before-classify design with recursive Gaussian refinement, temporal-variance regularization, subject-aware contrastive learning, and reconstruction. Experiments across multiple EEG datasets and four EFM backbones show that DEFT improves matched fine-tuning baselines. Further ablations and analyses support the intended content/style branch organization.

---

### Reels-RSI: Program-Level Self-Improvement for Code-Rendered Explainer Videos

Authors: Wei Lin
Published: 2026-10-10
Categories: cs.AI, cs.GR, cs.MM, cs.SE
arXiv: https://arxiv.org/abs/2610.14495v1

Explainer videos -- a worked geometry problem, why the sky is blue, how a dynasty fell -- are usually one more target for pixel video generation. We argue they are a different kind of artifact: a program (script, storyboard and per-frame animation code) that a deterministic renderer turns into frames. A program can be read, diffed, tested and audited; pixels can only be watched. Reels-RSI is an open-source reel director for any agent harness; its self-improvement happens at the program level rather than in weights or prompts. Every revision pass is logged with a code snapshot; a retrospective pairs each check finding with the diff that fixed it and proposes lessons; a lesson that a static check can catch is compiled into a rule with a must-fail and a must-pass example, so the memory is unit-tested; and a rule learned today runs over everything made before it. A bounded, benchmark-gated step lets an agent rewrite the director's playbook, kept only when blind pairwise judgments on held-out topics prefer it; nothing enters memory without a human decision. Across 111 reels -- a field study, two playbook evolutions, a 2x2 memory ablation, a weaker director, a 48-reel learning curve and a judge-reliability study -- we find that the rule learned in the field caught four latent defects in frames authored before it existed and none after; that rules are an audit for a strong director (eight unguarded reels, no violation) and a guard for a weaker one (thirty violations unguarded versus five); that lessons shorten revision loops but do not compound or move the judge; and that the first rewrite's 11--1 verdict was a position bias in our vote design (judges pick the second-shown reel 62--81% of the time), masking scripts that traded thirteen facts for craft. We document four failure modes of the preference signal, which check caught each, and the fixes. Code, rules, lessons and data are public.

---

### A Gated Catchment Hydrology Model (GCHM) for Physically-Interpretable Rainfall-Evapotranspiration-Runoff Simulation Across Diverse Hydroclimatic Regimes

Authors: M Jawad, HV Gupta, MA Farmani, A Behrangi, GY Niu
Published: 2026-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.14493v1

Recent advances in machine learning have improved rainfall-runoff prediction, yet most data-driven models remain difficult to interpret and do not explicitly enforce physical conservation laws. We present the Gated Catchment Hydrology Model (GCHM), a mass-conserving architecture based on the Mass-Conserving Perceptron (MCP). GCHM uses five storage states representing canopy interception, snow, soil moisture, groundwater, and streamflow generation. Constrained gates govern precipitation partitioning, evapotranspiration, infiltration, seepage, runoff, and baseflow while learning context-dependent process relationships. We evaluate GCHM across 50 CAMELS catchments spanning snowmelt-dominated, recent-rainfall, historical-rainfall, and arid regimes, using NOAA AORC meteorological forcing and remotely sensed vegetation and land-surface data. Performance is assessed using the Jawad-Kling-Gupta Efficiency (JKGE) metric, which accounts for stationary and non-stationary hydrologic behavior. We compare the full architecture with ablated variants and recurrent neural network benchmarks. GCHM reproduces streamflow dynamics across diverse hydroclimatic regimes, including arid catchments, while maintaining interpretable states and fluxes. Ablation experiments show that simplified architectures may perform adequately under wet conditions but struggle to represent flow persistence and recession during dry periods, when groundwater storage sustains low flows. Learned gates recover hydrologically plausible relationships governing snowmelt, runoff, seepage, and evapotranspiration. GCHM combines competitive predictive performance with mass conservation and interpretability, providing a foundation for hydrologic modeling and counterfactual analysis.

