本篇博文主要内容为 2026-08-11 从Arxiv.org论文网站获取的最新论文列表，自动更新，按照NLP、CV、ML、AI、IR、MA六个大方向区分。
说明：每日论文数据从Arxiv.org获取，每天早上12:30左右定时自动更新。
提示: 当天未及时更新，有可能是Arxiv当日未有新的论文发布，也有可能是脚本出错。尽可能会在当天修复。
目录
概览 (2026-08-11)
今日共更新250篇论文,其中:
- 自然语言处理共31篇(Computation and Language (cs.CL))
- 人工智能共81篇(Artificial Intelligence (cs.AI))
- 计算机视觉共78篇(Computer Vision and Pattern Recognition (cs.CV))
- 机器学习共48篇(Machine Learning (cs.LG))
- 多智能体系统共5篇(Multiagent Systems (cs.MA))
- 信息检索共9篇(Information Retrieval (cs.IR))
- 人机交互共7篇(Human-Computer Interaction (cs.HC))
多智能体系统
[MA-0] Algorithmics for Safe Bicycle Network Design with Bounded Detours in Rural Areas
【速读】：该论文旨在解决在农村道路网络升级中构建安全自行车网络（Safe Bicycle Network with Bounded Detours, SBNBD）的问题，即在给定预算和允许的最大绕行因子 α 的条件下，如何选择性地升级不安全路段，使得每对终端节点之间均存在一条安全路径，其长度不超过原网络中最短路径距离的 α 倍。该问题的核心挑战在于在有限预算下平衡升级成本与路径可接受的绕行程度。解决方案的关键在于引入并利用图的结构性参数——反馈边数（feedback edge set number, fes）与终端对数量（p），通过构造一个等价实例，将原问题规模压缩至 $ O(\mathrm{fes} + p) $ 个顶点和边，从而实现关于 $ \mathrm{fes} + p $ 的固定参数可解性（fixed-parameter tractability）。这一结构化压缩不仅为算法设计提供了理论基础，还支持了基于整数线性规划（ILP）的精确求解方法，并在真实世界道路网络数据集上的实验验证了其有效性，揭示了绕行因子 α 与升级成本之间的权衡关系。
链接: https://arxiv.org/abs/2608.09472
作者: Till Fluschnik
机构: Till Fluschnik
类目: Data Structures and Algorithms (cs.DS); Multiagent Systems (cs.MA)
备注:
Abstract:We introduce the \emphSafe Bicycle Network with Bounded Detours (\emphSBNBD) problem, motivated by upgrading rural road networks for bicycle traffic. Given an undirected graph with safe and unsafe edges, edge lengths, upgrade costs, terminal pairs, a budget, and a detour factor \alpha , the task is to upgrade unsafe edges so that each terminal pair is connected by a safe path of length at most \alpha times its shortest-path distance in the original network. We study SBNBD from a parameterized perspective. We prove strong NP-hardness on restricted graph classes, including planar graphs of treewidth two, graphs with feedback vertex set number one, and graphs of maximum degree three, and complement these lower bounds with polynomial-time algorithms for trees and graphs of maximum degree two. We show fixed-parameter tractability for the number of unsafe edges and prove matching SETH-based lower bounds, a polynomial-kernel lower bound, and W-hardness for natural parameters. Our main structural result maps any instance to an equivalent instance with O(\mathrmfes+p) vertices and edges, where \mathrmfes is the feedback edge number and p the number of terminal pairs; this yields fixed-parameter tractability for \mathrmfes+p . Finally, we evaluate ILP-based algorithms on OpenStreetMap road networks for small German municipalities and their surroundings. The instances have small treewidth upper bounds and moderate feedback edge structure. Preprocessing based on the \mathrmfes+p reduction and tree-decomposition-based cut generation both improve exact solving, especially on harder instances. Experiments with different detour factors show that increasing \alpha can reduce the upgraded-edge length, revealing trade-offs between upgrade cost and allowed relative detours. Overall, structural graph parameters provide a useful algorithmic lens for safe bicycle-network design. Subjects: Data Structures and Algorithms (cs.DS); Multiagent Systems (cs.MA) Cite as: arXiv:2608.09472 [cs.DS] (or arXiv:2608.09472v1 [cs.DS] for this version) https://doi.org/10.48550/arXiv.2608.09472 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[MA-1] LAVE: Latent Visual Evidence-Enhanced Planning for Video Tool-use Agents
【速读】：该论文旨在解决长视频理解中视频工具使用智能体（video tool-use agents）因依赖文本化观测而导致的**工具观测瓶颈（Tool observation bottleneck）**问题，即在多步规划过程中，由于仅通过文本接口传递视觉工具计算结果，导致未被语言描述的潜在视觉证据被丢弃，无法在后续决策中复用。其核心解决方案是提出一种无需训练的框架——潜在视觉证据增强规划（Latent Visual Evidence-Enhanced Planning, LAVE），关键在于引入双通道观测接口：可见通道保留原始文本轨迹，潜在通道则存储已完成工具调用所产生的预表述视觉更新信息，包括工具角色、源帧时间戳和视觉位置等元数据。在规划阶段，LAVE 通过基于时间戳对齐的有限更新与熵约束的帧时路由机制，检索并融合当前规划状态所需但未被文本覆盖的潜在视觉证据，从而实现对已有视觉计算成果的高效复用。该方法无需额外训练、帧重播或修改原有调度逻辑，在 Video-MME、LongVideoBench 和 CG-Bench 等基准上均显著提升性能，尤其在相同帧预算下使 Video-MME 总分提升 3.76 分，验证了潜在视觉证据复用在多步视频智能体规划中的有效性。
链接: https://arxiv.org/abs/2608.07585
作者: Zijian Wang,Junnan Zhu,Rongzhen Li,Xiao Liu,Guohui Xiang,Quan Lu,Lijia Liu,Yining Wang,Jiang Zhong,Kaiwen Wei
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG); Multiagent Systems (cs.MA)
备注: 16 pages, 6 figures, 9 tables. Includes appendix
Abstract:Long-video understanding requires models to efficiently acquire and reuse sparse visual evidence from long and redundant video streams. Recent video tool-use agents address this challenge by iteratively invoking visual Tools at different temporal scales, but their Tool-Planner communication typically relies on textual observations. Such text-only interfaces provide lossy summaries of Tool computations, causing previously computed visual evidence not verbalized to be discarded and unavailable for subsequent planning. We identify this limitation as the Tool observation bottleneck and propose Latent Visual Evidence-Enhanced Planning (LAVE), a training-free framework for reusing latent visual evidence from completed Tool calls. LAVE introduces a dual-channel observation interface: the visible channel preserves the original textual trajectory, while the latent channel stores pre-verbal visual updates with their Tool roles, source-frame timestamps, and visual locations. During planning, LAVE retrieves evidence relevant to the current Planner state but not covered by textual observations, and integrates it through bounded timestamp-aligned latent updates with entropy-constrained frame-time routing. This enables video agents to reuse existing visual computation without additional training, frame replay, or modifications to the original orchestration. Extensive experiments on Video-MME, LongVideoBench, and CG-Bench show that LAVE consistently improves video tool-use agents across backbones. Under a comparable frame budget, LAVE improves the Video-MME overall score by 3.76 points over the strongest baseline, demonstrating the effectiveness of latent visual evidence reuse for multi-step video-agent planning.
[MA-2] MasDrift: Benchmarking Authorization Preservation Across Multi-Agent Architectures
【速读】：该论文旨在解决多智能体系统（Multi-agent Systems, MAS）在长期任务分解过程中存在的授权边界模糊问题，即委托目标在跨层级或同级智能体间传递时，其原始权限约束未能有效保留。现有安全评估主要聚焦于对抗性攻击，缺乏对约束漂移（constraint drift）在架构层面的可控评估。为此，作者提出MasDrift基准，涵盖8个领域共600个良性生产力任务，每项任务均明确区分所需工作与保留操作。该基准通过对比单智能体、集中式与分布式协调模式，在不同层级深度和同级宽度条件下，量化任务完成率与授权保留度。实验表明，在通用多智能体配置下，集中式层级结构的任务完成率达93.9%–98.6%，显著优于同级网络的85.7%–87.0%；但未经授权行为的发生率也从0.6%–0.8%上升至2.7%–19.8%，且随层级加深而加剧。进一步比较两种防御机制：一种将每个待处理调用重新锚定至原始用户请求，可有效降低未经授权行为，但导致综合完成率下降1.6分；另一种沿委托链传递衰减策略，虽能维持授权一致性，却可能阻塞必要工作，最高造成36.3分完成率损失。异构案例研究证实，失败根源在于协作机制而非模型能力。因此，MasDrift揭示了集中化设计中的性能与安全权衡，并将授权保留性确立为多智能体系统设计中可量化的关键属性。
链接: https://arxiv.org/abs/2608.07556
作者: Zhuoning Xu,Xiucheng Zhang,Hanjun Luo,Yingbin Jin,Yinpeng Dong,Hanan Salam
机构: University of California, Berkeley (加州大学伯克利分校); Tsinghua University (清华大学); Alibaba Cloud (阿里云); Meta (Meta)
类目: Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
备注: 22 pages, preprint
Abstract:Multi-agent systems (MAS) decompose long-horizon tasks across supervisors and subagents, but delegated goals do not necessarily carry their original authorization boundaries. Existing safety benchmarks mainly study adversarial compromise, while work on constraint drift lacks controlled architecture-level evaluation. We introduce MasDrift, a benchmark of 600 benign productivity tasks across eight domains. Each task pairs required work with reserved actions. MasDrift compares single-agent, centralized, and decentralized coordination while varying hierarchy depth and peer width, measuring task completion and authorization preservation. Across generic multi-agent conditions, centralized hierarchies achieve 93.9–98.6% task completion versus 85.7–87.0% for peer networks, while unauthorized actions occur in 2.7–19.8% of tasks versus 0.6–0.8%, a gap that widens with hierarchy depth. We further compare two defenses that differ in where authorization evidence resides. One re-anchors every pending call to the original user request. The other carries an attenuated policy along the delegation chain. Re-anchoring reduces unauthorized actions in every model configuration we evaluate, at a cost of 1.6 points of pooled completion. Chain propagation blocks required work instead, forfeiting up to 36.3 points. A heterogeneous case study confirms that the failure follows from coordination rather than model strength. MasDrift exposes a centralization tradeoff and makes authorization preservation a measurable property of MAS design.
[MA-3] An AI Scientist that Doesnt Drift: Taste Structure and Falsifiable Findings in a Quadruped Navigation Research Loop
【速读】：该论文旨在解决由大型语言模型驱动的自主研究循环在大规模机器学习实验中普遍存在的“局部优化漂移”问题，即研究过程倾向于不断优化特定指标，而忽视了对初始假设的根本性检验。其解决方案的关键在于引入一种结构化设计：首先，采用不可变的实验卡片（immutable experiment card），以固定格式记录每次迭代的预测与实际结果，防止因事后修改而扭曲假设验证；其次，引入具有特定机械角色的子代理（specialised subagents），确保任务分工清晰且职责受限；最后，核心创新是提出kkanbu——一个将用户研究偏好编码为类型化知识图谱（typed knowledge graph）的偏好型判断代理（preference oracle），仅此组件被授权进行主观决策。通过在11个研究流中并行运行相同的研究循环（有/无kkanbu），实验证明两组均未出现显著漂移，约75%的假设被成功证伪；但关键差异在于方向选择：仅有含kkanbu的一侧探索了测试时适应（test-time adaptation），主导生成了最优策略，并实现了跨研究流的知识迁移，避免重复推导。因此，该框架的核心在于通过结构约束保证研究诚实性，而由kkanbu决定研究探索的方向。
链接: https://arxiv.org/abs/2608.07542
作者: Yiwen Zhang,Eloise Zeng,Jaeha Lee,Tony Yue Yu
机构: 未知
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Multiagent Systems (cs.MA); Robotics (cs.RO)
备注:
Abstract:Autonomous research loops driven by large language models can run machine-learning experiments at scale but tend to drift toward local refinements of whichever metric they optimise rather than testing the hypotheses that motivate the experiments. We address this structurally and present an AI Scientist for studying generalisation in quadruped robot navigation policies in simulation. Building on the autoresearch paradigm of Karpathy, our loop adds three components: an immutable experiment card that pairs each iteration’s prediction with its outcome under a fixed schema, so a falsified hypothesis cannot be retconned; specialised subagents restricted to mechanical roles; and kkanbu, a preference oracle that holds the user’s research taste as a typed knowledge graph and is the only component permitted to make subjective judgements. To isolate the oracle we run the identical loop twice across eleven research streams, with and without kkanbu. Neither arm drifts: both falsify roughly three quarters of their own hypotheses, and the best trained policy comes from the oracle-less arm. What the oracle changes is direction, not score: it alone explores test-time adaptation, it authored the winning designs where its arm led, and it carried lessons across streams that the other arm repeatedly re-derived. The scaffold keeps the loop honest; kkanbu decides where it looks.
[MA-4] NeuroPilot: An Agent -Driven Smart Pipeline for Processing Quality Control and Managing Neuroimages
【速读】：该论文旨在解决神经影像数据处理流程中三大脆弱环节——数据标准化、模态特异性预处理及质量控制（QC）——在实际应用中依赖项目定制脚本、环境适配调参以及大量人工干预所导致的效率低下与可复现性差的问题。其核心解决方案是提出NeuroPilot，一个基于大语言模型（LLM）驱动的多智能体系统，将神经影像处理、质量控制与数据管理的专业知识数字化为三个可调用的技能模块：dcm2bids-skill、neuroimage-pre-skill 和 qc-agent-skill。该系统通过LLM实现工作流的自主编排，能够泛化多种基础设施配置并统一管理，显著提升可扩展性。在实践中，NeuroPilot首先通过dcm2bids-skill完成数据标准化，随后根据数据模态和队列特征动态调度最优预处理流程（如将T1w和fMRI数据交由fMRIPrep处理，或为婴幼儿队列启用专用流程），再由qc-agent-skill驱动基于证据的半自动化质量控制，借助三维浏览器仪表盘与多层级验证机制，有效识别并优化失败案例，复杂问题则自动升级至人工审核。定量评估显示，该QC代理对558例生产数据进行筛查，其自动标记结果与FreeSurfer拓扑缺陷指标高度一致；在婴幼儿处理流程中实现了100%（201/201）的高质量完成率。更重要的是，NeuroPilot将传统需2–3个月的人工培训与数据处理周期压缩至单周内完成，极大提升了神经影像分析的效率与可推广性。
链接: https://arxiv.org/abs/2608.07541
作者: Yiyao Chen,Yucheng Li,Jungong Tong,Shaoqi Wang,Kunhao Zhou,Ziquan Wei,Monica Murea,Marissa DiPiero,Tingting Dan,Guorong Wu
机构: 未知
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Multiagent Systems (cs.MA)
备注: 21 pages, 6 figures
Abstract:Transforming raw neuroimage archives into analysis-ready derivatives relies on three brittle stages: data standardization, modality-specific preprocessing, and quality control (QC). While individual neuroimaging tools are well developed, their orchestration requires project-specific scripts, environment-adaptive tuning, and labor-intensive manual QC. To address this, we introduce NeuroPilot, a multi-agent system that digitalizes the expertise of neuroimage processing, QC, and data management into three LLM-invocable skills: dcm2bids-skill, neuroimage-pre-skill, and qc-agent-skill. The LLM-driven agent autonomously orchestrates workflows, generalizing various infrastructure settings into a single configuration to achieve the highest scalability. Demonstrating the system’s generalizability, we deployed NeuroPilot across 17 cohorts (123,000 subjects) spanning infant to aging populations and multiple MRI modalities (structural, diffusion, functional). In practice, after standardizing data via the dcm2bids-skill, the agent dynamically routes datasets to the optimal neuroimage-pre-skill based on available modalities and cohort traits (e.g., dispatching T1w and fMRI data to fMRIPrep, or selecting specialized pipelines for infant cohorts). The qc-agent-skill then drives an evidence-based, semi-automated QC via a 3-D browser dashboard, utilizing a multi-tiered verification system to optimize failed cases and escalate complex issues for supervisor inspection. Quantitatively, our QC agent screened 558 production subjects, validating its automated flags against FreeSurfer’s topology-defect metrics. The infant processing pipeline achieved a 100% (201/201) completion rate on QC-validated inputs. Importantly, NeuroPilot compresses the traditional 2–3 month timeline for training staff and processing complete datasets into a single week. NeuroPilot is deployed in this https URL.
自然语言处理
[NLP-0] Build it Break it Repeat: Benchmarking and improving LLM -manipulated disinformation detection in social media posts
【速读】： 该论文旨在解决生成式人工智能（Generative AI）背景下，社交媒体上机器生成虚假信息（machine-generated disinformation）检测系统在面对持续迭代的对抗性攻击时鲁棒性不足的问题。传统静态基准评估无法反映检测模型在真实对抗环境下的表现，尤其当攻击者通过系统性改写手段主动规避分类时。为此，本文提出并应用“构建-破坏-重复”（Build it, Break it, Repeat, BiBiR）的迭代框架，以模拟对抗者对虚假信息内容进行多轮演化攻击的过程，从而更真实地测试检测器的抗干扰能力。其解决方案的关键在于：首先采用基于回译（back-translation）与大语言模型角色化重写（LLM persona-based rewriting）相结合的策略生成高隐蔽性的对抗样本，实现高达95%的标签翻转率（Label Flip Rate, LFR），同时保持原始语义不变；其次，构建端到端的三元组对比学习模型并引入动态锚点切换（Dynamic Anchor Switching, DASS）架构，显著提升了检测器在复杂对抗攻击下的准确率（平均达72.68%），相较强基线模型（fine-tuned e5-small-LoRA）提升15个百分点。研究结果表明，迭代式对抗测试能有效暴露检测系统的脆弱环节，并推动鲁棒性改进，但仍需结合语义保真度分析，以区分真正有效的对抗规避行为与语义篡改的无效攻击。
链接: https://arxiv.org/abs/2608.09510
作者: Kevin Thomas,Milosz Kasprzyk,Reuel C Igbokwe Onuigbo,Elliott Pert,Cameron Tovey,João A. Leite,Olesya Razuvayevskaya,Carolina Scarton
机构: University of Sheffield (谢菲尔德大学)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Social and Information Networks (cs.SI)
备注: Under review
Abstract:Detecting machine-generated disinformation on social media is increasingly difficult as large language models (LLMs) make it easier to generate and rewrite misleading content at scale. Static benchmark evaluations, measuring detector performance on fixed held-out datasets, do not capture how detectors behave when posts are deliberately transformed to evade classification. This paper adapts the Build it, Break it, Fix it framework into Build it, Break it, Repeat (BiBiR): iterative sessions designed to stress-test detectors’ robustness under iterative adversarial conditions, evaluating whether models remain reliable when disinformation posts are systematically transformed to evade classification. Across five iterations, the findings show that the best adversarial breakers’ transformations came from a combination of back-translation and LLM persona-based rewriting, with the best performing technique achieving a 95% label flip rate (LFR), whilst still preserving the meaning of the original posts. The best builders’ model was a triplet contrastive model with a dynamic anchor switching (DASS) architecture, which achieved an average accuracy of 72.68%, outperforming the strong baseline (a fine-tuned e5-small-LoRA) by 15 percentage points on the most robust set of breakers’ adversarial attacks. The results demonstrate that an iterative framework best exposes detector weaknesses and pushes robustness improvements; however, it may still require semantic preservation analysis to distinguish valid adversarial evasion from transformations that changed the original disinformation claims’ meaning.
[NLP-1] Learning Preference Adaptation for Large Language Model Personalization via Verbal Reinforcement Learning
【速读】： 该论文旨在解决通用用户偏好摘要在下游任务中因包含无关信息而导致上下文容量浪费与跨任务干扰的问题。其核心挑战在于如何在不依赖人工设计的前提下，高效生成与特定任务相关的偏好表示。解决方案的关键是提出一种无需训练的元学习框架\textscAlignXada，通过语义强化学习迭代优化可复用的文本精炼策略（textual refinement policy），将通用偏好摘要自适应地转化为任务相关的表示。该方法在13个任务和3种下游模型（共39个任务-模型组合）上平均提升3.82点，仅保留原始偏好令牌的22.8%，且在36个场景中优于检索增强生成（RAG）。扩展的忠实性分析表明，精炼后的偏好仍基本源自原始输入并保留了关键个性化信号，验证了从偏好端进行任务适配作为终身个性化智能体中通用记忆构建的实用补充。
链接: https://arxiv.org/abs/2608.09507
作者: Yuting Liu,Wei Wu,Jianzhe Zhao,Guibing Guo
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Natural language user preferences provide an interpretable interface for LLM personalization. However, universal preference summaries often contain information irrelevant to a particular downstream task. Directly supplying the full preference summary therefore wastes context capacity and introduces cross-task distraction, while manually designing task-specific preference views is difficult to scale. In this work, we study \emphtask-specific preference adaptation: given a universal user preference summary and a downstream task, derive a task-conditioned representation that preserves sufficient decision-relevant evidence while removing redundant context. To this end, we propose \textscAlignXada, a training-free meta-learning framework that induces reusable textual refinement policies for adapting universal preference summaries to task-specific ones. The refinement policy is iteratively optimized by a meta learner through verbal reinforcement learning. Across 13 tasks and three downstream models (39 task–model cells), \textscAlignXada achieves an average gain of 3.82 points, improving 33 cells while retaining only 22.8% of the original profile tokens and outperforming RAG in 36 cells. An extended faithfulness analysis further shows that the refined profiles remain largely grounded in the source preferences while preserving task-relevant personalization signals, suggesting that profile-side adaptation serves as a practical complement to universal memory construction for lifelong personalized agents.