---

### Multi-modal Representation Fusion for Long-term Microsurgical Anastomosis Trajectory Anticipation

Authors: Yuxuan Liu, Yating Luo, Yunfei Luan, Xinyao Zhou, Yuyang Zhuge, Yao Guo, Guang-Zhong Yang
Published: 2026-10-10
Categories: cs.RO, cs.CV
arXiv: https://arxiv.org/abs/2610.14492v1

Microsurgical anastomosis is one of the most challenging and sophisticated procedures for surgical robotics, requiring accurate manipulation of surgical instruments within a limited space to achieve precise operations. Accurate surgical trajectory anticipation is essential for surgical scene understanding, intraoperative planning, robotic manipulation, and reliable behaviour analysis. However, existing approaches mainly focus on features extracted from surgical videos and historical trajectories, while overlooking rich semantic and motion information that can provide more discriminative multi-modal representations for complex surgical instrument-tissue interactions. In this work, we propose a cross-modal feature fusion framework for long-term 2D and 3D microsurgical anastomosis trajectory anticipation. The proposed framework consists of a multi-modal representation extractor and a cross-modal feature fusion network. Semantic recognition of surgical objects and consecutive motion estimation are incorporated as multi-modal visual cues, providing complementary spatial and temporal guidance for instrument-tissue interaction characterization. A coarse-to-fine progressive fusion strategy with a long-short-term memory mechanism is introduced to capture temporal dependencies step by step. The training and evaluation are conducted on our established robotic platform and a comprehensive dataset containing both phantom and ex-vivo surgical scenarios is constructed. Extensive quantitative and qualitative experiments demonstrate that the proposed framework consistently outperforms existing methods and exhibits improved robustness across different surgical scenes, highlighting its potential for both intelligent robot-assisted and autonomous surgical systems.

---

### HiPPT: Physiology-Grounded Wearable PPG Representation Learning via Multi-Scope Text Alignment

Authors: Jiaqi Wang, Wei Chen, Xiaoyu Li, Lexian Xia, ShuaiLong Tang, Chenran Zhang, Donglin Xie, Guangpu Zhu, Yelei Li, Yi Zhou
Published: 2026-10-10
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.14489v1

Reusable wearable PPG representations should capture local pulse morphology and broader physiological regularities. Existing pretraining objectives, however, usually apply supervision at the recording or pair level, overlooking that different concepts become reliable over different scopes of waveform evidence. We propose HiPPT, a physiology-grounded framework that matches each supervision source to its appropriate evidence scope. We construct a PPG-Text corpus of over 4 million wrist-worn recordings from 202K users, pairing recordings with waveform-derived morphology and metadata-derived context descriptions. HiPPT combines record-level morphology alignment, which grounds individual recordings in pulse characteristics, with cross-recording context alignment, which aggregates recordings sharing the same metadata tuple to capture context-consistent regularities. Waveform-guided text modeling further recovers masked contextual concepts from PPG patch representations, providing fine-grained waveform-text supervision. Across six downstream tasks spanning blood pressure, vascular stiffness, physiological attributes, and cross-dataset heart-rate estimation, HiPPT achieves the best result on the primary linear-probing metrics. The aligned space also supports zero-shot profiling of age group, sex, and BMI category, individually and jointly, with performance improving as more recordings are aggregated. These results show that evidence-scope matching yields transferable and semantically accessible wearable PPG representations. Code and pretrained weights will be released.

---

### AlignPoison: Conditional Targeted Poisoning of Generative Recommenders via Decoding-Aligned Prefixes

Authors: Zhuoxuan Zhang, Jinhao Duan, Xinyu Zhao, Jinduo Guo, Shuai Shao, Rui Li, Luke Simon, Zhun Deng, Tianlong Chen, Bo Sun
Published: 2026-10-10
Categories: cs.AI
arXiv: https://arxiv.org/abs/2610.14487v1

Generative recommendation models based on autoregressive decoding have been adopted as alternatives to classical ranking-based systems. While their modeling flexibility has been widely studied, their resilience to data poisoning remains largely unexplored. In this work, we ask: how susceptible are generative recommenders to conditional targeted poisoning? Under a realistic setting where an adversary can inject limited malicious interactions and query a recommendation API as a black box, our empirical study shows that generative recommenders are vulnerable: injecting fewer than 1% synthetic interactions significantly increases the likelihood of promoting malicious targets under the trigger. To characterize this vulnerability, we analyze the synthesis of malicious interactions from two perspectives: (i) decoding alignment and (ii) prefix popularity. We observe that short, context-aligned prefixes that are both common under the model's induced transition dynamics and strongly linked to a frequent trigger consistently amplify targeted promotion. Motivated by this analysis, we develop AlignPoison, a black-box framework that discovers and samples context-aligned prefixes from API outputs. We conduct experiments on benchmarks in unimodal and multimodal generative recommendation settings. Results show that these prefixes outperform competitive baselines, improving ASR lift by up to 16.3 percentage points and highlighting a decoding-dependent, context-sensitive poisoning vulnerability in generative recommendation models.

---

### AphaVoice: Text-to-Speech Model for Aphasia Patient Simulation

Authors: Al Varrel Putra Kusuma, Fernando Bello, Joshua Brown
Published: 2026-10-10
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2610.14479v1

While language therapy is the primary treatment for post-stroke aphasia, limited clinical resources often restrict patient access to proper care. Addressing this gap requires reliable training simulations for medical professionals. However, current AI efforts focus almost exclusively on speech processing, leaving speech generation largely unexplored. This study introduces AphaVoice, a novel text-to-speech (TTS) framework explicitly engineered to synthesize highly realistic aphasic speech. Evaluated against state-of-the-art TTS systems, AphaVoice demonstrates a superior ability to capture the clinical feature of dysfluency, outperforming existing models in temporal metrics while consistently maintaining speaker identity. By leveraging latent space shifting, the model also supports multi-speaker generation without sacrificing phonetic realism. Ultimately, AphaVoice provides a scalable, high-fidelity tool for medical training and clinical simulation related to aphasia.

---

### What Counts as the Same Task? Executable Contracts for Robustness Evaluation of Tool-Using Agents

Authors: Shaikh Akib Shahriyar, MD. Akil Raihan Iftee, Armon Barton, Matthew Wright
Published: 2026-10-10
Categories: cs.SE, cs.AI, cs.CR, cs.MA
arXiv: https://arxiv.org/abs/2610.14477v1

Robustness tests for tool-using agents often change the agent's environment, for example by hiding a hostile note in an invoice, and then judge the agent by what it does. The result tells us about the original task only if the changed test case is still a valid test of that task. Many evaluations decide this with weak rules, such as accepting any well-formed edit or any edit that stays close to the original in embedding space. We propose executable task contracts. A contract states which facts define a task, which edits are allowed, and what finishing the task means, and it checks each changed test case before the agent runs. We compare validity rules by scoring the same 14,336 stored agent runs under each rule. Across four encoders, embedding thresholds allowed 75-100% of edits that changed the task, such as swapping the payee, yet rejected 24-45% of valid hostile notes. With a MiniLM cosine threshold, Mistral appeared vulnerable on all 64 tasks, but only 26 had a successful valid attack. Replaying a saved solution and allowing edits to benchmark fields fail in similar ways. Evaluations should state, and check, what counts as the same task.

---

### Scale, Drivers, and Limits of Representational Alignment Across Large Language Models

Authors: Jialin Wu, Meenakshi Khosla
Published: 2026-10-10
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2610.14476v1

Comparing large language models can reveal shared representational structure and systematic differences between them. However, aggregate similarity measures can obscure how agreement varies across geometric scales and which linguistic properties support it. We introduce two complementary methods to resolve where models agree and what supports that agreement. Standard representational similarity analysis (RSA) summarizes similarity across all stimulus pairs, potentially obscuring differences in alignment between nearby and distant inputs. Neighborhood-Weighted RSA (NW-RSA) addresses this limitation by weighting stimulus pairs according to their proximity in representational space, placing alignment from local neighborhoods to global geometry on a common scale. Counterfactual RSA (CF-RSA) tests which linguistic properties support alignment by comparing intact representations in one model with representations under controlled linguistic perturbations in another. Across diverse language models, we find strong agreement in coarse representational geometry but substantial divergence in local neighborhoods. Alignment is observed both across broadly sampled natural text and within linguistic equivalence classes that preserve particular aspects of meaning or structure. Counterfactual analyses reveal selective sensitivity: changes to logical scope reduce alignment substantially more than the tested changes to hierarchical syntax, event roles, or discourse structure. Alignment also decreases when sentence structure or lexical meaning is disrupted in word lists and jabberwocky sentences, suggesting that the regularities of natural language support stronger convergence. These findings refine claims of representational universality by showing that strong overall similarity need not imply agreement in how models represent finer linguistic distinctions.

---

### EdgeLLM-Nano: Auditing Polar-Guided Key-Cache Compression and Consistency-Anchored Pruning on a 33.6 MB On-Device Language Model

Authors: S M Mahbub Murshed, Hayden Bright, Rezaul Karim, Ali Shiri Sichani
Published: 2026-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2610.14475v1