[NLP-2] Depth-adaptive Inference of Looped Language Models via Continuous Depth Batching
【速读】： 该论文旨在解决循环语言模型（Looped Language Models, LLMs）在实现深度自适应推理（depth-adaptive inference）时面临的高效批处理难题。传统批处理机制依赖于统一的前向传播，而自适应深度机制使得同一批次中的不同标记（token）需执行不同次数的循环迭代，导致无法进行统一计算，严重制约了推理效率。现有解决方案如基于循环层级的调度虽被提出，但缺乏端到端实现。其核心挑战在于：循环架构中包含非循环的边界阶段（如词元嵌入和语言模型输出头），这些部分需以不同于循环步骤的频率进行调度。为此，论文提出连续深度批处理（Continuous Depth Batching, CDB），以单个循环迭代为调度粒度，将边界阶段与循环步骤分别置于独立的优先队列中，提前一步做出退出决策，并将所有调度工作与GPU计算重叠执行。实验结果表明，在Ouro 1.4B和Huginn 3.5B模型上，CDB可实现高达99%的理论最大加速比，相较静态批处理带来1.5–1.9倍的离线吞吐量提升，并在动态服务负载下降低45%–90%的归一化延迟。
链接: https://arxiv.org/abs/2608.09444
作者: Kristian Schwethelm,Daniel Rueckert,Georgios Kaissis
机构: Technical University of Munich (TUM), Germany; Imperial College London, UK; Munich Center for Machine Learning (MCML), Germany; Hasso Plattner Institute for Digital Engineering, University of Potsdam, Germany
类目: Machine Learning (cs.LG); Computation and Language (cs.CL); Distributed, Parallel, and Cluster Computing (cs.DC)
备注:
Abstract:A main promise of looped language models (LMs) is depth-adaptive inference. By iterating a block of shared layers a variable number of times, the model can use less compute for “easy” tokens and more for “hard” ones. However, this adaptivity breaks standard batching: tokens in the same batch now require a different number of loops, so there is no unified forward pass, making efficient inference difficult. Standard inference frameworks like vLLM schedule on the token level and cannot handle this because tokens need to be removed from the batch within the forward pass. Loop-level scheduling has been proposed as a solution, but never implemented end to end. The key challenge is that looped architectures also contain non-looped boundary stages (e.g., token embedding and LM head) that must be scheduled at different frequencies than the loop. We introduce continuous depth batching (CDB), which schedules at the granularity of individual loop iterations. CDB handles boundary stages and loop steps in separate priority queues, makes exit decisions one step ahead, and overlaps all scheduling work with GPU computation. On Ouro 1.4B and Huginn 3.5B, CDB can realize up to 99% of the theoretical maximum speed-up from adaptive-depth, translating to 1.5 - 1.9\times higher offline throughput and 45 - 90% lower normalized latency under dynamic serving load.
[NLP-3] ZetaGPT : A Reference Implementation of Positional–Encoding–Free State–Space–Attention Language Models
【速读】： 该论文旨在解决基于Transformer的语言模型中自注意力机制（self-attention）因具备排列等变性（permutation-equivariant）而缺乏内在的词元顺序表征能力的问题。现有方法通过引入可学习的位置嵌入或手工设计的位置编码（如旋转位置编码，Rotary Positional Encoding, RoPE）来显式补充位置信息，将位置感知视为架构层面的外加能力而非模型本身的固有属性。为实现无位置编码（positional-encoding-free）的架构设计，本文提出一种创新解决方案：在自注意力计算前引入因果状态空间方程（causal state-space equations），通过递归状态动态隐式地将序列信息编码至词元表示中。这一机制使后续注意力层在处理位置感知的表示时无需依赖显式的位置编码，同时保持了自注意力的强表达能力。核心创新在于将位置信息的编码过程从显式架构设计转化为由动态系统内在演化所驱动的隐式过程。研究团队据此构建了名为\textscZetaGPT的小型混合语言模型，其不仅实现了无位置编码的架构设计，还提供了一个完全开源、端到端的训练流程，涵盖数据集构建、分词器训练、预训练、监督微调、基于人类反馈的强化学习（RLHF）以及纯强化学习驱动的思维链（Chain-of-Thought, CoT）推理，成为首个无显式位置编码且可复现的小型语言模型参考实现，为相关领域的研究与实验提供了重要基础。
链接: https://arxiv.org/abs/2608.09432
作者: Róisín Luo
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Transformer-based language models rely on self-attention, whose computation is permutation-equivariant and therefore lacks an intrinsic mechanism for representing token order. Existing architectures address this limitation by explicitly incorporating positional information through learned positional embeddings or hand-crafted positional encodings, such as rotary positional encoding (RoPE), treating positional information as an architecturally acquired capability rather than an inherent property of the model. Motivated by the pursuit of positional-encoding-free architectures, this work explores a language model architecture that integrates causal state-space equations to implicitly encode positional information before attention computation. Specifically, each model block applies a causal state-space equation before self-attention, allowing recurrent state dynamics to encode sequential information into token representations. Consequently, subsequent attention layers operate on position-aware representations without requiring explicit positional encodings while retaining the expressive modeling capacity of self-attention. We present \textscZetaGPT, a compact hybrid language model designed for research, rapid prototyping, algorithm verification, and educational applications. In addition to the proposed architecture, \textscZetaGPT provides a fully open-source, end-to-end training pipeline encompassing dataset construction, tokenizer training, pretraining, supervised fine-tuning, reinforcement learning from human feedback (RLHF), and chain-of-thought (CoT) reasoning via pure reinforcement learning. To the best of our knowledge, \textscZetaGPT is the first open-source small language model without explicit positional encoding and establishes a compact, reproducible reference implementation for the development and empirical study of positional-encoding-free language models.
[NLP-4] Reducing Pretraining-Generation Mismatch in Diffusion Language Models
【速读】： 该论文旨在解决扩散语言模型（diffusion language model, dLLM）在预训练过程中存在的提示延续（prompt continuation）不匹配问题：传统dLLM的原生预训练会随机同时破坏提示（prompt）和后续内容（continuation）的令牌，导致其无法有效维持生成时所需的“干净前缀”接口，从而削弱了提示条件生成的能力。其解决方案的关键在于提出一种名为前缀条件扩散（Prefix-Conditioned Diffusion, PCD）的新预训练目标，通过在训练层面重构注意力掩码、污染掩码与标签构建机制，实现对干净前缀部分的自回归监督，仅对未知的延续部分应用扩散去噪过程。这一设计使训练阶段的局部上下文接口更贴近块扩散模型在推理时的实际使用方式，无需引入额外解码器、验证器或新的推理模式。此外，PCD还显式分离了样本内前缀条件监督与样本间目标混合机制，便于独立分析局部对齐信号。实验结果表明，在LLaDA2-Mini和Qwen-1.7B等不同架构上，PCD均显著优于同族原生dLLM基线，分别在主要基准平均得分上提升4.2%（+2.56点）和14.2%（+4.86点），证明了通过对齐预训练上下文分布与提示条件生成范式，可有效弥合扩散语言模型的生成性能差距，且无需改变推理流程。
链接: https://arxiv.org/abs/2608.09424
作者: Xiaocheng Lu,Huabin Liu,Song Guo,Jianguo Li
机构: 未知
类目: Computation and Language (cs.CL)
备注: 12 pages, 9 figures, 1 table
Abstract:Autoregressive language models align training and use: generation conditions on a clean prompt, and training predicts future tokens from clean left context. Diffusion language models offer parallel denoising, but native dLLM pretraining can randomly corrupt prompt and continuation tokens together, weakening the clean-prefix interface needed for prompt-conditioned generation. We identify this mismatch for prompt continuation and propose PCD (Prefix-Conditioned Diffusion), a pretraining objective that combines AR prefix supervision with no-shift suffix denoising. At the training-objective level, PCD changes the attention mask, corruption mask, and label construction in continued pretraining; it does not require an autoregressive decoder, verifier, or new inference mode. By supervising the clean-prefix side autoregressively and applying diffusion only to the unknown continuation, PCD makes the local training interface resemble how block-diffusion models are queried at evaluation time. We further separate intra-sample prefix conditioning from inter-sample objective mixing, allowing us to identify the local alignment signal separately from the optional batch-level mixing knob. Across LLaDA2-Mini and Qwen-1.7B backbones, PCD consistently improves over same-family native dLLM stable baselines, reaching a 4.2% relative gain on the main LLaDA2-Mini six-benchmark average (+2.56 points) and a 14.2% relative gain in the primary Qwen mechanism comparison (+4.86 points). These results suggest that aligning the pretraining context distribution with prompt-conditioned generation can recover a measurable part of the dLLM continuation gap without changing inference.
[NLP-5] Intent Speaks Louder: Controllable User Simulation Beyond Response Imitation
【速读】： 该论文旨在解决交互式助手训练与评估中用户模拟（User Simulation）存在的核心问题：当前的用户模拟模型在生成下一回合用户话语时，面临“一对多”（one-to-many）的语义歧义性，即同一用户画像和对话上下文可能支持多种合理的后续回应，但模型难以准确控制所期望的局部交互意图（local interaction intent），导致生成的响应可能偏离预期意图（如将修复意图误判为接受意图），从而影响对话系统的训练质量。其解决方案的关键在于提出一种可调控的用户模拟框架——UserIDA（User Intent-Directive Alignment），通过将交互意图作为每轮对话的显式指令（per-turn directive）进行分离建模，实现对意图表达的精准控制。具体而言，UserIDA构建了一个六类交互意图接口，采用监督微调学习条件化生成，并在基于群体的强化学习中引入意图校准策略优化，确保奖励函数不仅衡量回复的整体质量，还强制意图违规的候选回复在混合群体中排名低于合规项。实验结果表明，该方法在LMSYS-USP数据集上达到86.6%的意图准确率，显著优于最强基线24.3个百分点，同时提升语义与风格相似性；在上下文内干预任务中，91.7%的对话状态能成功实现至少四种目标意图，远超基线的22.9%。这验证了逐轮意图控制作为用户模拟中与响应保真度并行的重要维度，具有显著提升对话系统训练有效性与可控性的潜力。
链接: https://arxiv.org/abs/2608.09420
作者: Bo Wang,Ruixing Zhang,Yunqi Liu,Yang Zhang,Liangzhe Han,Tongyu Zhu,Leilei Sun
机构: 未知
类目: Computation and Language (cs.CL)
备注: 26 pages, 7 figures, 16 tables. Code: this https URL
Abstract:User simulators are widely used as scalable environments for training and evaluating interactive assistants. Generating the next user turn is inherently one-to-many: the same profile and dialogue context may support multiple plausible continuations with different local interaction intents. A fluent response may therefore advance the dialogue through an inappropriate intent, such as acceptance rather than repair. Our key insight is that controllable user simulation should separate which local interaction intent the next user turn should realize from how that intent is expressed in language. We introduce UserIDA (User Intent-Directive Alignment), which exposes interaction intent as an explicit per-turn directive. UserIDA defines a six-way intent interface, learns directive-conditioned generation through supervised fine-tuning, and uses intent-calibrated policy optimization during group-based reinforcement learning. The reward preserves composite response quality while ensuring that intent-violating candidates rank below compliant alternatives in mixed groups. On LMSYS-USP, UserIDA achieves 86.6% intent accuracy, outperforming the strongest dedicated user-simulator baseline by 24.3 percentage points while improving semantic and stylistic similarity. In within-context interventions, it realizes at least four of the six target intents in 91.7% of evaluated dialogue states, compared with 22.9% for the strongest external baseline. These results establish per-turn intent control as a complementary dimension to response fidelity in user simulation.
[NLP-6] Universal or Language-Family-Specific Script Unification for Cross-Lingual Transfer? A Case Study on Turkic Languages
【速读】： 该论文旨在解决多语言模型在书写系统差异较大的亲属语言间跨语言迁移能力受限的问题，尤其关注使用不同文字系统的突厥语族语言之间的表面形式重叠不足所导致的性能瓶颈。其核心解决方案是通过脚本统一（script unification）增强语言间的共享表示能力，具体比较了两种方法：通用型罗马化工具uroman与面向突厥语族的专用统一突厥文字（Common Turkic Script, CTS）。实验基于11种突厥语的转写维基百科语料训练匹配的fastText模型，并在WikiANN命名实体识别（NER）和通用依存句法标注（POS）任务上进行评估。结果表明，CTS与uroman在NER任务上表现相当且均显著优于官方单语fastText基线；而在POS任务中，无绝对优势方案，语言内字符n-gram覆盖度与目标语言监督信号的可用性共同决定性能表现。尽管CANINE-c在整体POS平均得分上更优，但结构简单的fastText系统在多个语料库上仍具竞争力。研究揭示，脚本统一的有效性取决于目标语言特性、生成的子词重叠程度以及监督资源的丰富程度。
链接: https://arxiv.org/abs/2608.09356
作者: Zijie Zhang
机构: The Chinese University of Hong Kong, Shenzhen
类目: Computation and Language (cs.CL)
备注:
Abstract:Closely related languages written in different scripts expose little surface overlap to multilingual models, limiting cross-lingual transfer. We compare two approaches to script unification: the general-purpose uroman romanizer and the family-specific Common Turkic Script (CTS). We train matched fastText models on transliterated Wikipedia corpora from 11 Turkic languages and evaluate them on WikiANN named entity recognition and Universal Dependencies part-of-speech tagging. CTS and uroman show no significant difference on NER, while both substantially outperform the official monolingual fastText baselines. POS results reveal no universal winner: language-specific differences are associated with the cross-lingual character n-gram coverage induced by each representation, while within-language coverage becomes more important when target-language supervision is available. Although CANINE-c achieves higher overall POS averages, the substantially simpler fastText-based systems remain competitive on several treebanks. Overall, the effectiveness of script unification depends on the language, the induced subword overlap, and the available supervision.
[NLP-7] Evo-Bench: Can Language Models Improve Agent Harness?
【速读】： 该论文旨在解决当前大型语言模型（LLMs）在自主智能体评估中长期存在的核心问题：如何有效、系统地衡量智能体自主优化其自身运行框架（harness）的能力。现有评估方法无法将框架改进效果与基础模型性能相分离，易导致任务特定过拟合，并难以捕捉长周期迭代式研究过程中的演化规律。为此，论文提出Evo-Bench——首个针对搜索（Search）、办公（Office）和通用（General）三大智能体领域设计的基准测试体系，专门用于评估模型内在的框架演化能力。其关键解决方案在于引入一种新型“框架引导式构建”（harness-guided construction）范式：首先通过辅助任务演化识别对框架改进真正敏感的任务，再结合感知敏感性的分层划分策略，确保跨任务套件的鲁棒泛化能力。实验结果表明，顶尖模型在该基准上实现了高达16.6分的绝对性能提升，接近人工设计最优框架的水平；同时分析揭示，自主演化在通用任务和搜索任务中优于人工框架，但在需高度定制化处理流程的办公任务中表现受限；此外，研究还发现早期饱和等时间维度异常现象，并证实所生成的框架具备强可迁移性，能持续提升多种策略模型的推理效能。
链接: https://arxiv.org/abs/2608.09096
作者: Lisheng Huang,Chen Yang,Hao Zhou,Huatong Song,Zongchao Chen,Ran Le,Yang Song,Wayne Xin Zhao,Tao Zhang
机构: Renmin University of China (中国人民大学); BOSS Zhipin (Boss直聘)
类目: Computation and Language (cs.CL)
备注:
Abstract:Large Language Models (LLMs) have driven rapid progress in autonomous agents, yet standard evaluations remain confined to static task solving. An emerging frontier is harness evolution—the agent’s capacity to autonomously optimize its own operating harness. However, systematically benchmarking this capability remains challenging, as existing evaluations fail to isolate harness improvements from base model strength, prevent task-specific overfitting, or capture long-horizon iterative research. To address these challenges, we introduce Evo-Bench, the first benchmark designed to evaluate models’ intrinsic harness-evolving capabilities across Search, Office, and General agent domains. To rigorously isolate this capability, Evo-Bench employs a novel harness-guided construction framework: it leverages auxiliary-task evolution to identify tasks genuinely sensitive to framework improvements, followed by sensitivity-aware stratified splitting to ensure robust cross-suite generalization. Extensive evaluations across nine frontier and open-weight models reveal that top models achieve massive absolute gains reaching 16.6 points, closely approaching state-of-the-art human-engineered baselines. Crucially, while autonomous evolution outpeforms artificial harness in General tasks and excels in Search tasks, it struggles in Office tasks that demand highly specific processing workflows. Furthermore, our analysis exposes critical temporal anomalies like early saturation, while demonstrating that the synthesized harnesses act as highly transferable reasoning structures, consistently boosting diverse policy models.
[NLP-8] he Announcement Carries the Cue: Markup Boundaries and the Notation of Pre-Training Corpora
【速读】： 该论文旨在解决当前自然语言处理数据集在标注（notation）层面缺乏系统记录与评估的问题，即文档排版结构的书写方式这一关键训练变量未被数据卡片（dataset card）所捕获。尽管已有研究证实文本提取策略会影响模型行为，但从未对这些策略所引入的标注形式进行量化测量。其解决方案的关键在于提出“清洁窗口生存率”（clean-window survival）这一确定性指标，用于衡量文本流中仍需依赖边界推断的残余部分，并从三个维度进行实证分析：首先，在十三个公开语料库的普查中发现，视觉转换的PDF片段中生存率仅为0.153，远低于C4语料库的0.889，揭示出稀缺资源并非无标记文本本身，而是长段无标记文本；其次，预注册供应测试表明，剩余内容多为机构生成而非消费者生成；再次，作者的预注册预测失败——文本转换器不会在散文中虚构结构，这一“零效应”反而促使研究构建了具备鲁棒性的可靠性机制。进一步实验显示，删除结构声明会使后续散文的可预测性显著下降，而仅改变其标注形式则无影响，说明真正起作用的是声明内容本身，而非符号形式（sigil），从而将变量焦点从“标注”转移至“声明”。最后，在作者施加层面，基线模型不会在原文基础上强加标记格式，即使移除所有声明，也几乎不会自行恢复，其行为与零基准无异。基于上述发现，论文提出“纯框架”（pure frame）格式：以原始顺序排列段落，所有声明移入可逆的侧车文件，通过是否包含声明而非其标注形式来混合数据。研究强调应根据格式操作符所训练的能力选择格式，而非追求保真度，并主张在数据卡片中记录提取器身份与生存率。
链接: https://arxiv.org/abs/2608.09093
作者: E. M. Freeburg
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 44 pages, 10 tables, 7 figures. Pre-registered protocols and their amendment history ship with the repository. Code, data, and instruments: this https URL
Abstract:How a document’s arrangement is written down, its notation, is a training variable that no dataset card records. The field has established that text-extraction choices change model behaviour, and has never once measured the notation of what those choices put into the corpus. We define clean-window survival, a deterministic count of how much of a stream still demands the boundary inference, and measure notation on three fronts. What corpora carry: a census of thirteen public corpora, where survival falls to 0.153 in a vision-converted PDF slice against 0.889 in C4; the scarce resource is not unmarked text but long unmarked text; a pre-registered supply test finds what remains institutional, not consumer. Our own pre-registered prediction failed: converters do not fabricate structure on prose, and that null forced the reliability mechanism that survives it. What readers use: across five base models spanning 0.6B to 8.2B and two pipelines, deleting a structural announcement makes the following prose measurably harder to predict, while swapping its notation moves nothing. That zero does not make notation unimportant; it relocates the variable: the operative cue is the announcement, not the sigil. What writers impose: a bounded null. Base models do not impose the marked register above the authored baseline, and handed prose with every announcement deleted they do not put one back, at a rate indistinguishable from zero against an authored reference of zero. We ship the format those measurements imply: the pure frame, paragraphs in authored order, every announcement deleted into a reversible sidecar, mixed against the marked copy over announcement presence rather than notation. Choose format operators by the capability they train, not by the fidelity they preserve, and record extractor identity and survival on data cards.
[NLP-9] Security and Privacy Taxonomy Generation from Mobile App Reviews
【速读】： 该论文旨在解决移动应用评论中隐私与安全（privacy and security）相关议题的自动化分类难题。现有分类体系依赖人工构建，难以适应用户反馈数据的快速演化。其核心挑战在于如何在大规模评论语料（数百万级）上实现可扩展的自动分类体系构建。论文提出的解决方案关键在于提出TaxoScale框架：首先通过过滤筛选出超过60万条与隐私和安全相关的应用评论，构建大规模专用语料库；其次，采用基于专家定义初始结构的递归分层聚类（Recursive Hierarchical Clustering）结合大语言模型（LLM）进行节点命名的方法，实现高效且可扩展的层次化分类体系构建。该方法在路径准确率、层级深度、覆盖率及新颖性等指标上均优于现有基线，成功挖掘出原有分类体系中缺失的新分支，显著提升了自动化分类的全面性与适应性。
链接: https://arxiv.org/abs/2608.09049
作者: Moghis Fereidouni,Vinaik Chhetri,Umar Farooq,A.B. Siddique
机构: University of Kentucky (肯塔基大学); Louisiana State University (路易斯安那州立大学)
类目: Computation and Language (cs.CL)
备注:
Abstract:Mobile app reviews are a rich, continuously renewing source of how users experience privacy and security, yet existing taxonomies of these concerns are hand-crafted and cannot keep pace with the evolving nature of the data. Automating taxonomy construction is the natural response, but scalability is the core challenge: current LLM- and clustering-based methods are developed for scientific corpora of a few thousand documents and do not extend to app review collections numbering in the hundreds of thousands. We address this gap in two ways. First, we filter app reviews for privacy- and security-related content, yielding a comprehensive corpus of over 600K reviews. Second, we introduce TaxoScale, a pipeline that handles taxonomy construction at this scale by extending an expert-defined taxonomy via Recursive Hierarchical Clustering and LLM-based node naming. TaxoScale outperforms strong automatic-taxonomy baselines on path, level, coverage, and novelty metrics, and discovers novel branches absent from prior taxonomies.
[NLP-10] Measuring the Tokenization Premium: A Cost Audit for Underserved Language Communities IJCAI2026
【速读】： 该论文旨在解决大语言模型（Large Language Models, LLMs）在多语言应用中因分词（tokenization）不平等导致的技术与经济壁垒问题。尽管生成式 AI 已被广泛用于教育和技术辅助场景，但其底层基础设施对不同语言的支持存在显著差异，其中分词效率的不对称性尤为突出：语义等价的内容在不同语言间可能产生数量级差异的标记（token）数，进而影响 API 成本、响应延迟及实际可用上下文长度。为此，论文提出可复现的分词公平性审计工具（Tokenization Equity Audit, TEA），针对英文编程调试语料库（120项 Python 调试任务）进行多语言翻译评估，涵盖孟加拉语、印地语、阿拉伯语、泰米尔语和约鲁巴语，并对比 GPT-4o 的 o200k base、Qwen2.5-7B 与 Mistral-7B 三种主流分词器的表现。结果显示，孟加拉语在 GPT-4o 下需 1.56 倍于英语的 token 数，使原本 128k token 的上下文窗口有效容量降至 82k 英语等效量；而在 Qwen2.5 与 Mistral 模型下，该比例最高达 4.5 倍。约鲁巴语虽使用拉丁字母，却在 GPT-4o 下呈现 2.37 倍的最高分词溢价，表明分词不公不能仅由书写系统或语言家族解释。研究揭示分词机制已成为影响边缘语言社区获取低成本或离线部署型 AI 教育工具的关键非功能性障碍，强调应将分词视为具有公平性意义的基础设施层级，亟需在模型设计与部署策略中予以系统性考量。
链接: https://arxiv.org/abs/2608.09046
作者: Avijit Roy,Proma Roy,Hrishitva Patel
机构: John Jay College of Criminal Justice, City University of New York(纽约城市大学约翰杰伊刑事司法学院); The City College of New York, City University of New York(纽约城市大学城市学院); University of Texas at San Antonio(圣安东尼奥德州大学)
类目: Computation and Language (cs.CL); Computers and Society (cs.CY)
备注: Accepted at IJCAI 2026 Workshop ( this https URL )
Abstract:Large language models are increasingly deployed as general-purpose educational and technical assistance systems, but their underlying infrastructure does not treat languages equally. One underexamined source of disparity is tokenization: semantically equivalent content can require substantially different token counts across languages, affecting API cost, latency, and usable context length before a model is invoked. We introduce the Tokenization Equity Audit (TEA), a reproducible benchmark for measuring tokenization premiums in technical tutoring content. TEA evaluates three widely used tokenizers, GPT-4o’s o200k base, Qwen2.5-7B, and Mistral-7B, on a 120-item Python debugging corpus translated from English into Bengali, Hindi, Arabic, Tamil, and Yoruba. Bengali and Hindi serve as the primary validated cases, while the remaining languages provide exploratory cross-script and cross-family comparisons. Across this corpus, Bengali requires (1.56\times) as many GPT-4o tokens as English, reducing a nominal 128k-token context window to an effective 82k-token English-equivalent capacity for the same semantic content. With the Qwen2.5 and Mistral tokenizers, Bengali requires up to (4.5\times) the English token count. Yoruba, despite using the Latin script, exhibits the highest GPT-4o tokenization premium at (2.37\times), indicating that tokenization inequity cannot be explained by script family alone. These results demonstrate that tokenization can create measurable economic and functional barriers, highlighting the need to treat tokenization as an equity-relevant infrastructure layer for underserved language communities, particularly where educational systems depend on low-cost or offline-capable AI tools.
[NLP-11] Bridging the Gap Between Semantics and Reconstruction:Unifying Sign Language Translation and Production
【速读】： 该论文旨在解决将手语理解（Sign Language Understanding, SLU）与手语生成（Sign Language Production, SLP）统一于同一框架中的关键挑战。现有研究虽已实现多种手语理解任务（如孤立手语识别、连续手语识别及手语翻译）的整合，但将手语翻译（SLT）与手语生成（SLP）统一仍面临显著困难，因其涉及相反方向的模态映射——前者为手语到文本，后者为文本到手语。为此，该研究提出统一框架Uni-SLTP，其核心创新在于：（1）设计一个共享的手语分词器（shared sign tokenizer），能够将连续手语动作序列转换为离散符号与潜在表征，同时兼顾语义抽象与动作重建；（2）构建一个统一的条件自回归生成模型，可接受手语或文本作为输入，并在反向模态中生成目标序列。该方案通过共享表示学习与双向生成建模，有效弥合了手语与文本之间的模态鸿沟，实现在手语生成任务中更高的动作准确性，同时保持手语翻译性能的竞争力。
链接: https://arxiv.org/abs/2608.09045
作者: Xiao Liu,Shiwei Gan,Yafeng Yin,Jiaxin Yin,Bowen Guo,Yaqi Sun,Zhiwei Jiang,Lei Xie
机构: Nanjing University (南京大学); State Key Laboratory of Novel Software Technology (新型软件技术国家重点实验室)
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Multimedia (cs.MM)
备注:
Abstract:Recent advances in sign language (SL) research have shown a trend toward unifying multiple sign language understanding (SLU) subtasks, such as isolated sign language recognition (ISLR), continuous sign language recognition (CSLR), and sign language translation (SLT), within a single framework, leading to substantial progress. Meanwhile, sign language production (SLP), which generates sign sequences from text, has also attracted growing attention. This naturally raises an important question: can sign language understanding and production be unified within a single framework? Compared with unifying SLU subtasks, this problem is substantially more challenging. Existing SLU tasks largely share the same direction of mapping, namely from sign inputs to linguistic outputs, whereas SLT and SLP lie in opposite directions of sign-text mapping. A unified framework must therefore address two key challenges: (1) bridging the modality gap between continuous sign motions and discrete text tokens through a shared sign tokenizer that supports both linguistic abstraction and motion reconstruction; and (2) learning a single conditional autoregressive model that can take either sign or text as input and generate the corresponding target sequence in the opposite modality. To this end, we propose Uni-SLTP, a unified framework for SLT and SLP with two key components: (1) a shared sign tokenizer that converts sign sequences into discrete tokens and latent representations, capturing both semantic and reconstructive information; and (2) a unified autoregressive generation model that formulates both tasks as conditional sequence generation. Experiments on widely used public datasets show that Uni-SLTP achieves superior motion accuracy for SLP while maintaining competitive SLT performance.
[NLP-12] ree-of-Experience: Hierarchical Experience Management for Self-Evolving Agents
【速读】： 该论文旨在解决大语言模型（LLM）智能体在持续自我演化过程中，如何将环境交互转化为可信赖且可复用的经验这一核心问题。现有方法通常仅对单个推理轨迹进行优化或从相关轨迹中抽象共享知识，但其经验表征与底层推理过程脱节，导致反馈归因困难、跨任务迁移能力弱以及更新与检索效率低下，尤其在以结果层面反馈为主的复杂推理任务中表现不佳。为此，本文提出经验树（Tree-of-Experience, ToE）框架，其关键在于将经验组织结构与LLM智能体的层次化推理过程对齐：通过构建共享的分析视角与推理路径树，并基于环境反馈结果校准各节点的可靠性，从而实现系统性的经验更新、高效跨任务迁移与快速检索。实验结果表明，ToE在\textscGame of 24和\textscFinEvolveBench两个基准上显著提升性能与效率，在\textscGame of 24上相较无经验基线相对准确率提升31.4%，在\textscFinEvolveBench上平均指标提升达41.24%，而传统经验管理方法反而常劣于无经验基线。
链接: https://arxiv.org/abs/2608.09044
作者: Zihao Deng,Yining Zhu,Leiming Wang,Jingfei Lu,Junbo Wang,Chuncheng Ran,Yu Yang,Dixuan Yang,Jikun Shen
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Continual self-evolution requires LLM agents to transform environmental interactions into reliable and reusable experience. Existing methods typically refine individual trajectories or abstract shared knowledge from related trajectories, but their experience representations are often disconnected from the underlying reasoning process. This limits feedback attribution, cross-task transfer, and update and retrieval efficiency, particularly in complex reasoning tasks with outcome-level feedback. To overcome this limitation, we propose \textbfTree-\textbfof-\textbfExperience (ToE), a structured experience-management framework that aligns experience organization with the hierarchical reasoning process of LLM agents. Specifically, ToE organizes the experience into a shared tree of analytical perspectives and reasoning paths, whose reliability is calibrated through environmental outcomes to support systematic updating, transfer, and efficient retrieval. The experimental results on \textscGame of 24 and \textscFinEvolveBench show that ToE substantially improves both problem-solving performance and efficiency. On \textscGame of 24, ToE achieves a 31.4% relative improvement in accuracy over the experience-free ToT baseline. On \textscFinEvolveBench, ToE improves tsIC by an average of 41.24% over the experience-free pipeline across 12 evaluation settings, whereas conventional experience-management methods often underperform experience-free baselines.
[NLP-13] Dont Scroll Back: Missing-Evidence Memory for Streaming Dialogue Summarization
【速读】： 该论文旨在解决现代对话平台中用户频繁需要对近期对话进行摘要，但当前窗口上下文信息不足、难以独立理解的问题。其核心挑战在于：在固定记忆预算下，如何从无界历史中选择性地保留关键证据，以补全当前对话窗口所隐含的缺失背景。解决方案的关键在于提出ReMEMBER框架——一种基于未解决窗口依赖关系的检索条件机制，并在固定预算内将检索到的内容精炼为高信息密度的记忆单元，从而提升对关键证据的召回率与上下文断点的补全能力。实验表明，在长达160K token的历史对话上，ReMEMBER显著优于现有基线方法，在记忆召回和断点填补完整性方面表现更优。
链接: https://arxiv.org/abs/2608.09043
作者: Hyangsuk Min,Hwanjun Song
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注: 36 pages, 17 figures, 10 tables
Abstract:Users of modern platforms repeatedly need summaries of recent dialogue, but the window rarely contains enough context to be interpreted on its own. We formalize this setting as streaming dialogue summarization, where a system must summarize a current window using selective memory from an unbounded history under a fixed budget. We show that the central challenge is not how much history is accessed, but whether memory recovers the evidence that the current window presupposes. We construct a benchmark and evaluation protocol that separately assesses whether memory contains gap-resolving evidence and whether the generated summary reflects it. We propose ReMEMBER, a missing-evidence memory framework that conditions retrieval on unresolved window dependencies and refines retrieved chunks into evidence-dense memory under a fixed budget. Experiments on dialogues with histories up to 160K tokens show that ReMEMBER improves memory recall and gap-resolution completeness over memory construction baselines under the same budget.
[NLP-14] PolicyKG: An Agent ic LLM Pipeline for Translating Institutional Policies into SHACL Knowledge Graphs
【速读】： 该论文旨在解决机构政策（institutional policies）以自然语言表述，而合规性检查系统需要机器可读的约束条件之间的语义鸿沟问题。现有方法依赖人工进行规则提取与形式化，效率低且易出错。为此，论文提出PolicyKG——一个基于大语言模型（LLM）的智能体式流水线，其核心解决方案是通过四阶段的LangGraph状态机实现从政策文本到SHACL（Shapes Constraint Language）约束的自动化转换。其中最关键的部分是“语料适配器”（Corpus Adapter），它是一个基于YAML的词汇注册表，将大语言模型生成的谓词锚定至目标本体（ontology），从而实现领域可迁移性：只需更换注册表即可适配新领域，无需重新训练模型。在亚洲理工学院政策与程序语料库上的实验表明，该方法在德奥尼提克逻辑（deontic logic）分类上达到86.9%准确率（Cohen’s kappa = 0.709），SHACL形状正确性F1达0.866；绝大多数规则（79.2%）可通过一阶逻辑（FOL）路径处理，其余则通过自然语言直接转为SHACL的降级路径。审计显示高阶逻辑（HOL）规则比例不超过0.67%的95%置信上限，验证了该框架对制度性政策形式化的有效性。此外，通过切换注册表实现对GDPR等新领域的适配，显著提升了属性对齐度，证明了该架构的可扩展性与模块化设计优势。
链接: https://arxiv.org/abs/2608.09028
作者: Ponkrit Kaewsawee,Chaklam Silpasuwanchai,Chutiporn Anutariya
机构: 未知
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Databases (cs.DB); Logic in Computer Science (cs.LO)
备注: 23 pages, 3 figures, 4 tables. Under review at IJCKG 2026, Bangkok, Thailand
Abstract:Institutional policies stay in natural language while the systems that check compliance demand machine-readable constraints. Bridging that gap is still done by hand. PolicyKG closes the loop. It is an LLM pipeline that reads a policy PDF, classifies each sentence as an obligation, permission, or prohibition, lifts the label into first-order deontic logic, and emits SHACL constraints. Four stages run on a LangGraph state machine with per-stage validators. The piece that matters most is the Corpus Adapter: a YAML vocabulary registry that grounds LLM predicates in a target ontology. Retargeting to a new domain means swapping the registry, not retraining a model. On the Asian Institute of Technology Policies and Procedures corpus (1,663 sentences, 443 rules), PolicyKG reaches 86.9% deontic classification accuracy (Cohen’s kappa = .709). Three annotators independently re-label a 50-item sample and agree at Fleiss’ kappa = .844. SHACL shape correctness on a 69-shape subset is F1 = .866. The FOL path handles 79.2% of rules; the rest go through a direct NL-to-SHACL fallback. We audited every one of the 443 rules for second- or higher-order constructs. An automated regex checklist flagged none, and a first-author pass on the 92 FOL-fallback cases confirmed the same. The exact upper 95% Clopper-Pearson bound on the true HOL rate is 0.67%. This is an audit finding for one corpus, not a proof of FOL sufficiency for institutional policy. Swapping the AIT registry for a GDPR registry raises exact property alignment from 1/15 to 11/15 (Fisher’s exact p .001; Cohen’s h = 1.53). On the LexDeMod lease-contract benchmark (N = 200), Macro F1 drops to .370 because lease English uses “shall be entitled” for permission – exactly the vocabulary mismatch registry swap is meant to fix. Repeated runs produce hash-identical SHACL outputs. Comments: 23 pages, 3 figures, 4 tables. Under review at IJCKG 2026, Bangkok, Thailand Subjects: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Databases (cs.DB); Logic in Computer Science (cs.LO) ACMclasses: I.2.4; I.2.7; H.2.8 Cite as: arXiv:2608.09028 [cs.AI] (or arXiv:2608.09028v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.09028 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Ponkrit Kaewsawee [view email] [v1] Mon, 10 Aug 2026 02:28:57 UTC (146 KB) Full-text links: Access Paper: View a PDF of the paper titled PolicyKG: An Agentic LLM Pipeline for Translating Institutional Policies into SHACL Knowledge Graphs, by Ponkrit Kaewsawee and 2 other authorsView PDFHTML (experimental)TeX Source view license Current browse context: cs.AI prev | next new | recent | 2026-08 Change to browse by: cs cs.CL cs.DB cs.LO References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[NLP-15] ELICITED: EHR-grounded Longitudinal Interactive Conversations for Information-seeking Triage Evaluation and Decision-making
【速读】： 该论文旨在解决急诊科分诊（Emergency Department Triage）中临床信息动态获取与决策过程建模不足的问题。传统分诊评估通常基于静态的临床数据快照，无法反映医护人员在实际诊疗中通过交互式对话逐步识别信息缺口、提出针对性问题并动态更新判断的动态过程。为此，论文提出EHR2Dial-Triage——一个基于MIMIC-IV-ED数据集的代理式对话生成框架与基准评测体系。其核心创新在于构建了具有角色分工和时间顺序约束的分诊对话场景，将每次患者披露的信息精确关联至电子健康记录（Electronic Health Record, EHR）中的对应事件及首次可获取的对话轮次。该框架实现了对信息获取效率、证据利用、五级急诊严重程度指数（Emergency Severity Index, ESI）预测以及面向患者的沟通质量的可控评估，为研究分诊作为临床信息采集、推理与沟通的动态过程提供了结构化实验环境。
链接: https://arxiv.org/abs/2608.09024
作者: Haohao Zhu,Xiaolin Shi,Jiayu Zhou
机构: University of Michigan (密歇根大学); Ellipsis Health (椭圆健康)
类目: Computation and Language (cs.CL)
备注:
Abstract:Emergency-department (ED) triage requires clinicians to rapidly identify patients who need immediate attention, determine who can safely wait, and prioritize limited clinical resources. At presentation, however, information may be limited to a chief complaint and initial vital signs. Clinically important details, including symptom onset and progression, associated symptoms, medical history, and medication use, are often obtained through focused conversation. Effective triage therefore requires clinicians to identify information gaps, ask appropriate follow-up questions, and update their assessment as new evidence becomes available. Most existing ED benchmarks evaluate acuity prediction from a fixed clinical snapshot. Although this formulation measures predictive performance after patient information has been assembled, it does not capture the interactive process through which triage-relevant evidence is elicited and interpreted. Existing medical dialogue datasets support the study of clinical communication, but dialogue statements are not always linked to temporally ordered events in the electronic health record (EHR). We introduce EHR2Dial-Triage, an agentic conversation-generation framework and benchmark grounded in MIMIC-IV-ED. The framework constructs triage conversations under explicit role-based and temporal information boundaries. Each accepted patient disclosure is linked to its supporting EHR event and the first dialogue turn at which it becomes available. EHR2Dial-Triage enables controlled evaluation of information elicitation, evidence use, five-level Emergency Severity Index prediction, and patient-facing communication across models and patient personas. It provides a structured setting for studying conversational triage as a dynamic process of clinical information acquisition, reasoning, and communication.
[NLP-16] Automated Generation of Complexity-Validated Decision Scenarios Using Large Language Models
【速读】： 该论文旨在解决认知决策研究中决策场景人工构建效率低、一致性差且存在偏见的问题。其核心解决方案是提出一种基于大语言模型（LLM）的自动化流水线，能够生成结构化决策场景，并通过融合既有任务复杂性理论的复合框架对场景复杂度进行验证。关键创新在于构建了一个具备严格心理测量学标准的评估体系：在多领域、多层次的4,238个场景上验证后，结果显示模型间一致性极高（组内相关系数ICC=0.997，Kappa=0.971），已知组别效度显著（η²=0.587，所有成对比较p<0.001），因子分析揭示复杂度为首要主成分（因子载荷0.87–0.96），互动性为次要维度（载荷0.34）。尽管文本长度与复杂度存在较强部分相关性（partial r=0.86），影响构念纯度，但不影响层级划分的有效性。模型性能分析表明吞吐量与模板通过率呈负相关（r=-0.967, p=0.007），体现速度-质量权衡，其中Llama 4 Maverick虽生成速度最快（134/分钟），但复杂层级场景产出不足；而DeepSeek Chat V3.2在领域覆盖与模板合规性之间取得更好平衡。整体系统展现出优良的心理测量特性，可实现对决策场景的可靠分层（简单、中等、复杂），为下游人工智能系统认知能力评估提供了可扩展的测量基础设施。
链接: https://arxiv.org/abs/2608.08822
作者: Abdalla Doleh,Toni Somers,Ratna Babu Chinnam
机构: Wayne State University (韦恩州立大学)
类目: Artificial Intelligence (cs.AI); Computation and Language (cs.CL)
备注: 38 pages, 7 figures
Abstract:Cognitive decision-making research depends on diverse scenarios with carefully controlled complexity, yet manual production is slow, inconsistent, and biased. We developed an automated pipeline that uses LLms to generate structured decision scenarios and validates their complexity through a composite framework rooted in established task-complexity theory. We evaluated 4,238 scenarios across multiple domains and complexity tiers. Measurement validation met rigorous psychometric standards. Agreement among five independent model families was nearly perfect, with an intraclass correlation coefficient of 0.997 and a kappa of 0.971. Known-groups validity demonstrated large separation between tiers, with an eta-squared of 0.587 and all pairwise comparisons significant at p less than .001. Factor analysis revealed a dominant complexity construct, with loadings between 0.87 and 0.96 across three frameworks, while interactivity formed a weaker secondary dimension at 0.34. Discriminant validity was limited by a strong relationship between complexity and text length that persisted after controlling for tier, yielding a partial correlation of 0.86. This constrains construct purity but does not undermine the instrument’s tier-grading function. Model analyses showed a negative association between throughput and schema pass rate (r = -0.967, p = .007, n = 5), suggesting a speed-quality trade-off, though largely driven by one high-throughput model. Llama 4 Maverick generated scenarios fastest at 134 per minute versus 25 for DeepSeek Chat V3.2, but underproduced complex-tier scenarios, whereas DeepSeek Chat V3.2 balanced domain coverage with high schema compliance. The system demonstrated strong psychometric properties, enabling reliable classification into Simple, Moderate, and Complex tiers and providing the measurement infrastructure needed for downstream cognitive assessment of AI systems
[NLP-17] vatron-Elastic: A Unified Abstraction for Training Elastic Retrievers and Rerankers
【速读】： 该论文旨在解决生成式检索系统中单一模型规模难以兼顾灵活性与效率的问题：不同应用场景对模型速度、索引大小的需求各异，且最优权衡随负载动态变化。现有方法（如减少层数、提前退出、嵌入向量压缩等）虽可分别实现模型轻量化，但各自独立实现，需不同的代码与训练配置，难以融合或适配新模型。本文提出一个统一的抽象框架——通过一个简洁的对象表示模型可运行的所有尺寸，并以简短调度表定义需训练的尺寸组合，使一次训练即可产出覆盖多尺寸的单一检查点，部署时用户可任意选择所需尺寸。该抽象兼容检索器与重排序器、编码器与解码器架构，仅通过配置变更即可支持新骨干网络，无需修改建模代码；此前的马特约什卡嵌入（Matryoshka embeddings）、提前退出（early exit）、二维马特约什卡（2D Matryoshka）及逐层词元压缩等方法均成为其特例。此外，该框架还支持联合训练多种词元压缩比率的马特约什卡-词元压缩（MLTC），实现更高效的资源利用。实验验证表明，基于三种骨干网络和两个任务训练的20个检查点具备平滑的质量曲线，单个检查点的额外开销极低，且控制实验确认了实际推理速度的显著提升。研究团队已开源该框架及所有检查点，为构建弹性检索系统提供基础资源。
链接: https://arxiv.org/abs/2608.08809
作者: Yu Wang,Shengyao Zhuang,Xueguang Ma,Zongyu Wu,Jimmy Lin,Vivek Srikumar,Zhichao Xu
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:A single model scale challenges the flexibility of a production retrieval system: some settings need it faster, others need a smaller index, and the right trade-off changes with the workload. In the context of information retrieval (IR), a transformer-based model can be made smaller in three ways—using fewer layers, passing fewer tokens through the upper layers, or producing a shorter embedding—and each way saves a different compute resource. These options have been studied one at a time, each as its own method with its own code and training setup, which makes them hard to combine or adapt to a new model. We present~\ours to bring all three under one simple abstraction: a single object names any size the model can run at, and a short schedule lists the sizes to train. Training then produces one checkpoint that serves all of those sizes, and at deployment the user picks any of them. The same abstraction covers both retrievers and rerankers and both encoder and decoder models, as it works through interfaces that Hugging Face transformers already expose; a new backbone is a configuration change, not new modeling code. Prior methods—Matryoshka embeddings, early exit, 2D~Matryoshka (e.g., Starbucks), and layerwise token compression—become special cases of our unified abstraction. The same interface also enables Matryoshka~LTC (MLTC), which jointly trains several token-compression ratios in one retriever checkpoint. To validate our framework, we train 20 checkpoints across three backbones and two tasks: the quality curves are smooth, one checkpoint costs little over a model trained for a single size, and a controlled study confirms the wallclock speedups. We release the framework and all checkpoints as a resource for building elastic retrieval systems.