Deploying language models (LMs) on the edge presents unique memory, latency, and power challenges absent in high-performance computing. KV-cache compression emerges as a promising approach to address these challenges. However, the effectiveness of current KV-cache compression methods has not been thoroughly examined on edge devices. Moreover, we show that existing evaluation methods do not always involve the KV-cache under test. We propose EdgeLLM-Nano, a measurement study of KV-cache compression using a nano-scale decoder-only LM (transformer) testbed that has 8.39M parameters, i.e., 33.6 MB, to enable end-to-end examination of each backend, seed, and power sample. Our results give a negative verdict for a polar-guided key-cache quantizer (round-to-nearest (RTN) control causes an order of magnitude less damage), as well as a consistency-anchored progressive weight pruner (one-shot magnitude-based pruning performs as well as, if not better). We also introduce Long-Context Decode Fidelity (LCDF) evaluation, a token-by-token examination of teacher-forced log-losses using the KV-cache under test. Unlike perplexity that gives the same 5.461 score to all methods, LCDF distinguishes different KV-cache compression backends thanks to its paired statistics across seeds and effect sizes. Our companion Raspberry Pi 5 measurement shows all compression backends we tested to be less energy-efficient per token compared to uncompressed FP16 inference. Except for KIVI that is close to lossless in terms of LCDF, our findings indicate KV-cache compression can be improved. LCDF offers a simple and effective way of shedding light on KV-cache compression methods, whose negative impacts have been overlooked by perplexity.

---

### RES-MR: Risk-Aware Reasoning for Explainable and Safe Medication Recommendation

Authors: Cong Wang, Jin Li, Shoujin Wang, Yishuo Li, Huilin Gu, Wenpeng Lu
Published: 2026-10-10
Categories: cs.LG, cs.AI, cs.IR
arXiv: https://arxiv.org/abs/2610.14474v1

Medication recommendations seek to deliver personalized drug combinations based on patients' clinical information, where ensuring transparent decision-making and personalized safety are of paramount importance. However, existing approaches predominantly operate as ``black boxes'', offering limited interpretability and hindering clinician trust. Furthermore, they typically rely on fixed, patient-agnostic safety constraints, neglecting the heterogeneity of risk tolerance across individuals, which poses significant risks to vulnerable populations. To address these gaps, we propose \textbf{RES-MR}, a novel Risk-aware Reasoning framework for Explainable and Safe Medication Recommendation using LLMs. RES-MR follows a two-stage training paradigm: (1) an explainable clinical reasoning distillation stage to elicit diagnostic capabilities by fine-tuning on synthesized reasoning trajectories, grounded in patient-specific knowledge from graph retrieval; and (2) a risk-aware policy optimization stage to dynamically balance therapeutic efficacy with personalized safety. This stage employs disentangled drug factors to construct patient-specific reference prototypes that capture desired therapeutic drug features and risk features to avoid. These prototypes enable risk boundary calibration and guide the model toward safe recommendations via reward shaping and clipping. Extensive experiments on benchmark datasets demonstrate that RES-MR significantly outperforms state-of-the-art baselines in terms of accuracy, safety, and explainability. Our code and data are available at: https://github.com/wangcong2001/RES-MR.

---

### Diffusion-Based Auctions for Multi-Winner Banner Ads

Authors: Lilian Sun, Henry Huang, Fucheng Warren Zhu, Giannis Daras, Constantinos Daskalakis
Published: 2026-10-10
Categories: cs.GT, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2610.14473v1

We propose a new framework for banner ad auctions, leveraging diffusion models to allow for multiple bidders to co-win a single banner ad opportunity, and in particular influence the ad shown on the banner depending on the strength of their bids, enabling a far richer outcome space compared to standard banner ad auctions, where there is a single winner who shows their ad. In our framework, the outcomes are images (what ad to show) and each bidder expresses their preferences over outcomes by declaring their value per unit of alignment between the image chosen by the mechanism and a textual description of their desired image. Our proposed mechanism combines the advertisers' bids and textual descriptions into a score that drives the reverse process of a diffusion model. This is used to sample a set of candidate images from which the welfare maximizing one is eventually chosen. Extensive experimentation evaluates our proposed mechanism around the following axes: (1) efficiency improvements of our allocation rule relative to the single-winner allocation rule; (2) monotonicity of our allocation rule; and (3) approximate incentive compatibility when combining our allocation rule with Myerson lemma-based or Clarke rule-based payments. Improvements along all these axes are shown with the number of Monte Carlo samples. Meanwhile, these benefits are attained while preserving the image quality.

---

### Co-adaptive Rule Routing and Constraining for Structured Knowledge Retrieval

Authors: Zhisheng Qi, Haoyu Han, Jiaxi Li, Bing He, Huiyuan Chen, Franck Dernoncourt, Ryan A. Rossi, Mahantesh M. Halappanavar, Jin Lu, Yu Zhang, Tianming Liu, Yu Wang
Published: 2026-10-10
Categories: cs.IR, cs.AI
arXiv: https://arxiv.org/abs/2610.14465v1

Structured knowledge retrieval acquires task-relevant evidence from structured knowledge sources for downstream tasks. Existing methods rely on unconstrained retrieval or predefined query-rule supervision, such as author-paper rules for scientific discovery and anatomy-drug rules for precision medicine. Such supervision is rarely available in practice. We propose Co-Adaptive Rule Routing and Constraining Retrieval (CoRE), which jointly learns which reasoning rule to assign to each query and how to retrieve evidence under that rule, without predefined query-rule pairs. A routing agent selects a rule, and its corresponding retriever traverses the knowledge base and ranks evidence that satisfies the rule's constraints. Because routing and retrieval are interdependent, CoRE alternates between reinforcement learning for the routing policy and contrastive learning for the rule-constrained retriever. The selected rules provide supervision for retriever training, while retrieval performance rewards the routing policy. This closed-loop process enables both components to co-adapt without explicit query-rule supervision. Experiments on STaRK-QA datasets across scientific discovery, biomedical reasoning, and e-commerce knowledge management demonstrate CoRE's effectiveness. Ablations show the benefit of co-adapting the router and retriever. Generation-based routing transfers across datasets and benefits from multi-domain pretraining followed by target-specific fine-tuning. Analysis of rule usage across queries also informs automatic rule refinement and discovery.

---

### Evaluating Bias in Chinese Large Language Models across Tasks and Registers: Task Operationalization Effects and Irony Classification Risks

Authors: Mo Zhou
Published: 2026-10-10
Categories: cs.CL, cs.CY
arXiv: https://arxiv.org/abs/2610.14464v1

In Chinese content governance, large language models are increasingly used to identify toxic and discriminatory expressions, yet both task operationalization and ironic expression can alter bias measurement readings. We conducted a preregistered multi-task behavioral audit of multiple Chinese large language models. In the content classification experiment, the core panel's classification error rate rose from 10.9% in the formal register to 23.5% in the yinyang (an echoic-irony register) version, a risk difference of +12.7 percentage points. (95% CI (+9.7, +15.7)). Inter-rater agreement was lower in the yinyang register (majority-vote kappa = 0.299), so the increase contains a component of label uncertainty propagation, but the direction of increase was consistent under both adjudicated and majority-vote reference standards. Explicit irony-aware prompts reduced the intervention panel's error rate by 9.6 percentage points. We operationalized the 29 shared proposition anchors into three task forms: social consensus estimation (Form A), high-stakes decision (Form B), and controlled association rating (Form C); independent manipulation checks found that all 58 A-C comparison units received unanimous agreement, whereas the low A-B convergence was confirmed by calibrated majority adjudication (87.8% supporting the minority judgment); on the irony-probe materials, the interactions of B and C relative to A were +0.577 and +1.121, respectively. Thus, when evaluating large language models for Chinese content governance, the effect of task operationalization on bias readings is an empirical question that requires data to answer, and safety evaluation should incorporate the ironic register as a test condition.

---

### Re-executing Coding-Agent Evaluations: Unchanged Decisions and Unresolved Policy Risk

Authors: Tianyu Ding, Yaocheng Tong, Muzhi Yang, Fei Jiang, Lei Yang, Boya Yu
Published: 2026-10-10
Categories: cs.SE, cs.CL
arXiv: https://arxiv.org/abs/2610.14460v1

Executable tests grade large language model (LLM) coding-agent patches for leaderboards, rewards and self-improvement. Two preregistered re-execution studies returned negative results. SWE-bench Pro v1's offline option rejected none of 59 accepted Go patches whose grading configuration changed. Their exploratory, unweighted upper confidence limit is 4.95% under candidate independence, leaving a 3% exposed-patch risk unresolved; the preregistered estimator uses inverse sampling-probability weights. An extension was inconclusive. In a selected stratum, a reconstructed module cache built from the reference solution's dependencies rejected accepted alternatives from four models that passed with their own dependencies. Re-executing 1,929 selected agent-version/task evaluations from the Darwin Gödel Machine's archive preserved best-agent selection and system-variant ranking, with search history and scores of evaluations not rerun fixed. Exploratory host, image and adapted-grader comparisons also preserved them. Seven of 23 disagreements with archived verdicts did not recur on a second host; rebuilt images changed 16 three-run majority verdicts (0.83%) under four-CPU execution. We provide an evidence-to-claim summary, lessons for grader audits and an anonymous replication package. These results concern accepted patches and fixed-history decisions; general policy safety and adaptive-learning robustness require broader evaluations.