[NLP-18] IDRAAK: From Multi-Agent NLP to Few-Shot Prompting for Semantic Drift Detection in Technical Requirements
【速读】： 该论文旨在解决技术需求在跨语言翻译过程中可能出现的语义漂移（semantic drift）问题，尤其是数值约束、极性、情态等关键语义信息的失真。其核心解决方案是提出一种语言无关的语义需求表示（Semantic Requirement Representation, SRR），并基于此构建可解释的检测框架IDRAAK。该框架的关键在于利用单一大语言模型（LLM）调用结合六种少样本示例（few-shot prompting）即可实现高效精准的语义漂移检测，在890个合成扰动数据上达到马修斯相关系数（MCC=0.888）和F1分数（F1=0.983），显著优于多阶段结构化方法与复杂多智能体验证方案。研究还揭示：增加代理复杂度并不必然提升性能，而简单的少样本提示策略在技术需求场景下已具备强大且高效的检测能力，同时通过后处理Platt校准进一步优化置信度可靠性。
链接: https://arxiv.org/abs/2608.08801
作者: Shiva Ahir
机构: 未知
类目: Computation and Language (cs.CL); Hardware Architecture (cs.AR); Emerging Technologies (cs.ET)
备注:
Abstract:Translating technical requirements across languages can introduce semantic drift, altering numerical constraints, polarities, modalities, or other specification-critical meaning. IDRAAK is presented as an interpretable framework for detecting such drift using a language-independent Semantic Requirement Representation (SRR), with six detection workflows evaluated, ranging from deterministic comparison to multi-agent verification and few-shot prompting. On 890 synthetic perturbations across 300 requirements from 10 engineering domains, a single LLM call with six few-shot examples achieves MCC=0.888 and F1=0.983, outperforming the evaluated structured and multi-stage alternatives. Further evaluation on PAWS-X (805 pairs, 5 languages) and XNLI (700 pairs, 7 languages) exposes complementary strengths and limitations of structured and LLM-based approaches. Deterministic SRR comparison performs strongly on technical requirements (F1=0.898) but poorly on general-domain text (F1=0.012), while structured evidence improves performance on adversarial paraphrases. Post-hoc Platt scaling further improves confidence calibration. The results demonstrate that increased agentic complexity does not necessarily improve semantic-drift detection and that simple few-shot prompting can provide a strong and efficient alternative.
[NLP-19] Instability of LLM Pre-Pretraining: It Doesnt Always Help. An Investigation on Multiple Languages
【速读】： 该论文旨在验证在人工语言上进行大语言模型（LLM）的预训练（即“预预训练”）是否能够显著提升分词效率，从而减少达到特定性能水平所需的训练分词量。研究通过在涵盖四大语系的多个自然语言上开展实验，采用两种不同的分词器与不同规模的模型，系统评估了该方法的有效性。其解决方案的关键在于：通过引入基于形式语言理论的128-Dyck人工语言进行预训练，并分析其对不同语言在句长、形态丰富度及依存句法树结构特征（如树深、子节点数、交叉依存数量）方面的分词效率影响。研究发现，尽管部分条件下可观察到稳定的分词效率增益，但整体结果高度依赖于实验设置与随机种子的选择，表明先前报告的33%效率提升具有不稳定性。因此，论文强调应至少对部分实验进行多次重复训练，以确保方法的可靠性，避免社区采纳不可复现的技术路径。
链接: https://arxiv.org/abs/2608.08800
作者: Sofiia Riazhskykh,Nam Luu,Ondřej Bojar
机构: Charles University, Faculty of Mathematics and Physics (查尔斯大学数学与物理学院)
类目: Computation and Language (cs.CL)
备注:
Abstract:Pretraining LLMs on artificial languages (“pre-pretraining”) is a technique that could reportedly increase token efficiency by 33%, i.e., save up to 33% of training tokens needed to reach a certain performance. We validate this prior result for English on a larger set of natural languages across four language families, using two different tokenizers and varying model sizes. We also relate the observed gains (or losses) in token efficiency to quantified linguistic properties of the languages, such as sentence length, morphological richness, and features of dependency syntactic trees (tree depth, number of children, number of crossing dependencies). Our empirical results indicate that the reported gains depend heavily on the experiment setup and the choice of random seed, although we can confirm the trend of stable gains with 128-Dyck pretraining of small models with the Llama tokenizer for most of the examined languages. On a general note, we argue that multiple training runs should be carried out at least for a subset of experiments to avoid the community adopting unstable approaches.
[NLP-20] oward Metacognitive One-Shot Indirect Prompt Injection: Strategy Abstraction Via Outcome-Conditioned Reflection
【速读】： 该论文旨在解决工具型大语言模型（Tool-using Large Language Model, LLM）代理在面对间接提示注入（Indirect Prompt Injection, IPI）攻击时的脆弱性问题，尤其关注现实场景中攻击者仅拥有一次与未知目标代理交互机会的挑战。传统自适应攻击依赖于反复查询和迭代优化目标代理，但在实际环境中难以实现。本文提出SAVOR（Strategy Abstraction Via Outcome-Conditioned Reflection），其核心解决方案是将攻击适应过程从测试阶段的迭代查询转变为离线策略蒸馏。SAVOR通过在独立训练环境收集的成功与失败轨迹进行结果条件反射（outcome-conditioned reflection），验证上下文条件化的候选策略，并迭代整合形成可复用的策略记忆库。在测试阶段，冻结的记忆库指导生成针对未见目标的单次有效载荷，仅需一次目标代理查询且无需反馈。实验表明，在两个基准测试和三个受害模型上，SAVOR在全部六组设置中均达到最高平均攻击成功率，相较最强基线提升2.5至11.8个百分点，且在新引入的OpenClaw-IPI可执行基准上比无策略学习的同通道攻击高出28.6个百分点；此外，基于一种防御机制学习的策略具有良好的跨防御迁移能力。
链接: https://arxiv.org/abs/2608.08795
作者: Sihan Hou,Xinmeng Hou,Zhijun Zhang,Zehao Wang,Xuhong Ren,Sibo Qin,Kuntharrgyal Khysru,Qing Guo
机构: 1. Institute of Automation, Chinese Academy of Sciences (中国科学院自动化研究所); 2. School of Artificial Intelligence, University of Chinese Academy of Sciences (中国科学院大学人工智能学院); 3. Department of Computer Science and Technology, Tsinghua University (清华大学计算机科学与技术系); 4. School of Computing, University of Utah (犹他大学计算机学院)
类目: Cryptography and Security (cs.CR); Computation and Language (cs.CL)
备注:
Abstract:Tool-using large language model (LLM) agents are vulnerable to indirect prompt injection (IPI), in which malicious instructions embedded in external observations manipulate subsequent agent decisions and actions. Most existing adaptive attacks rely on repeatedly querying and refining against the target agent, whereas realistic attackers may have only a single opportunity to interact with an unknown target agent. We propose SAVOR (Strategy Abstraction Via Outcome-Conditioned Reflection), which shifts attack adaptation from test-time iteration to offline strategy distillation. SAVOR performs outcome-conditioned reflection over successful and failed trajectories collected from disjoint training environments, validates context-conditioned candidate strategies, and iteratively consolidates them into a reusable strategy memory. At test time, the frozen memory guides the generation of a single payload for each unseen target, requiring only one target-agent query and no target-agent feedback. Across two benchmarks and three victim models, SAVOR attains the highest average attack success rate in all six settings, leading the strongest prior attack by 2.5 to 11.8 points and the same injection channel without strategy learning by 23.1 points on Agent Security Bench, which holds out attacker tools, and 28.6 points on OpenClaw-IPI, an executable benchmark we introduce that holds out attack goals and verifies attacks through tool interactions and execution receipts. A memory learned under one defense also transfers to another.
[NLP-21] Evidence-Calibrated Runtime Reconstruction for Agent Skills Across Heterogeneous Coding Agents
【速读】： 该论文旨在解决生成式 AI（Generative AI）代理在使用工具时，因缺乏对技能（Skill）生命周期状态的准确可观测性而导致的故障边界识别困难问题。现有基于会话、模型或工具中心的追踪方法无法有效刻画技能从发现到激活再到结果验证的完整流程，导致诸如“技能可被发现但无法激活”“已激活但无指令”或“看似成功但无独立验证结果”等语义不一致现象。其核心解决方案是提出技能运行时智能（Skill Runtime Intelligence），一种被动式运行时情报系统，能够在异构执行环境（harnesses）中重建受支持的技能生命周期阶段，并将未支持的阶段标记为未知，从而保持语义完整性。该系统通过“运行全景图（Run Panorama）”将不可变事件、确定性关系、推断诊断与可控结果四类信息以四个证据等级进行分离，实现细粒度的状态追溯；同时支持可选的追踪导入及OTLP/HTTP导出，兼容现有可观测性基础设施。实验表明，在六组冻结仓库配置、三类编码代理及七种清洁或注入故障条件下，126次执行均保留源工作树且一一对应原始会话，但不同适配器暴露三种截然不同的语义行为：无技能运行、完整运行但无失败事件、或在所有操作失败和清洁会话中均出现失败类似事件。在七模板诊断研究中，语义别名与全景图能定位相同的六个非清洁边界，但在具体行为与状态表现上存在差异；而原始视图在全部18个清洁案例中均误报失败状态，全景图则零误报。进一步验证显示，已知规则图符合126/126冻结契约，而另一模型仅完成228/378调用，揭示了事件存在性并非边界保真性的充分条件，复合精确度分数会掩盖本质错误差异，且模型解释不应覆盖确定性事实。
链接: https://arxiv.org/abs/2608.08793
作者: Xueping Gao
机构: 未知
类目: Computation and Language (cs.CL)
备注: 17 pages, 1 figure, 6 tables. Submitted to PROFES 2026. Code and artifacts: this https URL
Abstract:Agent Skills package reusable instructions and assets for tool-using language-model agents. Progressive loading creates failure boundaries poorly represented by session-, model-, or tool-centric traces: a Skill can be discovered but not activated, activated without instructions, or appear successful without an independently verified outcome. We present Skill Runtime Intelligence, a passive runtime-intelligence system that reconstructs supported Skill-lifecycle stages across heterogeneous harnesses while preserving unsupported stages as unknown. Its Run Panorama separates immutable events, deterministic relations, inferred diagnoses, and controlled outcomes with four evidence grades; optional trace import and OTLP/HTTP export support existing observability deployments. Across six frozen repository profiles, three coding agents, and seven clean or fault-injected conditions, all 126 executions preserve source worktrees and each correlates to exactly one source session. Yet adapters expose three distinct semantics: no Skill runs; complete runs but no failure-like events; or failure-like events in every operational-failure and clean session. In a seven-template diagnostic study, semantic aliases and Panorama localize the same six non-clean boundaries but differ in exact/status behavior; both Raw views emit a failure status on all 18 clean cases, while Panorama emits none. A known-rule graph conforms to 126/126 frozen contracts, whereas a second model completes only 228/378 calls. These observations motivate executable adapter qualification and show that event presence is not boundary fidelity, composite exact scores mask distinct errors, and model explanations must not overwrite deterministic facts. Comments: 17 pages, 1 figure, 6 tables. Submitted to PROFES 2026. Code and artifacts: this https URL Subjects: Computation and Language (cs.CL) Cite as: arXiv:2608.08793 [cs.CL] (or arXiv:2608.08793v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.08793 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[NLP-22] Unsure but Certain: Uncovering the Representation-Confidence Gap in Diffusion Language Models
【速读】： 该论文旨在解决生成式 AI（Generative AI）在输入噪声环境下可靠性下降的问题，特别是扩散语言模型（Diffusion Language Models）虽具备较强的上下文理解能力，但在面对输入噪声时表现出“表面高置信度”与“实际准确性下降”的严重不一致现象。其核心问题在于：尽管模型内部能精准检测文本错误，但外部输出的置信度分数却未能反映这一内部信号，导致置信度维持在高位而排名能力退化至随机水平，形成所谓的“表示置信度差距”（Representation Confidence Gap）。解决方案的关键在于：虽然传统的置信度校准、输入级错误信号或训练匹配等方法均无法恢复正确的答案排序，但研究发现，用于评估答案正确性的关键信息仍保留在模型的隐藏状态中。为此，作者提出一种轻量级提取工具，从隐藏状态中挖掘该信号以重构排序，该方法无需微调基础模型、不增加生成步骤，保持模型冻结，从而高效实现排序性能的提升。此工作揭示，在存在噪声的场景下，置信度的可靠性比整体准确率更具决定性意义。
链接: https://arxiv.org/abs/2608.08791
作者: Saurabh Yadav,Badri Narayana Patro,Vijay Srinivas Agneeswaran
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Diffusion language models use broad context to create text, suggesting they might handle input noise better than standard models. Testing reveals this is only partially true. Internally, diffusion models detect text errors highly accurately. Externally, their reported certainty ignores this signal. As accuracy drops due to noise, confidence stays near its maximum and the ability to correctly rank answers degrades toward random chance. We call this mismatch the representation confidence gap. The visible concentration of high certainty scores is a misleading surface symptom. Standard math adjustments remove this concentration but fail to fix the underlying loss of ranking order. This ranking deficit favors standard models under noisy conditions and resists common remedies. Matching training recovers accuracy but not ranking, while score recalibration and input level error signals cannot reorder the final answers. However, the information needed to properly evaluate an answer survives in the hidden states. A lightweight extraction tool uses this signal to improve ranking. This approach is highly efficient because it leaves the base model completely frozen and requires zero additional text generation steps. We present this tool to prove the signal exists, while clearly noting its limits. Ultimately, certainty reliability is a more pressing limit than overall accuracy under noisy conditions.
[NLP-23] OmnilingualGAIA2: Evaluating the Multilingual Gap in Frontier AI Agents
【速读】： 该论文旨在解决当前生成式AI代理（Agentic AI）评估基准在语言多样性上的严重缺失问题，即现有评估体系几乎完全基于英语，而实际部署的AI代理需服务于多语言用户群体，因此亟需验证其跨语言能力是否具备泛化性。其核心解决方案是提出OmnilingualGAIA2——一个通过机器翻译扩展并辅以部分人工专家校验的多语言版GAIA2基准，覆盖十种目标语言（涵盖五种书写系统），并配套本地化、人工校准的多语言验证器。研究发现，所有测试代理均存在普遍的跨语言性能差距（pass@3指标下降8.8–18.4分），且该差距在不同代理间不对称，主要集中在工具编排（tool-orchestration）而非量化推理能力上，且不随模型规模扩大而收敛。分层错误归因分析表明，该差距主要由模型自身能力局限驱动（占比55%），翻译污染仅贡献6.4%的上限影响。进一步的人工语言学分析揭示，非拉丁字母语言中的形态线索丢失与语义歧义加剧是导致失败的关键机制。研究强调，多语言代理评估应成为全球部署智能体的标准报告流程组成部分。
链接: https://arxiv.org/abs/2608.08775
作者: Andrea Caciolai,Pere-Lluís Huguet Cabot,Chierh Cheng,Albert Ventayol-Boada,Gabriel Mejia Gonzalez,Christophe Ropers,Lucas Bandarkar,Sebastian Ruder,Darlene Sakakihara,Elliot Yun,Pierre Andrews,Grégoire Mialon,Romain Froger,Marta R. Costa-jussà
机构: 未知
类目: Computation and Language (cs.CL)
备注:
Abstract:Agentic benchmarks aim to measure how well AI agents plan, search, execute, and recover within realistic multi-tool environments, but they are almost exclusively in English. As AI agents are globally deployed to a linguistically diverse user base, whether agentic competence measured in English transfers to other languages remains an open question. We introduce OmnilingualGAIA2, a machine-translated expansion (with partial human- expert validation) of the GAIA2 agentic benchmark, covering ten target languages spanning five writing systems, paired with a localised and human-calibrated multilingual verifier. Evaluating seven frontier and open-weight agents, we find a universal cross-lingual gap of 8.8-18.4 pass@3 points that is agent-asymmetric in magnitude, concentrates on tool-orchestration rather than quantitative reasoning, and does not close with model scale. A stratified error attribution decomposes the gap as predominantly model-driven (55%), with a bounded translation-contamination floor of only 6.4% of scenario-language pairs. Human-expert linguistic analysis further identifies morphological cue loss and amplified ambiguity as the primary failure mechanisms in non-Latin-script languages. Our results argue that multilingual agentic evaluation must become a standard part of the reporting protocol for globally deployed agents.
[NLP-24] Multilingual Emotion Neurons in Large Audio-Language Models
【速读】： 该论文旨在解决大音频-语言模型（Large Audio-Language Models, LALMs）在跨语言情感表达中是否存在语言特异性关联或语言无关表征的核心问题。其核心挑战在于：现有方法依赖单语情感神经元的独立识别，导致跨语言情感神经元（Multilingual Emotion Neurons, MLENs）的可迁移性差且重叠率低，难以实现统一的情感控制。论文提出的关键解决方案是一致性正则化融合（Consistency-Regularized Fusion, CR-Fusion），通过整合多语言联合证据，从跨语言数据中系统性地识别出具有稳定情感选择性和一致因果效应的神经元单元。实验表明，基于CR-Fusion识别的MLENs在零样本和低资源场景下均表现出更强的精确性与跨语言可迁移性；进一步的留一法消融分析揭示了非对称转移现象——各语言（包括低资源语言）贡献独特信息，而后者尤其受益于跨语言知识迁移。本研究首次提供了LALMs跨语言情感编码的因果性、神经元层级解释，确立了多语言神经元识别作为理解跨语言情感行为的有效机制。
链接: https://arxiv.org/abs/2608.08772
作者: Xiutian Zhao,Philipp Koehn,Björn Schuller,Berrak Sisman
机构: Johns Hopkins University (约翰霍普金斯大学); Imperial College London (帝国理工学院)
类目: Computation and Language (cs.CL); Audio and Speech Processing (eess.AS)
备注:
Abstract:Emotion is central to human communication, and its expression varies across languages. Large audio-language models (LALMs) achieve strong performance on multilingual speech tasks, yet it remains unclear whether they encode emotion through language-specific correlations or language-agnostic representations. We present the first neuron-level interpretability study of this question. We define Multilingual Emotion Neurons (MLENs) as functional units exhibiting stable emotional selectivity and aligned causal effects across languages, and introduce Consistency-Regularized Fusion (CR-Fusion) to identify them. Across four modern LALMs and 12 typologically diverse languages, emotion-sensitive neurons identified independently per language show minimal overlap, and additional monolingual identification data saturates quickly without isolating more transferable units, motivating identification from pooled cross-lingual evidence. Causal interventions demonstrate that MLENs identified by CR-Fusion provide more precise and transferable affective control than monolingual neuron sets in both zero-shot and low-resource settings. Leave-one-out ablations further reveal asymmetric transfer: individual identification languages, including low-resource ones, contribute non-redundant evidence, while several low-resource languages benefit most from the resulting cross-lingual transfer. Together, our findings provide the first causal, neuron-level account of how LALMs encode emotion across languages, and establish multilingual neuron identification as an effective mechanism for understanding cross-lingual affective behavior.
[NLP-25] Can We Optimize the Performance-Carbon Emission Break-Even Point?: The Quest for Greener LLM s
【速读】： 该论文旨在解决大语言模型（Large Language Model, LLM）在推理阶段持续累积碳足迹的问题，而现有大多数效率优化方法主要聚焦于预训练规模或后期压缩，未能有效控制推理过程中的碳排放。其核心解决方案是将一个经过校准的、可微分的能量代理（energy surrogate）融入微调目标函数中，通过引入与模型参数量、计算量（FLOP）和内存占用相关的线性能量代理，并基于实际硬件能耗测量进行拟合，实现对推理阶段碳排放的量化建模。研究提出一种联合损失机制，包含每模型的碳排放参数，以期在不增加额外碳成本的前提下提升任务准确率，达到“零或近零碳成本”的性能增益。实验针对Gemma-2 2B、Llama-3.1 8B和Qwen-2.5 14B三个架构差异显著的模型，在抽象代数、哲学和形式逻辑三个MMLU子任务上评估推理F1分数与二氧化碳排放。结果表明，碳项在不同任务中表现出双重作用：在某些任务中作为有害干扰，而在另一些任务中则起到有益正则化效果，说明碳感知微调具有模型与任务依赖性的“盈亏平衡区域”。因此，该方法被定位为一种轻量级、可直接嵌入的正则化策略，具备潜在的低碳优化价值，目前仍处于持续研究中，代码库即将开源。
链接: https://arxiv.org/abs/2608.08744
作者: Sourav Das,Tanmay Joshi,Kripabandhu Ghosh
机构: IIIT Kalyani; BITS Pilani Goa; IISER Kolkata
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Computational Engineering, Finance, and Science (cs.CE); Emerging Technologies (cs.ET); Machine Learning (cs.LG)
备注: 13 Pages, 6 Figures, Submitted to ARR Cycle
Abstract:The carbon footprint of any deployed Large Language Model (LLM) accumulates during inference, where repeated use of the model substantially exceeds the one-time cost of fine-tuning. Yet most efficiency interventions target either pre-training scale or post-hoc compression. We ask whether folding a calibrated, differentiable energy surrogate into the fine-tuning objective can produce inference behavior that gains task accuracy at zero or near-zero carbon cost, a break-even configuration. We propose a joint loss mechanism with a per-model carbon-emission parameter, a linear surrogate over parameter norm, FLOP proxy, and a memory proxy, fit from on-hardware energy profiling. We fine-tune three architecturally distinct families: Gemma-2 2B, Llama-3.1 8B, and Qwen-2.5 14B, and evaluate inference F1 and CO _2 emissions on three MMLU subjects: abstract algebra, philosophy, and formal logic. We discover from several outcomes that the carbon term behaves as either harmful interference or beneficial regularization depending on the task structure. We position calibrated carbon-aware fine-tuning as a lightweight, drop-in regularizer with a non-empty but model and task-dependent break-even region. This is an ongoing work, and we will release our codebase soon.