---

### Learning to Extrapolate in Depth with Looped Transformers

Authors: Huitao Yang, Jianhao Huang, Zixuan Wang, Baharan Mirzasoleiman
Published: 2026-10-10
Categories: cs.LG
arXiv: https://arxiv.org/abs/2610.14456v1

Looped transformers reuse a single block across iterations, raising a natural question: how far can a block trained at one depth be safely extrapolated to greater depths? We study this question for all-pairs graph connectivity, where harder graphs require longer-range information propagation. We show that under exchangeability and symmetric positive initialization, population training at depth K converges to an update rule governed by a single effective parameter that balances graph propagation against fitting structure unresolved at training depth. This trade-off creates a finite extrapolation window: too few test-time loops provide insufficient propagation, while too many eventually create false connections. We derive explicit endpoints for this window and bound the graph hardness solvable beyond training as a function of training depth and hardness. In particular, a test depth chosen solely from the training distribution can guarantee recovery on substantially harder graphs, with a certified depth gain that grows exponentially as training depth outpaces training hardness. Experiments validate the predicted windows and show how compute budget and initialization affect the optimal training depth.

---

### Jevolution: A Framework for Studying Communication and Collective Dynamics in AI Agents

Authors: Xirui Huang, Nikhil Hooda, Utkarsh Gupta
Published: 2026-10-10
Categories: cs.MA, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2610.14449v1

Experiments with remote decision-making agents depend on more than the policy: the world may change during inference, the simulator determines which actions are available, and births and immigration alter who can act. We present Jevolution, an instrumented spatial predator-prey framework that makes these dependencies explicit. Its reference implementation separates the evolving world, asynchronous request scheduling, controller adapters, and experiment records. The model combines local observations, energy and resource constraints, predefined signals, and inherited scalar traits. We describe these mechanisms using the ODD protocol and distinguish implemented controls from research questions requiring additional interventions. Free synthetic fixtures exercise scheduling, signaling, and inheritance against the pinned implementation; they validate specified software behavior rather than model intelligence or ecological realism. A completed communication case study illustrates the end-to-end workflow using Jev to control both species. Across 6 complete habitat-seed pairs, the mean signals-on minus signals-off difference in normalized rabbit abundance was -0.0095 (nominal 95% pair-bootstrap interval (-0.0772, 0.0602)). This small study does not resolve the direction of the signal-system effect. The contribution is an inspectable experimental system and its evidence boundaries, not a new policy-learning algorithm or a demonstration of emergent language, adaptive selection, or biological predictive validity.

---

### KV4-RL: Unlocking Faster Asynchronous Rollouts for RLVR with INT4 KV Cache Quantization

Authors: Sihwan Park, Sung-Yub Kim, Doohyuk Jang, Eunho Yang
Published: 2026-10-10
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.14448v1

Asynchronous rollout improves the efficiency of reinforcement learning with verifiable rewards (RLVR) by reducing under-utilization during long-tail generation. Sustained concurrency also concentrates context processing into fewer decode steps, increasing the KV workload per step. We characterize this \emph{KV concentration} and find that KV cache quantization delivers substantially greater acceleration than weight-only and weight-activation quantization in our matched-workload comparisons. However, INT4 KV rollout can degrade learning even with token-level truncated importance sampling (TIS). We propose \emph{KV4-RL}, which combines calibration-free INT4 KV quantization with selective INT8 protection of attention sinks and recent context. Protecting a bounded number of positions reduces sampler mismatch while retaining most of the speedup from INT4. Across 400 iterations of Qwen3-8B-Base GRPO training, KV4-RL achieves $1.80\times$ cumulative rollout speedup and $1.27\times$ cumulative full-iteration speedup over BF16 rollout, reaching $1.97\times$ rollout speedup over the final 50 iterations. Under a common BF16 evaluator, its final six-benchmark mean accuracy is within 0.5 percentage points of BF16-rollout training for both Qwen3-4B-Base and Qwen3-8B-Base.

---

### FreeCHROMA: Generative Chroma Keying from Unaligned Background References

Authors: Jiayi Wu, Tianfu Wang, Rui Zhao, Florian Schiffers, Zhimin Chen, Cornelia Fermuller, Yezhou Yang, Christopher Metzler, Yiannis Aloimonos
Published: 2026-10-10
Categories: cs.CV
arXiv: https://arxiv.org/abs/2610.14445v1