[NLP-26] Scaling Inherently Interpretable Language Models
【速读】： 该论文旨在解决传统语言模型中可解释性（interpretability）与模型能力之间存在权衡的问题，即通常将模型训练为“黑箱”系统，再通过事后解释方法进行分析，而这些解释方法的可靠性难以验证。其核心解决方案是将可解释性作为训练过程中的一个显式约束，与语言建模目标共同优化，而非事后补救。关键在于：在训练阶段就引入可解释性机制，使模型在规模扩大时不仅能力提升，其内部表征也趋向于更解耦（disentangled）且与人类可理解的概念对齐。研究通过构建名为Steerling-8B的扩散语言模型（diffusion language model），采用因果注意力掩码，在不同规模计算条件下均实现了可解释性随模型能力同步增长的现象。该模型能够为生成的输出追溯到相关输入、人类可理解的概念及训练数据，从而支持闭环干预——基于概念或特征归因诊断问题，检索相似训练样本，并通过概念引导（concept steering）修正行为，无需重新训练。实验表明，即使在远低于对比模型的计算量下，Steerling-8B仍保持竞争力，揭示了一种新的规模化范式：可解释性可被设计进训练流程，并随着规模扩展而增强。
链接: https://arxiv.org/abs/2608.07594
作者: Guide Labs Team,Andreas Madsen,Aya Abdelsalam Ismail,Giang Nguyen,Isaac Plant,Muawiz Chaudhary,Nathaniel Monson,Saqib Azim,Zhichen Guo,Julius Adebayo
机构: 未知
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI)
备注:
Abstract:Interpretability is often treated as a tax on capability: language models are trained as opaque systems, then explained after the fact, with methods whose reliability is difficult to establish. In this work, we challenge this premise. Rather than reverse-engineering a model, we make interpretability a constraint of the training pipeline, optimized alongside the language modeling objective. Across three orders of magnitude of compute, on both autoregressive and diffusion language models, interpretability scales with capability rather than against it. Surprisingly, model representations become more disentangled and aligned with human-understandable concepts with scale. We instantiate the training-time recipe with Steerling-8B, a diffusion language model with a causal attention mask. For any group of generated tokens, Steerling-8B attributes the output to relevant input tokens, human-understandable concepts, and training data. This enables closed-loop intervention: diagnose an output through its concept or feature attribution, retrieve similar training data, and correct the behavior through concept steering without retraining. Steerling-8B remains competitive with open peer models trained on substantially 2-16x more compute, suggesting a different scaling paradigm: interpretability can be designed into training, and it improves with scale. Subjects: Computation and Language (cs.CL); Artificial Intelligence (cs.AI) Cite as: arXiv:2608.07594 [cs.CL] (or arXiv:2608.07594v1 [cs.CL] for this version) https://doi.org/10.48550/arXiv.2608.07594 Focus to learn more arXiv-issued DOI via DataCite
信息检索
[IR-0] MetaStrategy: Generative Ranking with Executable LLM Strategies
链接: https://arxiv.org/abs/2608.09440
作者: Chengyu Lai,Jiuning Lin,Zhibo Xiao,Xiaodong Zhu,Ruiquan Lan,Bin Zhang,Zihong Huang,Wendong Zhang,Chuxin Chen,Yinjiang Cai,Shuai Zhong,Lingqing Zhang,Dimin Wang,Jialin Zhu,Han Zhu
类目: Information Retrieval (cs.IR)
备注:
Abstract:Industrial recommender systems rank heterogeneous content under coupled user, business, commercial, and experience objectives. Existing generative ranking methods typically construct item sequences directly, making them difficult to integrate with mature predictive models, operational rules, and field-level guardrails. We present MetaStrategy, a framework that instead generates a structured, executable ranking strategy. Conditioned on request context, a large language model (LLM) policy emits a typed JSON bundle controlling objective weights, content and category preferences, experience constraints, and position policies. A deterministic validator and compiler instantiate an isolated Generator that competes atomically with incumbents under the list-level Evaluator of the Generator-Evaluator (GE) architecture. We train the policy in a production-path replay environment that re-executes logged requests through the current re-ranking stack without user exposure. The method combines selection, relative-rank, and baseline-lift rewards, a self-competitive curriculum that feeds frequent strategies back as competitors, and Evaluator-routed reward-augmented on-policy distillation that transfers complementary 4B-parameter Teachers into a compact 0.8B-parameter Student. We deploy MetaStrategy in Taobao Homepage Guess You Like through diff-triggered nearline generation; LLM inference remains outside synchronous ranking, with no observable increase in response time (RT). In a seven-day user-randomized online A/B test, MetaStrategy wins 27.93% of treatment-side GE calls and significantly improves click page views (click PV) by 2.11%, item-detail page views (IPV) by 3.12%, and transaction amount by 2.83%.
[IR-1] DREAM Technical Report
链接: https://arxiv.org/abs/2608.09408
作者: Bin Zhang,Bowen Zheng,Chao Yi,Chengyu Lai,Dian Chen,Dimin Wang,Gaoyang Guo,Jialin Zhu,Jian Wu,Jing Yu,Jiuning Lin,Lingqing Zhang,Lingyun Zheng,Mao Zhang,Mingming Pan,Ruiquan Lan,Shuai Zhong,Wen Chen,Wendong Zhang,Xiaodong Zhu,Xuan Chen,Xunke Xi,Yifan Lu,Yiheng Wang,Yue Zeng,Yujie Luo,Yuning Jiang,Zhe Hu,Zhibo Xiao,Zihong Huang,Binbin Cao,Bo Zheng,Danning Wang,Dixuan Wang,Ge Fan,Haixia Wu,Han Zhu,Hao Fang,Haoming Chen,Huiping Chu,Jian Wang,Jianjun Wu,Jiawei Wu,Jiaxin Yu,Jingwen Liu,Jinzhe Shan,Kai Meng,Kai Zhang,Keqin Xu,Kewei Zhu,Lang Tian,Leihui Chen,Li Chen,Licheng Xu,Lide Xiao,Ruitong Zhang,Shiyao Peng,Silu Zhou,Tao Wang,Wei Shi,Wenjun Yang,Xiang Chen,Xiang Gao,Xiao Ren,Xu Liu,Xuwen Wang,Yang Li,Yeqiu Yang,Yi Hu,Yinnan Song,Yuan Liu,Yunqi Gao,Zhiliang Huang,Zhujin Gao,Zongyuan Wu
类目: Information Retrieval (cs.IR)
备注: Technical Report
Abstract:Industrial recommender systems commonly use cascaded retrieval, ranking, and re-ranking pipelines. Although efficient, these pipelines fragment information and objectives across modules, rely on rigid rules, and have limited awareness of real-time intent, leaving session-level shifts among browsing, comparison, and purchase insufficiently addressed. We present DREAM (Developing Recommender Engine with Agentic Methods), an autonomous optimization control architecture that adds a perception-aware, orchestrable, and auditable policy layer atop existing pipelines without replacing them. DREAM has two core components. First, a three-tier Intent Engine fuses on-device signals into structured L0/L1/L2 intent representations; its edge-cloud trigger chain reduces reporting volume to approximately 8.7%. Second, a Meta Engine uses a MetaModel for layered M1-to-M2-to-M3 reasoning: intent summarization, strategy planning informed by Strategy Memory, and parameter translation. It dispatches the resulting parameters through a unified outlet with safety guardrails. A Reward Dual Loop continuously optimizes both components by combining offline simulation for strategy-space exploration with online feedback for outcome calibration, forming a cycle of generation, execution, evaluation, and experience accumulation. Large-scale A/B tests on Taobao’s homepage feed show that re-ranking control alone improves IPV by 2.06%, Core IPV by 2.39%, and GMV by 0.88%. Extending control to fine ranking raises these gains to 2.71%, 3.06%, and 1.31%, respectively, while consistently improving PV by more than 1%. These gains require neither replacement of pipeline models nor compromise of serving stability, supporting agentic meta-control as a viable paradigm for industrial recommendation.
[IR-2] mporal Misgrounding in Legal RAG : A Versioned-Corpus Benchmark for French Tax Law ICML2026 ISCA
链接: https://arxiv.org/abs/2608.09393
作者: Rose Cymbler,Daniel Guez,Laurent Fabre
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: 13 pages, 1 figure, 4 tables. Accepted at the ICML 2026 Workshop on AI for Law (AI4Law), Seoul. Code and data: this https URL
Abstract:We identify and quantify temporal misgrounding: the systematic retrieval and citation of the currently in-force version of a legal article when the applicable version is an earlier or future one. Standard legal RAG treats the corpus as static; we argue legal question answering is a temporally-indexed retrieval problem. We introduce FiscalQA Pro, pairing a versioned corpus of 32,436 article-versions of the French tax code (93 years, 1938-2031) with an all-model-hard temporal-reasoning track: 209 scored, expert-reviewed questions across 33 CGI articles (221 released; twelve flagged out of the answerable scope). At selection time, no evaluated model recovered its date-applicable answer closed-book in any of four sampling draws, and the currently in-force text lacks the gold value for all but one of the scored questions. Answers are scored deterministically via atomic ground-truth “nuggets” (regex and numeric-with-tolerance), never LLM-as-judge: an LLM judge would inherit the temporal bias it is meant to score. Across eleven models (five frontier closed-API systems plus Gemini 2.5 Pro as a substitute entry, and five open-weight), parametric knowledge yields 3.0% mean strict accuracy and RAG over a static current-version corpus 2.7%. Static RAG retrieves the date-applicable version 0% of the time, confidently citing a real but inapplicable version. Our end-to-end retriever over a multi-version index, with no oracle, reaches 98.3% mean strict; an oracle-article ablation reaches 99.1%, locating the residual gap in first-stage recall, not version selection. We additionally release a version-aware jurisprudence dataset of 69,208 citation links, together with the corpus, benchmark, model responses, and pipeline code.
[IR-3] RVANNS: Mixed-Precision Indexing and Locality-Aware Graph Traversal on RISC-V
链接: https://arxiv.org/abs/2608.09077
作者: Chengying Huan,Yudong Liu,Jianguo Wang,Lizheng Chen,Renling Yin,Weijia Chen,Ji Qi,Jiageng Yu,Junjie Xu,Jie Zhang,Chen Tian,Yanjun Wu
类目: Information Retrieval (cs.IR)
备注:
Abstract:Approximate nearest neighbor search (ANNS) on CPUs is increasingly constrained by candidate-vector movement and decoding rather than peak arithmetic throughput. Although the RISC-V Vector Extension (RVV) provides vector-length-agnostic execution and LMUL-based register grouping, generic low-precision decoding still incurs conversion overhead, while irregular graph traversal generates scattered accesses that degrade cache locality and memory-level parallelism. We present RVANNS, an RVV-oriented ANNS engine that jointly optimizes vector representation and graph locality. Its Mixed-Precision Multi-Layer Index (MPMI) represents each vector with a dense 8-bit affine base and sparse FP16/FP32 residuals, fusing reconstruction with distance accumulation and aligning widening with LMUL-sized register groups. ROrder co-locates likely co-visited graph nodes and sorts remapped adjacency lists, transforming scattered payload probes into denser, predominantly forward-moving address streams. Integrated into Milvus, RVANNS achieves 3.39x and 4.94x speedups over scalar execution on real 128-bit and 256-bit RVV processors, respectively. Under controlled HNSW configurations, it improves throughput by 2.27–2.76x over RVV SIMD+FP32 and by 1.18–1.59x over the corresponding AVX-512 and SVE baselines. On Cohere10M, it further delivers 1.82–2.27x higher QPS/W than the evaluated GPU baselines. Subjects: Information Retrieval (cs.IR) Cite as: arXiv:2608.09077 [cs.IR] (or arXiv:2608.09077v1 [cs.IR] for this version) https://doi.org/10.48550/arXiv.2608.09077 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[IR-4] PreGress: Ranking-Native Pre-training and Prompting for Graph Node Ranking
链接: https://arxiv.org/abs/2608.09016
作者: Lujie Ban,Jiasheng shi,Yingli Zhou,Kaiwen Xue,Daiyin Wang,Xubin Li,Shuanghua Li,Chenhao Ma
类目: Information Retrieval (cs.IR); Machine Learning (cs.LG)
备注:
Abstract:Node ranking is a fundamental problem in graph information retrieval, measuring the relative importance of nodes and supporting a wide range of applications such as influence analysis, recommendation, and graph-based retrieval augmented generation. However, exact computation of graph-based ranking measures is often computationally prohibitive at scale. Existing GNN-based ranking methods provide scalable approximations, but they are typically tailored to individual ranking criteria and require retraining for each downstream task, which limits their transferability and efficiency. Recent graph pre-training approaches aim to enable knowledge transfer across tasks, yet their learning objectives are largely misaligned with node ranking, resulting in suboptimal adaptability to ranking-oriented applications. To address these limitations, we propose PreGress, the first ranking-native pre-training and prompting framework for supporting a wide range of node ranking tasks. PreGress performs multi-task pre-training using our carefully designed objectives, including degree centrality prediction and attribute reconstruction, to jointly capture structural and attribute information. To support heterogeneous ranking criteria, we design lightweight, task-specific prompt modules that adapt a frozen ranking backbone to downstream tasks without full retraining. Experiments on six public graphs and two real-world query-to-item benchmarks—Yelp2018 and MovieLens-100K—together with a controlled five-criterion graph-access study demonstrate strong ranking quality with low task-specific state overhead.
[IR-5] Automating Freshman Course Placement and Registration: A Case Study
链接: https://arxiv.org/abs/2608.08776
作者: Bharathwaj Vijayakumar,Samyukta Alapati,Sahana Varadaraju
类目: Computers and Society (cs.CY); Databases (cs.DB); Information Retrieval (cs.IR)
备注: Published in EdgeCon Proceedings 1(1), 2025
Abstract:This implementation report explores Rowan University’s effort to automate the process of freshman course placement and registration. Historically, Freshman Instructional Guides (FIGS) at Rowan was manually executed, requiring significant time from Testing Services, University Advising, and the Registrar’s Office to evaluate placement needs and assign students to courses. Given the 57% surge in first-time degree-seeking student enrollment over a decade, the manual processes became increasingly unsustainable. In response, a cross-departmental team developed a comprehensive automated process to integrate data from Banner (Student Information System), Google Sheets maintained by Advising, and other sources. This automated process classifies students based on program groupings, determines primary and secondary course placements, checks for real-time availability and constraints in Banner, and completes course registration for freshmen in bulk. The resulting system processed over 3500 incoming students with over 350 hours in annual time savings, reduced the potential for human error, and enabled staff to shift focus from administrative work to strategic advising. This report outlines the implementation context, design architecture, technical integration, assessment methods, lessons learned, and practical implications for institutions with similar challenges.
[IR-6] BOUND: Brief-Guided Corrective Preference Distillation at Search-Control Boundaries
链接: https://arxiv.org/abs/2608.08768
作者: Qingying Niu,Ruiyang Ren,Wayne Xin Zhao,Yaliang Li
类目: Information Retrieval (cs.IR)
备注: 15 pages
Abstract:Large language model (LLM)-based deep search agents solve tasks through iterative retrieval and reasoning, but locally relevant evidence can cause persistent wrong-anchor drift, constraint drift, or local-topic drift. Existing methods supervise trajectories, outcomes, or steps, but rarely distinguish task-aligned continuations from locally plausible ones that reinforce drift. We propose BOUND, a brief-guided corrective preference distillation framework for persistent search drift. For each student-induced decision-time state, BOUND constructs a teacher-side search-state brief that preserves the original search target and key constraints while summarizing confirmed evidence, missing information, and drift status. Guided by the brief, the teacher determines whether the student’s continuation contains a correctable local search-control error likely to affect subsequent decisions. Together with the rollout outcome, this assessment determines whether to construct a corrective contrast between a student-specific correction and the original continuation, or a termination contrast between a supported answer and an unnecessary retrieval continuation. Each validated state-matched preference pair operationalizes a search-control boundary. Direct preference optimization (DPO) distills these preferences into the student, while the brief and teacher-side computation remain confined to training. We evaluate BOUND on four multi-hop QA benchmarks and three deep-search benchmarks. Across the six benchmarks for which we reran baselines, BOUND leads on five datasets and 12 of 14 metrics. Under the same search-control interface and matched settings, BOUND outperforms Trajectory SFT by 5.6 EM points on Bamboogle and 4.8 accuracy points on BrowseComp-Plus. Code is available at this https URL.
[IR-7] AnchorFold: A Focus-Then-Fold Framework via Recursive Attention Propagation for Efficient Multi-Vector Visual Document Retrieval
链接: https://arxiv.org/abs/2608.08732
作者: Haoyu Zuo,Yibo Yan,Xin Zou,Shuliang Liu,Yi Cao,Mingdong Ou,Xuming Hu
类目: Computer Vision and Pattern Recognition (cs.CV); Computation and Language (cs.CL); Information Retrieval (cs.IR)
备注: 24 pages, 7 figures
Abstract:Multi-vector vision-language retrievers enable fine-grained Visual Document Retrieval (VDR) through late interaction, but storing and scoring hundreds of visual patch embeddings per page incurs substantial overhead. Existing training-free methods rely on pruning or merging: pruning degrades sharply under aggressive compression, whereas merging does not explicitly prioritize important regions when forming representatives. We introduce AnchorFold, a training-free focus-then-fold framework for document-side index compression. AnchorFold applies Recursive Attention Propagation over visual self-attention graphs, performing multi-step propagation within each attention head and integrating scores across heads and layers. The focus stage selects the highest-centrality tokens as anchors. The fold stage assigns remaining tokens to their most similar anchors in the normalized retrieval space and summarizes each anchor-centered group through centrality-weighted aggregation. This preserves non-anchor contributions while concentrating capacity on structurally important tokens. Across ViDoRe v1/v2 and REAL-MM-RAG with three diverse retrieval backbones, AnchorFold consistently outperforms all evaluated training-free baselines at \gamma \leq 0.20 . On ViDoRe v1/v2, it retains 98.3% of full-index NDCG@5 on average at 5\times compression, achieving near-lossless compression, and 92.4% at 20\times compression.
[IR-8] Weather- and Location-Aware Agent ic Dining Recommendation: Leverag ing LLM World Knowledge for Region-Sensitive Contextual Reasoning
链接: https://arxiv.org/abs/2608.07593
作者: Kadharmoideen Fadurudeen
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
备注: 5 pages. An agentic LLM system that reasons over combined location and weather context for region-sensitive dining recommendation. Working prototype implemented and briefly deployed end-to-end
Abstract:Context-aware recommender systems have long recognized that factors such as location, time, and weather shape where and what people choose to eat. Existing weather-aware food and point-of-interest recommenders, however, typically treat weather generically – mapping conditions to preferences through hand-crafted rules or specially trained context models – and do not capture that the culturally appropriate response to weather is itself region-specific: a rainy evening calls for hot tea and fried snacks in one culinary culture and for very different comfort food in another. Encoding such weather-by-region-by-cuisine interactions as explicit rules or training data is brittle and does not scale. We present a weather- and location-aware agentic dining-recommendation system that takes a different approach: a large language model (LLM) orchestrates tools for location and weather retrieval and then reasons in natural language over the combined context, drawing on the cultural and culinary world knowledge already latent in the model to produce region-sensitive, weather-appropriate recommendations without per-region rule tables or specialized training. We describe the agent architecture, the tool-orchestration flow (Google location services and a weather service feeding an OpenAI LLM), and the reasoning mechanism, and we report on a working prototype that was implemented and briefly deployed end-to-end. We discuss design trade-offs – cost, latency, ambiguity handling, and fallbacks – and we are explicit about limitations, including the absence of a formal user study and the risk of cultural stereotyping in locality-based inference. The contribution is architectural: a simple, extensible pattern for incorporating environmental and cultural context into agentic recommendation through LLM reasoning rather than engineered rules.
人机交互
[HC-0] LITEWAY: LIghtweight HAR via Temporal Efficient highWAY ISWC2026
链接: https://arxiv.org/abs/2608.09421
作者: Dominique Nshimyimana,Vitor Fortes Rey,Mengxi Liu,Bo Zhou,Paul Lukowicz
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC)
备注: accepted at UbiComp / ISWC 2026
Abstract:Wearable human activity recognition (HAR) remains challenging due to the computational and energy constraints of deep learning models on resource-limited devices. Existing lightweight approaches often rely on recurrent architectures (e.g., GRU and LSTM), limiting parallelism and increasing inference latency. We propose LITEWAY, a modality-agnostic, fully convolutional framework for multichannel sensor time series that replaces recurrent temporal modeling with structured convolutional decomposition. LITEWAY combines lightweight convolutional blocks, strided temporal processing, and convolution-attention pooling to efficiently capture temporal dependencies while reducing computational complexity. We evaluate LITEWAY on 16 HAR datasets against TinyHAR, TinierHAR, and MLP-HAR. LITEWAY achieves competitive macro F1 while reducing model size by 4.06x-9.52x (Light) and 3.87x-9.07x (Full) compared with TinyHAR and TinierHAR. Deployment experiments further show energy reductions of 2.29x-3.14x (Light) and 1.46x-2.01x (Full) compared with TinierHAR and MLP-HAR, highlighting efficient fully convolutional temporal modeling for wearable HAR. The source code is publicly available at this https URL.