In film production, chroma keying composites an approved foreground performance into a new environment, but conventional workflows require coordinated background plates and dedicated harmonization. Recent generative methods often require pre-aligned backgrounds, focus on background-conditioned foreground relighting, or represent the environment only through an edited first frame. These restrictions limit background selection, scene composition, and camera motion while adding preparation or editing effort. We present FreeCHROMA, which adapts a pretrained video diffusion model in a parameter-efficient manner for generative video compositing from one to three unaligned background references. The model integrates the foreground into the referenced environment, preserving foreground identity, motion, and fine visual details while generating a spatiotemporally coherent background. Foreground-background harmonization occurs within the same generation process, without a separate relighting stage. Optional text instructions help disambiguate complex spatial relationships and camera behavior. To train the model, we construct pseudo-paired data from readily available real monocular videos through foreground-background separation and augmentation with existing foundation models. The original videos serve as training targets, while automated processing and quality verification enable scalable data construction. Across reference fidelity, foreground preservation, scene integration, harmonization, and overall video quality, FreeCHROMA consistently outperforms recent compositing and conditional video-generation baselines, with particularly clear gains under camera motion. FreeCHROMA relaxes the constraints of existing tools to enable end-to-end generative chroma keying from unaligned inputs, with exploratory experiments demonstrating its potential in robotics.

---

### Specification-Based Testing of Large Language Models

Authors: Kausar Moshood, Jace Bolante, Roshan Aditya Murali, Manish Motwani
Published: 2026-10-10
Categories: cs.SE, cs.AI
arXiv: https://arxiv.org/abs/2610.14443v1

Model cards, system cards, and safety reports increasingly document LLM safety behaviors, yet safety evaluations typically start from predefined benchmarks rather than asking which documented claims those benchmarks actually test. We introduce LEST (LLM Evaluation through Specification-driven Testing), a framework that treats documented safety claims as specifications for black-box testing. In LEST, each claim determines which benchmark prompt tests it and whether a Refuse-to-Comply transition, where a model refuses a prompt but complies with its mutated version, counts as evidence against it. LEST mutates these prompts to search for counterexamples, manually validates each candidate, and traces validated counterexamples back to the claims they contradict. We evaluate LEST using the system cards of Claude Opus 4.8 and GPT-5.5 with 97 seed prompts from SORRY-Bench across Mental Health and Self-Harm, Harmful and Violent Content, and Fairness and Bias. Of 56 identified safety claims, 29 (51.8%) are prompt-testable, but only 17 (58.6%) are exercised by an applicable benchmark prompt. Both models refuse >95% of benchmark prompts. Evaluating 1,836 mutated prompts on both models reveals 312 candidate Refuse-to-Comply transitions flagged by automated response classifiers. However, human validation confirms only 19 (6.1%) as both intent-preserving and genuinely compliant, showing automated evaluation substantially overestimates behavioral failures. Tracing these 19 transitions back to the claims reveals counterexamples for 6 of 9 covered claims for which refusal is an applicable oracle. Our results show that benchmark prompts may not be sufficient to validate all documented LLM safety behavior, and automated failure detection requires manual validation. LEST provides end-to-end traceability from natural-language safety documentation to executable tests and human-validated counterexamples.

---

### Evaluating Semantic Relation Knowledge through Bidirectional Relation Prediction from Gloss Pairs

Authors: Zhihan Cao, Adam Nohejl, Hitomi Yanaka
Published: 2026-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2610.14442v1

Semantic relations organise word senses and are fundamental to lexical semantics. In order to evaluate semantic relation knowledge of large language models (LLMs), we propose a challenging task in which an LLM predicts the relation between two glosses (definitions of senses) presented in two different orders. The predictions for the two orders are evaluated for consistency with the symmetry or asymmetry of semantic relations. Our task is designed for both monolingual and cross-lingual evaluation, allowing us to examine whether relation knowledge remains accessible when the same senses are verbalised through glosses in different languages. We construct a dataset of 3,069 sense pairs, with each sense verbalised in English, Greek, Romanian, Turkish, and Japanese. Experiments with 17 LLMs in zero-shot and five-shot settings reveal generally poor performance. In each setting, only eight LLMs achieve a macro-averaged F1 score significantly above a random baseline, and only one (zero-shot) and two (five-shot) significantly outperform a neural baseline. In particular, LLMs often fail to generate bidirectionally consistent predictions for hypernymy and to make correct predictions for holonymy in either direction. Our results suggest limitations in current LLMs' semantic relation knowledge, both in identifying relations and in respecting asymmetry.