[HC-1] When Confidence Fails: Overconfidence in LLM s under Uncertainty and Missing Clinical Information
链接: https://arxiv.org/abs/2608.09080
作者: Maryam Tahermazandarani,Adnan Mahmood,Fahmida Islam,Quan Z. Sheng
类目: Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Machine Learning (cs.LG)
备注:
Abstract:Large Language Models (LLMs) have achieved strong performance in medical question answering and clinical reasoning tasks. However, their reliability under uncertainty remains poorly understood which raises critical concerns for deployment in high-stakes clinical settings. In such environments, incorrect predictions are inherently risky, but confident incorrect predictions can be particularly harmful as they may mislead clinical decision-making. In this paper, we conduct a systematic behavioral analysis of LLMs under clinical information uncertainty. We propose an evaluation framework based on the MedMCQA dataset consisting of two complementary uncertainty settings. First, we introduce linguistic uncertainty cues through prompt modifications to simulate ambiguous clinical contexts. Second, we construct an answer removal setting, wherein the correct option is deliberately excluded mandating the model to recognize insufficient information and abstain. We analyze both model accuracy and confidence behavior using multiple calibration metrics including calibration gap, Expected Calibration Error (ECE), and Unsafe Confident Error Rate (UCER) across 500 medical questions. Our results reveal a consistent failure mode, i.e., although accuracy degrades under increasing uncertainty, model confidence remains misaligned with accuracy. This leads to a substantial increase in unsafe confident errors, indicating that model confidence remains largely insensitive to clinically meaningful information loss. Furthermore, we observe significant variation across models in their ability to abstain when the correct answer is unavailable, with some models persistently producing high confidence hallucinated answers. These findings expose critical limitations in the epistemic reliability of current LLMs and highlight the need for uncertainty aware evaluation methods prior to their deployment in clinical workflows.
[HC-2] How People Evaluate AI- Expert- and Peer-Style Financial Advice
链接: https://arxiv.org/abs/2608.09019
作者: Aryan Ramchandra Kapadia,Eshwar Chandrasekharan,Koustuv Saha
类目: Human-Computer Interaction (cs.HC); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY); Social and Information Networks (cs.SI)
备注:
Abstract:As generative AI increasingly becomes a common source of daily decision-making, including financial choices, it is critical to understand how people evaluate AI-generated financial advice. We conducted a preregistered vignette experiment (N = 285) in which substantive financial content—including facts, numerical values, recommendation direction, and core reasoning—was held constant while communication style varied across AI Financial Assistant (AI), Certified Financial Planner (Expert), and Online Community Forum (OC) advice. Displayed source attribution was independently manipulated through correctly labeled, unlabeled, and mislabeled conditions, allowing us to separate attribution effects from source-specific communication cues. Expert advice was rated more favorably than AI advice on 9 of 10 outcomes (|d|=0.20–0.47), and this advantage remained visible without source labels, where Expert advice outperformed AI advice on 8 of 10 outcomes (up to d=0.60). Correct labels added limited differentiation, whereas mislabeling increased ratings of AI advice for situational fit and overall quality (d=0.42 for each) and attenuated the Expert advantage in situational fit (d=-0.36). Descriptive analyses further showed that AI advice was most responsive to displayed attribution and, conversely, that advice-style differences were most visible under an AI label. These findings show that financial-advice evaluations are shaped jointly by displayed attribution and message-level communication cues. We position disclosure not as a neutral transparency mechanism, but as an interpretive frame whose accuracy and interaction with message cues can shape trust and reliance.
[HC-3] From Scaffolding to Internalization: Enhancing CPR Training with In-Situ Visualization and Kinesthetic Feedback
链接: https://arxiv.org/abs/2608.08729
作者: Jiahe Dong,Shuhao Zhang,Yutao Ming,Jinkai Zhang,Yurui Zhang,Quan Li
类目: Human-Computer Interaction (cs.HC)
备注: In The 39th Annual ACM Symposium on User Interface Software and Technology (UIST '26), November 02-05, 2026, Detroit, MI, USA. ACM, New York, NY, USA, 15 pages. this https URL
Abstract:CPR training requires learners to not only understand explicit procedural targets, such as compression depth and rate, but also to internalize these targets as stable psychomotor skills. However, existing CPR training systems often rely on feedback presented outside the action space, which divides learners’ attention between performing compressions and monitoring external guidance. This separation weakens the coupling between action and bodily sensation and may lead to an over-reliance on external feedback, compromising skill retention once support is removed. To address this challenge, we conducted a formative study with novice trainees and certified BLS instructors, from which we derived three design goals: embedding feedback within the task space, providing active kinesthetic guidance, and gradually fading assistance based on learning phases. Informed by these insights, we designed Kinesthetic-CPR, a stage-adaptive multimodal mixed reality CPR training system, and evaluated it in a controlled user study across two sub-studies (N = 60). This work offers design implications for CPR training systems that aim to better support skill retention.
[HC-4] Enhanced Real-Time 6-DOF Extended Reality Catheter Tracking for Evaluating Potential Improvement in Efficiency Precision and Depth Perception for Cardiac Interventions
链接: https://arxiv.org/abs/2608.07606
作者: Mohsen Annabestani,Sandhya Sriram,Andrew Kuzemczak,S. Chiu Wong,Alexandros Sigaras,Bobak Mosadegh
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Human-Computer Interaction (cs.HC); Image and Video Processing (eess.IV)
备注:
Abstract:Despite advances in 3D ultrasound, most percutaneous cardiac interventions still rely on 2D visualization, limiting depth perception and spatial understanding. To address this challenge, we developed an Extended Reality (XR)-based platform that enables real-time six-degree-of-freedom (6-DOF) catheter tracking and visualization within a patient-specific 3D heart model. The system combines a custom machine-vision algorithm for 5-DOF catheter tracking with a 3D-printed electromechanical encoder that measures catheter roll, providing complete 6-DOF motion reconstruction. In a proof-of-concept study, 20 novice medical students navigated an intracardiac echocardiography (ICE) catheter to six anatomical targets using either immersive 3D visualization or a conventional 2D cathlab-style view. Participants in the 3D condition completed the task in 54.6 seconds and traveled 1,939 mm on average, compared with 267.5 seconds and 7,854 mm in the 2D condition. Therefore, the XR-based 3D system was more than 5x faster and required ~5x less catheter travel. The 3D mode also improved targeting precision and reduced performance variability. Participants consistently rated immersive visualization higher for accuracy, speed, usability, and clinical value. Kinematic analysis showed smoother depth-axis navigation in 3D, whereas 2D users relied on repeated corrective movements. These findings demonstrate that XR-based visualization can substantially improve procedural training efficiency, precision, and motor control.
[HC-5] “Always Want to Use it for Everything”: Understanding Young Adults Perceptions of AI Dependence AAAI
链接: https://arxiv.org/abs/2608.07592
作者: Ashlee Milton,Leah Ajmani,Amy Heger,Forough Poursabzi-Sangdeh,Mihaela Vorvoreanu,Jina Suh
类目: Human-Computer Interaction (cs.HC); Computers and Society (cs.CY)
备注: To appear in proceedings of the Ninth AAAI/ACM Conference on AI, Ethics, and Society (AIES '26)
Abstract:The growing integration of general-purpose AI chatbots into people’s daily lives has raised concerns about the potential for unhealthy dependence, particularly among young adults. As a first step toward understanding and characterizing AI chatbot dependence from the perspective of young adults, we collected testimonials from AI chatbot users aged 18 to 25 through an online questionnaire to capture their thoughts and experiences with this phenomenon. From participant responses, we identified three contributing factors of AI dependence: chronic use, efficiency, and delegation. The combination of these in a person’s interaction behavior was considered to indicate AI dependence. Participants also observed feelings of atrophy in abilities from AI dependence, leading to psychological impacts such as feelings of inadequacy. Interpreting these findings through the lens of self-determination theory reveals how AI chatbot dependence can impact young adults’ personal and social development. We argue that preventing lasting harm to young adults’ development is paramount, and provide implications for rethinking AI chatbot dependence grounded in this understanding.
计算机视觉
[CV-0] Renormalising Generative Models for Active Inference: Foundations Derivations and Verification
链接: https://arxiv.org/abs/2608.09512
作者: Karim Zaghw,Andrew Pashea,Marc Pritsch,Wouter Nuijten,Karl Friston,Lancelot Da Costa
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 25 pages, 1 figure. Accepted as a full paper at the 7th International Workshop on Active Inference (IWAI 2026). Supplementary material: this https URL . Code: this https URL
Abstract:Active inference offers a unified framework for perception, learning, and action, but scaling discrete active-inference models to rich spatial and temporal domains remains difficult. Renormalising generative models (RGMs) address this challenge by composing discrete generative models across spatial and temporal scales, coarse-graining lower-level states and paths into higher-level causes for objects, events, and action. However, fully reproducing and adapting the framework remains difficult: the mathematical exposition is compact, and the reference implementations are deeply integrated within specialized software environments, leaving many algorithmic details implicit. This paper addresses these challenges by providing a self-contained, derivation-oriented account of RGMs together with an open, verified implementation. We explain how the hierarchy is built, how beliefs and actions are updated within it, and how information is passed between levels. Where the published equations and implementation differ in emphasis, we make those choices explicit and explain their modelling consequences. By clarifying the theory and separating it from its original implementation context, this work lowers practical barriers to entry and makes RGMs more transparent, auditable, and reproducible, providing a foundation for future quantitative evaluation and development on machine-learning benchmarks.
[CV-1] SwissCrop25: A National Multi-Year Benchmark for Operational Crop Mapping ECCV2026
链接: https://arxiv.org/abs/2608.09497
作者: Thomas Lauber,Mehmet Ozgur Turkoglu,Sélène Ledain,Helge Aasen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted at the ECCV 2026 Workshop TerraBytes II. To appear in the workshop proceedings
Abstract:Operational crop mapping requires models that generalise across years, resolve fine-grained crop taxonomies, and distinguish cropland from surrounding landscapes. However, existing crop mapping datasets enable evaluation of these requirements only in isolation. We therefore introduce SwissCrop25, a national-scale crop mapping benchmark dataset spanning seven growing seasons (2019-2025). SwissCrop25 combines Sentinel-2 time series, daily temperature observations, a fine-grained 73 crop taxonomy including grassland management types, and 5 explicit non-crop land cover classes. To evaluate realistic deployment conditions, we define a leave-one-year-out protocol with joint cropland delineation and crop classification for benchmarking representative crop mapping architectures. Evaluating U-TAE (convolutional temporal-attention model), TSViT (transformer-based spatio-temporal model), and Galileo (EO foundation model) reveals differences between architectures hidden by conventional benchmarks. In this setting, domain-specific models outperform Galileo, with TSViT achieving the best overall performance and a 12 pp macro-mIoU advantage over U-TAE. SwissCrop25 also exposes substantial interannual distribution shifts and shows that incorporating temperature-derived phenological information improves robustness. Finally, in-season evaluation reveals a trade-off between models, with U-TAE performing better early in the season and TSViT gaining an advantage later through improved rare-class discrimination. SwissCrop25 provides a challenging testbed for evaluating crop mapping systems under realistic operational conditions and is publicly released at this https URL .
[CV-2] GeoRoute: Geometry-Aware Hybrid Inference for Traffic Future-Frame Prediction ECCV2026
链接: https://arxiv.org/abs/2608.09493
作者: Khang Minh Le,Hieu Dinh Trung Pham,Luu Thanh Danh,Nam-Tien Le,Hieu Anh Ngo,Phuong Huu Vu Tran,Son Nguyen Minh Le,Nguyen Trong Nghia,Tu Tran Thi Cam,Huy Minh Nhat Nguyen,Cuong Tuan Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: accepted to the ECCV 2026 AI City Challenge Workshop
Abstract:Long-horizon future-frame prediction is important for autonomous driving, traffic surveillance, and intelligent transportation systems, yet remains challenging due to temporal ghosting, geometry drift, and inconsistent object motion. Recent latent video diffusion models have achieved impressive visual quality, but directly applying them to structured traffic scenes often leads to unstable geometry and degraded temporal coherence over extended horizons. We present a training-free inference framework that stabilizes reliable static structure in pretrained video predictions through multi-frame temporal context and view-conditioned routing. For front-camera videos, our method refines generated futures with a multi-frame depth-layered renderer that projects static geometry from observed history frames while preserving dynamic regions from the generative base model. For heterogeneous traffic views, a frozen vision-language model infers a coarse camera group from the observed clip and selects a specialized motion-based predictor. The framework requires neither retraining nor fine-tuning of the underlying video model and can be applied directly to pretrained generators. We validate the proposed framework on the AI City Challenge Track 5 benchmark, where our final system achieves competitive performance among the top-ranked teams. These results demonstrate that geometry-aware inference-time refinement and view-conditioned hybrid inference can improve static-geometry stability and low-level structural fidelity without changing the original model architecture.
[CV-3] Beyond Uniform Restoration: Empowering All-in-One Restoration with Pixel-Level Multimodal Guidance
链接: https://arxiv.org/abs/2608.09482
作者: Chunxiao Liu,Wei Liu,Anbin Xiong,Erli Meng
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted by ACMMM2026 as Oral Paper
Abstract:All-in-one image restoration is a unified low-level vision task that aims to effectively recover high-quality images from inputs degraded by various types and levels of corruption using a single model. Recent works have achieved remarkable progress by learning degradation-adaptive prompts or network architectures. However, these methods typically apply a uniform restoration strategy across the entire image, neglecting the fact that different regions may suffer from distinct degradation types and varying degrees of severity. In contrast, we propose to perform restoration at the pixel level, thereby enabling more fine-grained and precise control over the restoration process. Specifically, we present MGN-AIR, a novel pixel-level restoration framework for all-in-one image restoration. Our approach first learns to estimate a pixel-level visual prompt. Then, it leverages both textual and visual prompts to provide global and local degradation cues, guiding the model on where to look and how to restore at each pixel. We conduct extensive experiments on multiple all-in-one image restoration benchmarks, covering a wide range of tasks including denoising, deraining, deblurring, dehazing, desnowing, and low-light enhancement. Experimental results demonstrate that our proposed method consistently and significantly outperforms existing approaches.
[CV-4] Agreement-Based Audio-Visual Segmentation:Champion Report for the MeViS-Audio Track in the 8th LSVOS Challenge
链接: https://arxiv.org/abs/2608.09475
作者: Yiwen Ren,Jianing Liu,Yingxin Wang,Kexin Zhang,Licheng Jiao,Lingling Li,Xu Liu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:The MeViS-Audio track asks a system to segment the objects described by a spoken motion expression throughout a video and to return empty masks when the described target is absent. We present a simple staged solution. Qwen3-ASR first converts speech into text. Several video mask tracks are then produced with complementary grounding and segmentation models. Instead of trusting a single prediction, we select the track that has the highest average mask agreement with the other candidates. A small set of explicit direction, count, and plural rules corrects queries that require more than ordinary single-object tracking. Finally, a video-level classifier combines visual, audio-visual, and within-video query scores to decide whether any target is present. The submitted system obtains 0.5952 J F, 0.7931 no-target accuracy, 0.9205 target accuracy, and a final score of 0.769589. The challenge organizers notified our team that this result ranked first in the track.
[CV-5] FaLCon: Facet-Anchored Retrieval with Late Consensus for Sim2Real Text-Based Person Anomaly Search ECCV2026
链接: https://arxiv.org/abs/2608.09474
作者: Hieu Dinh Trung Pham,Phuong Huu Vu Tran,Thuan Duc Mai,Son Nguyen Minh Le,Khang Le Minh,Hoang Vo,Minh-Chi Phung,Huy Minh Nhat Nguyen,Cuong Tuan Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: accepted to the ECCV 2026 AI City Challenge Workshop
Abstract:Text-based person anomaly search requires retrieving real-world pedestrian images from detailed natural-language descriptions using models trained primarily on synthetic data. This Sim2Real setting is particularly challenging because visually similar candidates may differ only in subtle actions, object interactions, or appearance attributes, while applying multimodal large language models to the entire gallery is computationally expensive. We propose an anchor-constrained coarse-to-fine retrieval framework that combines global semantic matching with fine-grained verification. First, each query is represented by its original caption, a structured concatenation, and several semantic facets. Heterogeneous vision-language retrievers are then integrated through robust per-query score calibration and soft claim-aware fusion. Full and concatenated captions serve as anchors to preserve candidate recall, whereas appearance, action, and object facets provide bounded corrective evidence. The resulting candidate pool is further refined by a discriminative Qwen3 reranker and two complementary semantic verification modules based on anomaly-aware cloze completion and multi-agent evidence reasoning. Finally, an uncertainty-gated consensus module adaptively reweights the three experts on ambiguous queries. Experiments on the PAB benchmark show that the proposed soft claim-aware retrieval achieves 86.44% mAP@10, substantially outperforming individual retrieval backbones. The complete framework further improves performance to 95.41% mAP@10, 94.44% R@1, and 99.09% R@5. These results demonstrate that preserving strong global retrieval while restricting expensive semantic reasoning to a small candidate pool is effective for fine-grained Sim2Real person anomaly search. Our code will be available on Github.
[CV-6] RecoverFly: A Failure-Aware Reinforcement Learning Post-Training Framework for Aerial Vision-Language Navigation
链接: https://arxiv.org/abs/2608.09467
作者: Boxiong Wang,Hui Kang,Geng Sun,Jiahui Li,Chao Yu,Daxin Tian
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Unmanned aerial vehicle vision-language navigation (UAV-VLN) requires agents to translate visual observations and language instructions into reliable flight actions in complex environments. Although recent end-to-end UAV vision-language-action (UAV-VLA) policies reduce reliance on separately designed perception, planning, and control modules, their behavior-cloning objectives provide limited corrective supervision for interactive closed-loop execution. Reinforcement learning (RL) offers a promising solution, while its effectiveness is constrained by inefficient use of samples, long-tailed scene distributions, and policy distribution shift during optimization. To this end, we propose RecoverFly, a failure-aware RL post-training framework for end-to-end UAV-VLA policies. Specifically, RecoverFly adapts token-level RL for stable optimization of grammar-constrained autoregressive UAV actions, revisits unresolved failure cases to strengthen corrective learning and sample utilization, and combines a two-stage long-tail scene curriculum with reference-policy regularization to improve scene adaptation while preserving acquired capabilities. Experiments on the TravelUAV benchmark demonstrate that RecoverFly achieves the best performance on the seen, unseen-map, and unseen-object splits. Moreover, compared to the AerialVLA initialization, RecoverFly improves success rate by 3.12 to 8.37 percentage points under a total rollout budget of about 30% of the training-set size, validating its effectiveness, robustness, and generalization capabilities.
[CV-7] Flow-based conditional cardiac anatomy generation for virtual cohorts
链接: https://arxiv.org/abs/2608.09460
作者: Konstantinos Kevopoulos,Beatrice Moscoloni,Benjamin Alheit,Cameron Beeche,Julio A. Chirinos,Alexander Heinlein,Mathias Peirlinck
类目: Machine Learning (cs.LG); Computer Vision and Pattern Recognition (cs.CV); Quantitative Methods (q-bio.QM); Tissues and Organs (q-bio.TO)
备注:
Abstract:Cardiac digital twin research is moving from subject-specific anatomical replicas toward virtual cohorts that represent clinically relevant population subgroups. Yet access to representative imaging-derived anatomy datasets remains limited by cohort size, subgroup sparsity, and data-sharing constraints. Conditional generative models could help address this gap, but virtual cohorts are useful only if they preserve realistic, metadata-dependent anatomical variability. Existing cardiac anatomy generators largely rely on conditional variational autoencoders (cVAEs), which couple representation learning and metadata conditioning through a shared regularized latent prior. We introduce CAN-FLOW, a two-step Conditional ANatomy generation framework based on normalizing FLOWs that first learns geometry-only latent representations of diffeomorphic cardiac shape momenta and then models their sex-, age-, and body-mass-index-dependent distribution with a conditional normalizing flow. We trained CAN-FLOW on 2,208 healthy UK Biobank subjects and compared it with cVAEs across regularization strengths. CAN-FLOW generated plausible stochastic biventricular anatomies that better reproduced clinical phenotype distributions, metadata-dependent trends, subgroup variability, point-cloud coverage, and high-dimensional shape variability. Together, these results establish CAN-FLOW as a shareable framework for generating realistic, stochastically varying, metadata-conditioned biventricular anatomies for virtual cohort construction and in silico clinical trial workflows.
[CV-8] A Content-Aware Pure Permutation with Intrinsic Avalanche Effect: Breaking the Diffusion-Permutation Dichotomy
链接: https://arxiv.org/abs/2608.09452
作者: Zahra Ghoraeian,Mohammad-Reza Sadeghi,Samaneh Mashhadi
类目: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR)
备注: 12 pages, 5 figures
Abstract:Pixel permutation is a fundamental tool in image processing, image encryption, and data hiding (including watermarking and steganography) that rearranges pixels without changing their values. A common assumption in the literature is that permutation alone cannot create differential sensitivity; changing one pixel merely relocates that pixel in the output, producing no avalanche effect. This paper challenges this by introducing the Triangular Content-Aware Permutation (TCA) algorithm. The method extracts edge points using Canny and applies Delaunay Triangulation to edges and corners, creating a unique partition. Since triangulation is highly sensitive to image geometry, changing a single pixel alters the edge map, resulting in a completely different triangulation and global permutation pattern. Unlike classical dimension-based permutations and advanced content-aware methods (2025-2026), which lack differential sensitivity, TCA increases NPCR from near-zero to 97.10% solely through pixel relocation. Experiments on 50 images show that TCA, with an average of 14.81 iterations, achieves NPCR = 97.10% and UACI = 20.06%, proving pure permutation can create significant differential sensitivity. Conventional methods maintain near-zero NPCR. The iteration threshold varies from 6.4 to 30.7 based on content complexity. Low PSNR (11.93 dB) and near-zero correlation (~10^-3) confirm superior statistical performance. Although slower than classical methods due to triangulation, this is a deliberate trade-off for stronger security. Given the non-analytic, content-dependent nature of the pattern, TCA is ideal for reference-based encryption, fragile watermarking, and non-blind steganography. Comments: 12 pages, 5 figures Subjects: Computer Vision and Pattern Recognition (cs.CV); Cryptography and Security (cs.CR) MSC classes: 68U10, 94A60 ACMclasses: I.4.9; E.3; K.6.5 Cite as: arXiv:2608.09452 [cs.CV] (or arXiv:2608.09452v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.09452 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[CV-9] Sekai2: From World Exploration to Interactive World Modeling
链接: https://arxiv.org/abs/2608.09449
作者: Kang He,Wenshuo Peng,Zihui Gao,Jiaming Tan,Kaipeng Zhang,Yongtao Ge
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Sekai2 dataset technical report. Developed at Alaya Lab
Abstract:Video world models must capture how scenes evolve over time and across viewpoints. Training them for long-horizon generation and camera control therefore benefits from long videos paired with camera trajectories and temporally grounded semantics. Existing corpora rarely offer the three together: large-scale web video provides broad visual diversity but no trajectories or time-aligned text, while pose-annotated datasets are typically short-range or reconstruction-oriented. We introduce Sekai2, a multi-source real-world video dataset that carries the world-exploration footage of Sekai toward interactive world modeling. The release contains 128,892 clips totaling 2,826 hours from 10,428 source videos across 113 countries or regions, and is deliberately weighted toward sustained observation: under a common 120-second decomposition, 43,594 segments reach the full two minutes and account for 51.4% of all footage. Every clip includes a released camera trajectory and hierarchical annotations disentangling subject motion, environment dynamics, static scene content, and camera behavior, resulting in 649,597 temporally grounded segments. Crucially, we further introduce 982 panoramic sequences captured along non-linear trajectories with loops and revisits. These revisits provide repeated observations of the same locations across time and viewpoints, offering essential supervision for learning persistent scene representations, long-term spatial memory, and geometrically consistent world models. Corpus-scale analyses demonstrate complete pose-and-caption coverage, broad geographic and semantic diversity, varied camera trajectories, and highly non-redundant temporal descriptions. Together, these properties make Sekai2 a scalable resource for long-horizon video generation, camera-controllable synthesis, and interactive world-model pre-training.