---

### Target-ESS Tempered Policy Optimization (TETPO): Smooth Variance Control via Bridge Importance Sampling

Authors: Shayan Mohajer Hamidi, Kayhan Behdin, Wenhui Zhu, Aida Rahmattalabi, Jincheng Cao, Jelena Markovic-Voronov, Alborz Geramifard
Published: 2026-10-10
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2610.14440v1

Asynchronous reinforcement-learning post-training of large language models is computationally efficient but inherently off-policy: the learner updates on rollouts generated by a stale policy, while correcting this mismatch requires importance weights that can be extremely heavy-tailed. A small number of trajectories may then dominate an update, collapsing the effective sample size (ESS) and increasing the variance and instability of the resulting gradient estimates. Existing stabilizers apply fixed interventions regardless of the realized severity of weight concentration, either discarding selected samples or reweighting them without adapting the correction strength. We introduce \emph{Target--ESS Tempered Policy Optimization} (TETPO), which replaces each importance weight $ρ$ by $ρ^α$ for a data-adaptive exponent $α\in(0,1)$: $α=1$ preserves the original importance weights, whereas $α=0$ assigns uniform coefficients. For each prompt group, TETPO selects the largest $α$ whose tempered weights satisfy a target ESS fraction $κ$, applying little tempering to healthy groups and stronger tempering when weight degeneracy is severe. At the population level, the tempered coefficients are exact importance weights for a geometric bridge between the sampler and learner policies. The exponent is efficiently found by one-dimensional bisection, while the resulting weights satisfy deterministic bounds on normalized-weight dispersion and weighted-gradient norm that are independent of policy drift, together with a fixed-$α$ bound on bridge-target discrepancy. Power tempering also introduces no explicit zero-weight mask. On mathematical-reasoning benchmarks under high staleness, TETPO outperforms strong on- and off-policy baselines.

---

### SceneForesee: Learning Scene Evolution for Spatiotemporal Reasoning

Authors: Chaoyue Li, Shoudong Han, Haoyang Wu, Boxue Yang, Sijia Chen, Yufan Chen, Ruping Zhang
Published: 2026-10-10
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2610.14439v1

Large multimodal models (LMMs) need to understand how entities move and interact as a scene develops. This requires following the same entities over time and relating their observed motion to subsequent changes. We formulate Scene Evolution Understanding (SEU), a task for predicting and interpreting this change. Given observed video, a model estimates entity histories and predicts one scene continuation through future trajectories and Events, accompanied by Visual Evidence from observed motion and a scene description. We construct SceneForesee, a benchmark for evaluating entity continuity and scene evolution reasoning through paired observed and future scenes with shared entity identities. It contains 144.57K temporal pairs, 1.05M entity correspondences, and 272.49K Events paired with observed evidence. We propose a spatiotemporal reasoning framework, EvoSee, that learns future geometry and Event semantics through shared candidate states and joint assignment. It selects one continuation per entity from observation and uses the selected changes for entity refinement, retention, and scene representation. On SceneForesee, EvoSee outperforms Qwen3.5-9B by 10.45 points in Combined Point-F1 and 17.24 points in Event Global F1. On the 4D reasoning benchmark Dyn-Bench, it improves question answering accuracy and grounding J&F by 4.8 and 3.58 points over their respective base systems. These results support scene evolution as both a prediction target and a dynamic representation that brings predicted entity changes into spatiotemporal reasoning.The project is available at https://github.com/mikubaka88/SceneForesee-Spatiotemporal-Reasoning.

---

### Neither True nor False: Do LLMs Represent Truth-Value Gaps in Plural Predication?

Authors: Duaa Nakshbandi, Zhiyu An, Yiping Li, Sachin Giri, Wan Du
Published: 2026-10-10
Categories: cs.CL
arXiv: https://arxiv.org/abs/2610.14438v2

Large Language Models (LLMs) are increasingly being used to summarize and verify claims in critical sources that inform consequential decisions. Such claims are full of plural definites. If three of five patients improved after taking drug A, the claim the patients improved after drug A is not true, since two did not, yet its negation, the patients didn't improve after drug A, is not true either, since three did. In formal semantics, this truth-value gap is known as the Homogeneity Gap: both the plural definite and its negation are neither true nor false when the property holds for only a fraction of the plural referent. A model that cannot represent this gap turns mixed evidence into a clean verdict, an error that can be costly in high-stakes use. We introduce HomogeneityNLI, a diagnostic dataset of 1,000 minimal pair items crossing determiner, polarity, and group composition across ten logical conditions. Evaluating seven open-weight models, we uncover a pervasive Bivalence Bias: models assign plural definites their homogeneous truth conditions. In the zero-shot setting, no model recognizes the gap for both polarities in more than 15% of scenarios, and negated definites are judged False in 84% of cases. Explicit linguistic rules, chain-of-thought, and enumeration prompts produce unstable, model-specific effects that often entrench the bias. Our findings indicate that current open-weight LLMs acquire the truth conditions of plural predication but not its trivalent logic, and cannot yet be trusted to summarize or verify claims whose evidence is mixed.

---