[CV-10] VANE: Reliable Test-Time Training for Vision-Language-Action Models via Future Visual Representation Prediction
链接: https://arxiv.org/abs/2608.09448
作者: Hongjin Ji,Guoyang Xia,Luoyang Sun,Fangxiang Feng,Lei Ren
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Test-time training (TTT) offers a lightweight way to adapt vision–language–action (VLA) policies from unlabeled deployment streams, but it remains difficult to use reliably in closed-loop manipulation. A shared adaptation space can mix incompatible task corrections, while an online update can alter subsequent actions before its consequences are known. We introduce a reliable TTT framework for VLA policies (VANE). VANE conditions prompt adaptation on the current vision–language context and learns from the future visual consequences of executed actions. Candidate updates are isolated from the live policy, evaluated on subsequent observations, and committed only when supported by future evidence, making adaptation selective and reversible. On SimplerEnv WidowX, VANE improves average success by 3.2 percentage points over the corresponding TTT baseline. Results on Google Robot further show that deployment-time gains remain task- and embodiment-dependent. Together, these results demonstrate a constrained, evidence-based approach to adapting VLA policies during interaction.
[CV-11] DiffSafeMerge: Mitigating Backdoor Inheritance in Diffusion Model Merging
链接: https://arxiv.org/abs/2608.09445
作者: Jiayang Zhang,Ji Guo,Jiachen Li,Wenshu Fan,Wenbo Jiang
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Unconditional diffusion checkpoint merging assumes benign sources, yet a compromised public checkpoint can transfer a dormant backdoor while clean generation appears normal. Mitigation is difficult without knowing the compromised source, trigger, or target, and broad sanitization may degrade image quality. We introduce DiffSafeMerge (DSM), which uses a small unlabeled clean set and fixed, attack-agnostic stress probes to score source blocks, shrink suspicious contributions toward a trusted reference, and select attenuation under a clean denoising-loss budget. We evaluate four attacks, two datasets, and 21 target conditions. Intended merging already has zero worst-target ASR in 10 of 14 source cases; DSM preserves these outcomes and records no target match in the remaining four over three seeds, including three with baseline ASR of 48–100%. Among methods with zero worst-target ASR on both datasets, DSM obtains the lowest case-averaged FID in the matched seed-0 comparison.
[CV-12] Unveiling the Secret of AdaLN-Zero in Diffusion Transformer
链接: https://arxiv.org/abs/2608.09438
作者: Jie Zhu,Mingyu Ding,Boqiang Duan,Leye Wang,Jingdong Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accept by IEEE TPAMI 2026, camera-ready version
Abstract:Diffusion transformer (DiT), a rapidly emerging architecture for image generation, has gained much attention. However, despite ongoing efforts to improve its performance, the understanding of DiT remains superficial. In this work, we delve into and investigate a critical conditioning mechanism within DiT, adaLN-Zero, which achieves superior performance compared to adaLN. Our work studies three potential elements driving this performance, including an SE-like structure, zero-initialization, and a “gradual” update order, among which zero-initialization is proved to be the most influential. Building on this understanding, we propose an analysis-guided initialization strategy, termed adaLN-Gaussian, which serves both as an empirical validation of our analysis and as a practical initialization method that consistently improves optimization efficiency. On the other hand, inspired by the SE-like structure, we introduce an improved conditioning mechanism called SE-adaLN-Zero. Extensive experiments following DiT on four datasets, especially on ImageNet1K demonstrate the effectiveness and generalization of adaLN-Gaussian and SE-adaLN-Zero. Beyond class-to-image generation, we also evaluate the generalization of the two improved methods on text-to-image generation.
[CV-13] Foundation Models are Implicit Deepfake Detectors
链接: https://arxiv.org/abs/2608.09427
作者: Stefan Smeu,Dragos-Alexandru Boldisor,Elisabeta Oneata,Dan Oneata
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Pretrained self-supervised representations have emerged as a core component of current deepfake detection methods, yet it remains unclear which of their properties make real and fake media distinguishable. In this work, we uncover a surprisingly consistent phenomenon: across multiple pretrained models, datasets, and both image and video domains, fake samples systematically produce lower-magnitude representations than their real counterparts. Motivated by this finding, we formulate deepfake detection as an anomaly detection problem and show that simple statistics of feature magnitude achieve competitive performance with far more sophisticated deepfake detection methods. We further investigate the origin of this effect and demonstrate that reduced feature magnitude is primarily associated with semantic shifts introduced by fake content, while low-level generative fingerprints play a comparatively smaller role. Finally, we show that this discriminative signal strengthens as the size of the underlying foundation model grows, suggesting that advances in representation learning naturally translate into stronger zero-shot deepfake detectors.
[CV-14] MeanSR: Restoration Trajectory Learning for One-Step Perceptual Super-Resolution AAAI2027
链接: https://arxiv.org/abs/2608.09405
作者: Axi Niu,Jiawei Kou,Kang Zhang,Qingsen Yan,Jinqiu Sun,Yanning Zhang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 7 pages, 8 figures. Submitted to AAAI 2027
Abstract:Diffusion-based super-resolution (SR) achieves strong perceptual quality but requires costly iterative denoising. Existing one-step distillation methods reduce inference time but depend on expensive pretrained teachers, whereas CTMSR avoids distillation through PF-ODE consistency training yet does not explicitly model the restoration dynamics from low-resolution (LR) inputs to high-resolution (HR) images. We propose MeanSR, a one-step perceptual SR method that learns an LR-conditioned average velocity field to directly capture the finite-time transition from degraded or noisy inputs to plausible HR outputs. We further reformulate distribution trajectory matching for average-velocity generation and introduce a Stage-Aware Temporal Sampling strategy to improve trajectory learning. Experiments on synthetic and real-world benchmarks show that MeanSR outperforms CTMSR on CLIPIQA, MUSIQ, and MANIQA while substantially reducing FLOPs and inference latency. MeanSR also reconstructs sharper structures and more realistic textures with fewer perceptual artifacts.
[CV-15] One Model to Magnify Them All: Efficient Scale-Invariant Histopathology via Conditional Normalization and Continuous Magnification Training
链接: https://arxiv.org/abs/2608.09403
作者: Agnieszka Florkowska,Henning Müller,Marek Wodzinski
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Whole slide images (WSIs) in digital histopathology are acquired at discrete magnification levels encoding complementary diagnostic information from global tissue architecture to fine-grained cellular morphology. Yet, deep learning models remain sensitive to scale variation. Existing magnification-invariant methods rely on multi-scale architectures at predefined discrete resolutions, while in clinical deployment the acquisition magnification varies continuously, rarely aligns with a model’s fixed training resolution, and intermediate scales are common, so robust coverage otherwise demands a costly ensemble of magnification-specific models. We propose Conditional Layer Normalization (CLN), a lightweight mechanism that generates affine normalization parameters from input pixel size via a small MLP, integrated into standard CNN architectures for both WSI classification and segmentation. Trained on patches sampled continuously across a range of pixel sizes, the model decouples inference from scanner-dependent magnification and generalizes to arbitrary, previously unseen scales at test time. On the PANDA prostate cancer dataset, our approach on average matches or exceeds independently trained single-magnification models and ranks among the top three performers at every evaluated magnification, including those unseen during training. This collapses a five-model ensemble into a single network and reduces training, and inference cost roughly 4-5 times, while leaving the multiply-accumulate count unchanged. The code is available at: this https URL.
[CV-16] Sign Language Recognition Using Original and Synthetic Depth Image Based Point Cloud Data Models
链接: https://arxiv.org/abs/2608.09400
作者: Rustem Ozakar,Eyup Gedikli
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Research regarding the sign language recognition mostly relies on RGB images, whileas sign language datasets that provide depth images are limited. Point clouds obtained from depth images can be used for sign language recognition with neural networks like PointNet. In recent years, various neural networks are used for generating realistic depth images from monocular RGB images. In this work, synthetic depth images were created from RGB images using Depth Anything V2 network. For this purpose, three sign language datasets (Real-time ASL Fingerspelling, KArSL, AUTSL) which contain both RGB and depth images were used. Classification accuracies of the point cloud data created from both original and synthetic depth images using various PointNet architectures were measured for sign language recognition. From the original and synthetic point clouds, frame based, Point Gesture Map and Long Short Term Memory data models were used for classification and their performances were compared. In the results, both original and synthetic based data achieved acceptable performance in most models. In general, original depth based point cloud models performed better than synthetic ones, however in some models synthetic depth based models performed better than the originals.
[CV-17] CableDex: Cable Length Estimation on Industrial Reels Using a Handheld Device
链接: https://arxiv.org/abs/2608.09392
作者: Francisco Guillén,Ricardo Almeida,Bruno Silva,João C. Neves
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:CableDex is a computer vision system that addresses the time-consuming and inaccurate manual measurement of cable length on industrial reels from a single photograph captured with a mobile phone. The system combines camera calibration, instance segmentation, pose estimation, and volumetric calculation to estimate the cable length across five different reel types and various cable sizes. This system is based on an instance segmentation model trained on 1,000 manually annotated images, achieving 99.5% mAP50 with an inference time of 5.66 ms per image. Evaluated on 75 reels across five reel types, the system achieves a MAPE of 4.90%, within the 10% error tolerance commonly accepted in industrial cable-reel measurement. The demonstration presents the end-to-end pipeline, from reel label scanning and image capture to segmentation and length estimation, through the mobile application.
[CV-18] CoInS-Net: A Continuous Position-Aware Network for Joint Medical Image Interpolation and Segmentation
链接: https://arxiv.org/abs/2608.09391
作者: Yujia Sun,Ningfeng Que,Peiting Shi,Rongrong Fu,Yingying Yang,Xinhang Li,Yin Dai
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Accurate medical image interpolation and anatomical structure segmentation are fundamental for computer-aided diagnosis and treatment planning. Anisotropic medical volumes with sparse through-plane sampling often suffer from structural discontinuity and boundary blur, hindering reliable clinical image analysis. Most existing methods implement interpolation and segmentation independently, which introduces redundant computation and fails to fully exploit complementary cross-slice structural information between sequential slices. To address these issues, we propose a continuous position-aware interaction network, termed CoInS-Net, for joint frame interpolation and lesion segmentation. Unlike conventional cascaded interpolation-then-segmentation paradigms, the framework enables bidirectional interaction under a shared Swin encoder with continuous spatial coordinate queries. A spatially continuous position interpolation module generates target-position features at every scale from the relative coordinate and physical spacing, and a prototype-based task mutual interaction module lets the segmentation and interpolation branches exchange global structure through a small set of shared prototypes rather than dense feature mixing. A multi-scale task-cooperative decoder further separates each scale into shared and task-specific components, so the two tasks reinforce common anatomy while preserving their distinct requirements down to the boundary level, without extra annotations. Experiments on four public medical imaging datasets with diverse modalities and anatomical regions demonstrate that the proposed method outperforms conventional single-task schemes. The joint optimization framework effectively realizes mutual promotion between interpolation and segmentation tasks, providing a reliable and universal technical scheme for intelligent clinical medical image analysis.
[CV-19] Efficient Human-Contact Representation for Human-Scene Interaction ECCV2026
链接: https://arxiv.org/abs/2608.09388
作者: Nghia Vu,Tuong Do,Binh X. Nguyen,Erman Tjiputra,Anh Nguyen
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted in ECCV 2026 Workshops
Abstract:Human-scene interaction is an active research topic with several industrial applications in virtual reality, gaming, robotics, and surveillance. Despite significant progress in network architectures to improve the results or optimize models’ parameters for fast inference speed, the efficient representation of contact between humans and their environments remains an open challenge. In this paper, we propose a new efficient human-contact representation for human-scene interaction. Our primary contribution is the introduction of sparse contact masks that strategically select essential contact information, significantly reducing redundant data in high-dimensional inputs. Leveraging this efficient contact representation, we propose a suite of sparse operators to replace traditional dense operators within deep network layers for faster computation. Our approach not only enhances computational speed but also filters out non-essential contact data, thereby improving the precision of human-scene interaction models. To validate the effectiveness of our method, we conduct intensive experiments across three public benchmark datasets, focusing on two critical tasks for human-scene interaction: contact prediction and scene synthesis. The experimental results show that our approach outperforms state-of-the-art models in reconstruction accuracy and achieves a computation speed-up of at least 12 times over recent baselines.
[CV-20] Imaginative Generative AI: Crossing the Entropy Wall into Worlds Beyond Imitation
链接: https://arxiv.org/abs/2608.09385
作者: Hossein Goli,Farzan Farnia,Amin Gohari
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Generative AI models are primarily designed to imitate the data distribution, an objective that neither corrects diversity lost by a learned generator nor defines how generation should extend beyond the diversity of the data itself. We introduce Imaginative Generative AI (IGA), a framework that makes diversity part of the target-distribution design problem: among distributions close to a reference, IGA selects one whose spectral diversity reaches a prescribed level. Diversity is measured by the von Neumann entropy of the generated distribution’s kernel covariance operator in a fixed representation space, providing a reference-free representation-guided measure of how broadly probability mass occupies embedding directions. The spectral entropy of the population data distribution defines an Entropy Wall. Below the wall, IGA performs diversity repair, recovering variation that a learned generator has lost while remaining within the diversity level of the data. Beyond the wall, the data distribution itself becomes infeasible, and IGA deliberately departs from it to produce distributions with greater representation-relative spectral diversity, an operational notion of imaginative generation. These regimes form a single regularization path from imitation to imagination and define an i.i.d. target distribution at each prescribed diversity level. We develop the theory of this entropy-constrained projection and show that, under a KL anchor to a pretrained generator, the optimum satisfies a self-consistent exponential-tilt relation. This characterization leads to IGA Guidance, a retraining-free inference-time method for score-based and diffusion models, including DDPM and DDIM samplers. Experiments on synthetic and vision benchmarks demonstrate diversity repair below the Entropy Wall and controlled spectral extrapolation beyond it.
[CV-21] CircuitReason -1k: Benchmarking Long-Horizon Visual-to-Symbolic Reasoning inElectrical Circuits
链接: https://arxiv.org/abs/2608.09374
作者: Xinqi Yang,Kang An,Tengyue Wang,Zhongyu Yang,Chenxu Du,Yuanchi Zhu,Hebao Zhu,Ziliang Wang,Faqiang Qian,Yunli Yang,Qibing Ren
类目: Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Electrical circuit analysis requires more than recognizing components in an image. A solver must ground symbols and labels, recover latent topology, select a physical model, formulate coupled equations, propagate intermediate quantities, and preserve units, signs, directions, and phase conventions. We introduce \benchmark, a benchmark of 1,000 authentic textbook problems for evaluating this complete long-horizon visual-to-symbolic reasoning process. Each problem pairs one or more circuit diagrams with a self-contained question, a typed or semantically specified answer, and a reference worked solution. An evidence-first construction pipeline aligns questions, figures, and solutions, while a reasoning-oriented taxonomy organizes problems by circuit type and dependency depth. Evaluation combines conservative typed scoring with identity-blinded multi-model semantic consensus, retaining every problem in the denominator. Across three commercial chatbot systems and six open-source multimodal large language models, the highest-scoring system reaches 84.8% accuracy. However, performance consistently deteriorates on long-horizon problems, and qualitative analysis exposes persistent failures in topology-to-target binding, physical conventions, and late-stage output propagation. \benchmark provides a focused testbed for measuring whether multimodal models can transform technical visual evidence into sustained, physically valid symbolic reasoning. Code are available at GitHub - CircuitReason/CircuitReason1K.
[CV-22] Preserve More Details: Mitigating Content Drift in Real-World Image Super-Resolution ACM-MM2026
链接: https://arxiv.org/abs/2608.09373
作者: Chunxiao Liu,Wei Liu,Anbin Xiong,Erli Meng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted to ACM MM 2026. This is the author’s accepted version. The definitive version is published in the Proceedings of ACM MM 2026
Abstract:Real-world image super-resolution (Real-ISR) aims to reconstruct high-quality (HQ) images from low-quality (LQ) inputs subject to diverse real-world degradations. Recent advances have leveraged the LQ inputs and natural image priors learned by Stable Diffusion models to achieve impressive results. However, existing methods often overlook insufficient clarity of LQ inputs inevitably induce content drift in the generated HQ images. This manifests primarily as visual detail degradation and textual semantic shift, severely compromising both fidelity and perceptual quality. To address this challenge, we propose FSP-Diff, a novel one-step diffusion model featuring a dual-pathway architecture. This architecture comprises a Detail-Conditioned Pathway for injecting structured details to recover fine structures, and a Detail-Modulated Semantic Pathway that refines semantic guidance using structured details to mitigate semantic deviations. Extensive experiments on standard Real-ISR benchmarks demonstrate that FSP-Diff surpasses existing one-step diffusion methods in both quantitative and qualitative metrics.
[CV-23] FeedbackTrack: Visual-Cortex-Inspired Cross-Frame Feedback for Transformer Tracking
链接: https://arxiv.org/abs/2608.09369
作者: Yueyang Cang,Xiaoteng Zhang,Zhiyuan Ning,Yuchen He,Li Shi
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Visual object tracking requires effective temporal integration, yet most Transformer trackers still rely on predominantly feed-forward feature extraction. Existing temporal mechanisms typically update templates, prompts, queries, or prediction states, while intermediate representations are rarely reused to modulate corresponding processing stages. We propose \textbfFeedbackTrack, a visual-cortex-inspired framework that introduces sparse, group-level layer-aligned cross-frame feedback into pretrained Transformer trackers. Previous-frame intermediate states are detached, cached, and returned to corresponding Transformer groups in the current frame through two lightweight pathways: Query Feedback for token-level query modulation and Gate Feedback for context-dependent feature modulation. FeedbackTrack preserves the original tracking pipeline with only a fixed-size one-frame cache. Across SPMTrack and ARTrackV2, FeedbackTrack consistently improves five backbone configurations on LaSOT and GOT-10k, achieving 83.4 AO and 79.1 AUC with SPMTrack-G while adding less than 1% parameters. Controlled comparisons show that cross-frame feedback outperforms same-frame modulation by 1.8–3.2 AO points, demonstrating that the gains mainly come from recurrent historical information. Further analysis reveals a non-uniform depth-dependent organization of learned feedback strengths, highlighting the effectiveness of recurrent feedback for Transformer tracking.
[CV-24] Deep Learning based Detection of Fishing Vessels and Fishing Monitoring using Nightlight Images
链接: https://arxiv.org/abs/2608.09360
作者: Shantakar Mohanty,Prasun Kumar Gupta,Raian Vargas Maretto
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Image and Video Processing (eess.IV); Applications (stat.AP)
备注:
Abstract:The demand for maritime surveillance has given rise to the need for monitoring fishing vessel activities, particularly in addressing the challenge of “dark vessels” that operate without Automatic Identification System (AIS) transmission. This study presents a novel approach for detecting small-scale fishing vessels using nighttime light (NTL) imagery from the SDGSAT-1 satellite, combined with deep learning techniques to enhance fishing monitoring awareness along the western coast of India. A dual-branch YOLO11 architecture was developed to exploit both the 10-meter panchromatic and 40-meter RGB imagery from SDGSAT-1. The custom model architecture was specifically optimized for small object detection in NTL imagery, featuring parallel convolutional backbones that process both modalities before concatenation for enhanced feature extraction. The dual-branch YOLO11 model demonstrated optimal performance with a precision of 0.99, recall of 0.93, F1-score of 0.96, and mAP@50 of 0.96, significantly outperforming single-branch implementations of YOLOv5s, YOLOv8s, and standard YOLO11s architectures. When applied to the western coast of India, the model detected 31525 vessel instances across the temporal dataset spanning 2022-23. Cross-matching analysis with AIS data revealed that only 7146 (22.7%) of detected vessels had corresponding AIS transmissions, while 24379 (77.3%) were identified as potential dark vessels. Spatio-temporal analysis showed peak fishing activity during January-April, with a primary activity corridor parallel to the coastline within 50-100 km, corresponding to productive continental shelf areas. This research contributes to maritime surveillance capabilities by highlighting the effectiveness of nighttime lights satellite imagery for fishing vessel detection and provides valuable insights into fishing patterns and potential regulatory compliance issues in Indian waters.
[CV-25] ControlRadio: Prompt-Driven Controllable Diffusion for Cross-Modal Radio Map Generation
链接: https://arxiv.org/abs/2608.09357
作者: Kangjun Liu,Xiying Pan,Shuhang Zhang,Xiang Xiang,Ke Chen,Yaowei Wang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages, 9 figures, 8 tables
Abstract:Radio maps describe how wireless signals propagate across space and are essential for wireless communication, sensing, and network planning. However, constructing accurate radio maps traditionally requires either dense measurements or computationally expensive physical simulations, which limits scalability and real-time deployment. Recent advances in generative artificial intelligence offer a promising alternative, but existing approaches lack fine-grained control and physical consistency when applied to real-world wireless environments. Here we present \textbfControlRadio, a controllable generative framework that produces radio maps from natural-language descriptions and environmental layouts, including building structures and transmitter locations. Joint semantic and spatial conditioning enables interpretable, propagation-plausible generation, while a controlled latent prior and layout-aware conditioning improve stability and structural consistency. Extensive experiments demonstrate that ControlRadio achieves state-of-the-art accuracy and strong generalization across diverse urban scenarios, while reducing computation time by more than four orders of magnitude compared with conventional simulation-based methods. Such results suggest a new paradigm for scalable and controllable wireless environment modeling, with broad implications for next-generation communication systems and data-driven radio sensing.
[CV-26] Alpha as an Efficiency Signal: Visibility-Routed RGBA Image-to-Video Generation
链接: https://arxiv.org/abs/2608.09355
作者: Zhe Li,Honghao Qiao,Zhixin Xu,Qijie Wang,Bo Peng,Dawei Li
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:RGBA videos combine RGB appearance with an alpha channel, enabling animated assets to be applied across arbitrary backgrounds, which are heavily used in gaming industry. However, generating high-quality RGBA animations for games remains challenging for two reasons. First, most existing RGBA video datasets are dominated by photorealistic content, with limited coverage of game assets. Second, the traditional generate-then-matte pipelines estimate alpha only after RGB synthesis, so semi-transparent regions are often blurred by background, resulting in unstable matting outputs. More recently, many methods have begun to model RGB and alpha jointly, but existing approaches are mostly text-conditioned, and still have unresolved issues in efficiency and quality. To address these challenges, we introduce GameAlpha-2.4K, a 2.4K-clip game-style RGBA video dataset built with matte-friendly synthesis, multi-hypothesis alpha recovery, and compositing-based quality gates. Using this dataset, we train a reference-conditioned RGBA video generator that jointly produces RGB frames and alpha mattes in a single pass. To improve efficiency, we propose a visibility router that identifies transparent tokens in an early stage and bypasses their later DiT updates, while x_0-lock guides them along the original flow-matching schedule toward self-predicted endpoints. Our model obtains lower FVD than traditional two-stage pipelines, and the visibility router skips 35% of token evaluations in the final two DiT denoising steps, providing a 1.2x backbone speedup with negligible quality degradation compared to dense inference.
[CV-27] One-Time Training for All Grains: Open-Set Grain Recognition and Quantitative Analysis
链接: https://arxiv.org/abs/2608.09345
作者: Qihe Su,Mengyu Sun,Yuxi Ke,Zhuoyan Jiang,Wanneng Yang,Chenglong Huang,Ziyuan Yang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 15 pages, 15 figures
Abstract:Advances in crop breeding have introduced an increasing number of grain varieties, creating a growing demand for efficient variety recognition and quantitative analysis. However, existing methods are typically trained on a fixed variety set, and incorporating newly introduced varieties requires additional data collection and model retraining. To address this limitation, we propose GROW, a framework for Grain Recognition and quantitative analysis in Open sets Without retraining. GROW first performs class-agnostic grain localization, converting mixed-grain images into individual instances for variety-wise counting and phenotypic measurement. It then combines visual embeddings and morphological descriptors into fused grain descriptors stored in an extensible GrainBank. Query grains are recognized through rank-similarity weighted top-k retrieval, and newly introduced varieties are incorporated by appending their descriptors without updating the deployed models. Extensive experiments under progressive variety expansion, varying grain densities, and background domain shifts demonstrate the scalability, robustness, and adaptability of GROW. Compared with joint retraining, GROW reduced the average category-registration time from 4153 s to only 39 s while maintaining competitive recognition performance. These results demonstrate that GROW provides an efficient and maintainable solution for extensible grain recognition, counting, and phenotypic analysis without repeated model retraining.
[CV-28] Beyond Global Editing: Per-Instance Disentangled Subspaces for Training-Free Hallucination Mitigation in LVLMs BMVC2026
链接: https://arxiv.org/abs/2608.09344
作者: Ali Cheraghian,Hamidreza Dastmalchi,Hamed Barzamini,Morteza Saberi,Mojtaba Golzan,Shafin Rahman,Hossein Rahmani
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: BMVC 2026
Abstract:Recent advances in large vision-language models (LVLMs) have enabled powerful multimodal reasoning by integrating visual encoders with large language models (LLMs). However, their reliability is frequently undermined by hallucinations, where generated text inaccurately describes the visual input. Although fine-tuning can mitigate this problem, it is computationally expensive and requires large, curated datasets, making training-free alternatives attractive. Among these, model editing is more promising than decoding-based approaches: decoding methods adapt outputs per input but introduce computational overhead and instability, whereas model editing modifies internal representations offline, providing a more efficient and stable solution. However, existing model-editing techniques typically rely on a single global subspace to correct hallucinations, treating all test samples identically and failing to capture diverse hallucination modes across inputs. To address this limitation, we propose a training-free hallucination mitigation framework for dynamic, per-instance suppression at test time. Our method first constructs a set of Disentangled Hallucination Subspaces, each isolating a distinct hallucination mode. During inference, the model adaptively calculates weights reflecting each input’s relationship to these subspaces, guiding a dynamically combined projection that selectively suppresses the most probable hallucination directions while preserving image-grounded semantics. Extensive experiments across multiple vision-language benchmarks and LVLM families demonstrate consistent improvements, highlighting the robustness, generalizability, and efficiency of our approach.
[CV-29] Revisiting the Current Frame: Physical-Trace-Guided Network Output Correction for Video Restoration
链接: https://arxiv.org/abs/2608.09342
作者: Yifeng Lin,Liuxiang Qiu,Guangming Ren,Tiesong Zhao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 6 figures
Abstract:Video restoration methods exploit temporal information to recover information missing from degraded observations. However, reference frames within the sequence may introduce inconsistent degradation, content discrepancy, or reconstruction errors due to physical image-formation variations, occlusion, and imperfect temporal aggregation. Existing approaches mainly focus on improving restoration networks, while the reliability of the generated outputs at different spatial locations remains largely unexplored. In this work, we propose ANCHOR, a model-agnostic framework that revisits the low-quality current frame as a temporally aligned anchor for video restoration correction. Specifically, ANCHOR estimates a spatial trust field from heterogeneous physical-trace evidence and adaptively balances the restoration proposal with the original observation. Experiments on High Dynamic Range video reconstruction and video deraining demonstrate consistent improvements across various state-of-the-art restoration models, validating the effectiveness of reliability-aware output correction for video restoration.
[CV-30] SI-Edit: Toward Sketch-Instruction Guided Local Image Editing with Pixel-Level Precision ACM-MM2026
链接: https://arxiv.org/abs/2608.09097
作者: Weixin Ye,Wei Wang,Hongguang Zhu,Xuecheng Nie
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: accepted by ACM MM 2026
Abstract:Despite rapid advances in generative models, achieving pixel-level precision in sketch-based image editing remains a persistent challenge, particularly for fine-grained local deformations. This gap stems primarily from the critical shortage of high-quality, publicly available benchmark datasets that jointly provide geometric constraints and semantic instructions. To address this issue, we first introduce SI-Data, a high-quality dataset specifically designed for instruction-guided local sketch editing. We develop an automated pipeline leveraging Multimodal Large Language Models (MLLMs) to synthesize comprehensive quadruplets comprising original images, local geometric sketches, semantic instructions, and corresponding edited images. By providing both reliable spatial anchors and explicit semantic intent, SI-Data uniquely enables collaborative spatial-semantic learning. Building upon this, we propose a collaborative framework called SI-Edit that integrates semantic instructions with precise geometric constraints. Furthermore, to address the lack of standardized evaluation, we establish a comprehensive set of metrics designed to measure both structural fidelity (e.g., sketch-to-edge alignment) and semantic adherence. Experimental results demonstrate that SI-Edit provides more reliable structural control than baselines for sketch-based image editing, and achieves precise, pixel-level local refinements aligned with user intent. The data and code are released on the [project page](this https URL).
[CV-31] LDChoiceNet: Quantitatively Choosing a Transfer Learning Dataset
链接: https://arxiv.org/abs/2608.09091
作者: Jing Ning,James D. Braza
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Transfer learning is particularly useful in settings with limited training data, and within image classification it is common to transfer learn upon massive datasets like ImageNet , CIFAR-100, or COCO . Qualitatively, it seems a transfer learning dataset should have both more classes and more examples per class than the fine tuning dataset; however, a quantitative method to choose the best transfer learning dataset does not currently exist. In this paper, we design TLDChoiceNet, a model to choose the best transfer learning dataset given a fine tuning dataset by predicting the test-set accuracy after fine-tuning. A simple version 1 achieves 0.154 MSE on the test dataset, while a version 2 leveraging an ImageNet pre-trained ResNet50 v2 embedding with per-class information attains a 5X lower MSE of 0.031. We further design two metrics that enable an unsupervised method of choosing an optimal transfer learning dataset: distribution distance (DD), which linearly regresses against fine-tune accuracy with an R2 of 0.89, and average class correlation (ACC), which improves the R2 to 0.97. Our results underscore that a dataset’s low-level statistics can explain the transfer learning effect, and that using a pre-trained ImageNet can embed different classes further apart in latent feature space.
[CV-32] Learning human joint torques from pixels
链接: https://arxiv.org/abs/2608.09083
作者: Chen Chen,Rui Cheng
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Estimating human joint torques from visual observations is a key step toward bringing biomechanical analysis from controlled laboratories to real-world movement scenarios. Existing torque estimation methods typically depend on surface electromyography, motion-capture markers, force plates, or simulated imitation data, which limits their applicability to ordinary RGB images. In this work, we introduce VID, a vision-based inverse dynamics dataset and benchmark for predicting human joint torques directly from real monocular images. VID contains 63,369 synchronized frames with real human images, kinematic annotations, anthropometric attributes, and OpenSim-derived dynamic labels, providing paired visual and biomechanical supervision for real-image inverse dynamics. We further define a standardized evaluation protocol covering overall torque estimation, joint-specific analysis, and action-specific prediction. To establish a strong reference model, we propose VID-Network, which combines pose-pretrained spatial probabilistic features, marker regression, and temporal torque inference to recover joint torques from image sequences. Experiments on VID show that VID-Network achieves an overall mPJE of 1.7612 N \cdot m/kg, improving over the best compared baseline by 39.81%, and obtains the lowest error across all evaluated joint types and most action categories. VID establishes a first practical benchmark for vision-driven human inverse dynamics and provides a foundation for studying biomechanical inference in less constrained environments.
[CV-33] Model the Edit Not the Image: Visual Autoregressive Editing from a Source-Centric Perspective
链接: https://arxiv.org/abs/2608.09057
作者: Hongyi Fang,Chuwen Xie,Benjia Zhou,Yu-Xuan Qiu,Chenggong Hu,Zhibin Wang,Chao Chen,Jianbin Qin,Rui Mao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Next-scale visual autoregressive models (VARs) have emerged as a powerful generative paradigm, producing high-quality images through efficient coarse-to-fine prediction. However, their potential for text-guided image editing remains largely underexplored. Existing training-free VAR editing approaches often formulate editing as target-conditioned regeneration guided or constrained by the source image, and may rely on inversion, test-time optimization, attention control, or user-provided masks. This generation-centric formulation does not fully exploit the multiscale source representations provided by VARs and may introduce additional computation or intervention. We instead take a source-centric perspective on VAR editing, in which the encoded source image tokens serve as the primary visual state and the editing process focuses on condition-induced changes. Based on this perspective, we propose \textbfEditMod, which compares source- and target-conditioned predictions under a shared autoregressive context, treats their difference as a scale-wise editing direction, and applies it as a residual update to source tokens at selected scales. Experiments show that EditMod achieves leading source-image fidelity while maintaining strong text alignment, and completes end-to-end editing of a 1K image in only 1.57 seconds on a single A100 GPU without per-image preparation.
[CV-34] riple Expert Learning from Noisy Labels for Semi-Supervised Vision Foundation Model Adaptation BMVC BMVC2026
链接: https://arxiv.org/abs/2608.09052
作者: Xuanyu Liu,Zheng Fang,Hongyang He,Yundi Hong,Daizong Liu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted for publication at the British Machine Vision Conference (BMVC) 2026. Official list of accepted papers: this https URL
Abstract:Semi-supervised adaptation of vision foundation models (VFMs) commonly freezes the pretrained backbone and updates lightweight modules such as LoRA. However, pseudo-labels have mixed reliability, and a single LoRA adapter must absorb reliable, ambiguous, and noisy gradients in the same low-rank space. This can make VFM adaptation sensitive to pseudo-label noise. We propose \textbfTriNoL, a \textbfTriple-expert learning framework from \textbfNoisy \textbfLabels for semi-supervised VFM adaptation. TriNoL routes unlabeled samples into three confidence regions and assigns them to three LoRA experts: a Positive Expert for high-confidence pseudo-labels, an Alignment Expert for medium-confidence ambiguous samples, and a Negative Expert for low-confidence noisy samples. The VFM backbone remains frozen, and only the LoRA experts and classifier head are updated. By separating different pseudo-label reliability regions into specialized adaptation paths, TriNoL improves robustness to noisy supervision while keeping the training cost low.
[CV-35] GeoAI-based post-segmentation quality validation of building footprints via spatial feature engineering
链接: https://arxiv.org/abs/2608.09048
作者: Shah Imran Ahsan Chowdhury,Kazi Jihadur Rashid,Rajsree Das Tuli,Rahul Saha,Bulbul Ahammad
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 19 pages, 10 figures, 8 tables
Abstract:Deep learning-based building footprint extraction from high-resolution imagery often produces topologically inconsistent vectors unfit for direct GIS database ingestion. To address this, we present a multidomain GeoAI quality control framework that automates error detection to systematically purify vector footprint databases. Candidate footprints were generated across five UAV survey sites in Bangladesh using U-Net (ResNet-34) and SAM-LoRA (ViT-B). The extracted raster masks were vectorized, geometrically regularized, and consolidated under a spatial-exclusivity constraint to eliminate duplicate representations. We used twenty-four predictors capturing geometric, spatial-contextual, and raster-derived spectral and texture properties. Machine Learning (ML) classifiers were trained on a development partition (Sites B-D) and rigorously validated on a spatially independent test set (Site E) excluded from hyperparameter tuning and class balancing. The experimental results demonstrate that geometric and spatial-contextual predictors using Decision Tree (DT) provide the most effective discriminatory evidence for identifying object-level boundary deformations. DT achieved an accuracy of 95.31%, an F1-score of 91.06%, and a Matthews correlation coefficient (MCC) of 0.880 on the unseen testing site. At the database level, this framework successfully identified 87.34% of erroneous footprints while maintaining 98.31% of acceptable structures, reducing the residual error proportion from 27.32% to 4.62% and improving final database purity to 95.38%. This translates into a relative error reduction of 83.09%. The findings indicate that post-segmentation object-level ML provides a highly transferable, robust mechanism for automated quality assurance in production-ready geographic information system (GIS) workflows.
[CV-36] Diversity Matters: Distributional Feature Coverag e Sample Selection for Data-Efficient Backdoor Attacks
链接: https://arxiv.org/abs/2608.09047
作者: Yi Yang,Xiaoke Chen,Jinyang Huang,Feng-Qi Cui,Yu-Tong Guo,Jia-Cheng Zhao,Haiming Jin,Xiaokang Zhou,Meng Li
类目: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Backdoor attacks compromise training data so that a model retains clean accuracy but predicts an attacker-chosen target on triggered inputs. At very low poisoning rates, only a few samples convey the trigger–target association, making poison-sample selection critical. Existing methods typically rank candidates using per-sample scores, which can select redundant samples from similar semantic regions, and many require task-specific surrogate training. We propose Distributional Feature Coverage Sample Selection (DFCS), a training-free, trigger-agnostic method that clusters fixed pretrained features into one region per poisoning slot and selects the centroid-nearest sample from each region. A local first-order analysis relates this allocation to feature-coverage and representative-mass terms. Across BadNets and Blended attacks on CIFAR-10, Tiny-ImageNet, and Imagenette, DFCS achieves the highest mean attack success rate among seven selectors in all six dataset–attack settings, averaging 96.30% and exceeding the strongest comparator in each setting by 4.60 percentage points on average while preserving clean accuracy. These results support distributional feature coverage as an effective selection principle for low-budget dirty-label backdoor attacks.
[CV-37] LogiShot: Logically Coherent Cross-Shot Video Generation
链接: https://arxiv.org/abs/2608.08820
作者: Shuai Guo,Yuhang Yang,Zeyu Zhang,Pengfei Yu,Wei Zhai,Yang Cao,Zheng-Jun Zha
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Generating cross-shot videos that are logically connected is essential for content creation. Currently, most cross-shot video-generation workflows, such as short-drama production, still rely on isolated textual scripts or explicit reference images to specify the generated content. Consequently, when user instructions are underspecified or ambiguous, a generated clip may appear visually plausible on its own but fail to align with the overall narrative, leading to disjointed content. We argue that achieving cross-shot logical coherence in video generation requires establishing logical connections across shots and maintaining visual consistency. To this end, we propose LogiShot, which incorporates information through two complementary paths: 1) LogiShot jointly encodes the context video and other conditioning signals, yielding dense multimodal cues that provide visual-semantic evidence for cross-shot generation; 2) the model maintains a visual memory of the context video throughout generation to preserve visual consistency across shots. Additionally, we construct a dataset with 110K samples and a dedicated benchmark for evaluating cross-shot logical coherence. Experiments demonstrate that LogiShot consistently outperforms existing baselines in terms of logical coherence across multiple shots. Model and data will be made publicly available.
[CV-38] MRI super-resolution in ten sampling steps using a diffusion bridge model
链接: https://arxiv.org/abs/2608.08819
作者: Mojtaba Safari,Hang Yu,Zach Eidex,Mingzhe Hu,Ryan J. Sanford,Alexandru Florea,Shansong Wang,Chih-Wei Chang,Erik H Middlebrooks,Aditya Juloori,Stanley L. Liauw,Ralph Weichselbaum,Xiaofeng Yang
类目: Computer Vision and Pattern Recognition (cs.CV); Medical Physics (physics.med-ph)
备注:
Abstract:Objective. MRI provides excellent soft-tissue contrast, but long acquisition times can cause patient discomfort and lead to motion artifacts, forcing a trade-off between spatial resolution and scan time. Diffusion-based super-resolution (SR) reconstructs high-resolution (HR) images from low-resolution (LR) inputs, but typically needs many sampling steps and initializes from a Gaussian prior ill-suited to image restoration. We developed an efficient diffusion framework that reconstructs HR MRI directly from LR data. Approach. We propose super-resolution diffusion bridge model (SR-DBM), a super-resolution diffusion bridge model that casts SR as a stochastic transport between the LR and HR image distributions. Through a Doob’s h-transform of a mean-reverting stochastic differential equation, SR-DBM pins the process to the paired HR and LR images at its endpoints, initializing reconstruction from the measured anatomy rather than from Gaussian noise. The HR image is recovered by a deterministic reverse trajectory in which a network predicts the clean image at each of only ten sampling steps. We evaluated SR-DBM on ultra-high-field 7T brain T1 MP2RAGE maps and pelvic T2-weighted prostate images against nine comparison methods using PSNR, SSIM, GMSD, and LPIPS. Main results. SR-DBM attained the highest PSNR and SSIM and the lowest GMSD on both datasets (brain: 27.66±1.52 dB, 0.96±0.02, 7.96±1.86 ; prostate: 27.87±2.29 dB, 0.80±0.05, 8.38± 1.44), with statistically significant gains over every comparison method (two-sided Wilcoxon signed-rank test with Holm correction, p0.05). The strongest baseline, SR-EMamba, ranked second. Qualitatively, SR-DBM produced the smallest residual errors and best preserved fine structures and lesions.
[CV-39] 360CityArena: A Realistic Virtual Urban Navigation Benchmark for Embodied Agents ECCV2026
链接: https://arxiv.org/abs/2608.08814
作者: Kenta Watanabe,Atsuyuki Miyai,Mizuki Takenawa,Kiyoharu Aizawa,Toshihiko Yamasaki
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注: ECCV2026. Project Page: this https URL
Abstract:We present 360CityArena, a benchmark for evaluating the urban exploration capabilities of embodied agents within a photorealistic environment constructed from 360-degree videos. Existing outdoor benchmarks either lack sufficient photorealism or complexity, resulting in a considerable gap from real-world urban environments. 360CityArena is built on a realistic reconstruction of the Akihabara district in Tokyo, Japan, using 602 360-degree video segments covering 85 streets, and consists of 175 meticulously human-crafted tasks. It encompasses three task categories: Environment Understanding, Path Reasoning, and Spatial Reasoning, covering fundamental abilities required for urban exploration, such as localization, landmark search, path planning, and relational spatial reasoning, thereby enabling comprehensive evaluation in realistic urban scenes. Our evaluation using state-of-the-art LMM-based agents shows that even the strongest model, Gemini 2.5 Flash, performs far below human level (human: 77.3% vs. Gemini 2.5 Flash: 17.1%), revealing substantial challenges that remain in city-scale embodied navigation and reasoning. 360CityArena provides a necessary and challenging testbed for photorealistic urban-district navigation and spatial reasoning.
[CV-40] AdapterMoE: A Two-Stage Hard-Routing Mixture-of-Experts Architecture for Multi-Crop Disease Recognition with Calibrated Rejection and Incremental Learning
链接: https://arxiv.org/abs/2608.08808
作者: Pin-Hsun Huang,Shaou-Gang Miaou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 17 pages, 7 figures, 14 tables
Abstract:Timely crop-disease identification is critical to food security. Multi-crop recognition suits Mixture-of-Experts (MoE), but conventional soft-routing MoE learns crop assignment freely end-to-end, letting a few experts dominate (expert collapse) with no semantic correspondence to crops, and facing high retraining costs, unstable rejection of non-target inputs, and a saturated accuracy ceiling. We shift the objective from accuracy toward a trade-off among deployment cost, scaling flexibility, and rejection stability, using deterministic hard routing. We propose AdapterMoE: a RouterHead classifies the crop and rejects non-target crops via a Maximum Softmax Probability threshold, with a dual-gate Energy+KNN out-of-distribution module catching distribution-shifted inputs; five per-crop Adapters atop a frozen EfficientNet-B0 backbone discriminate diseases, each calibrated via Temperature Scaling. Because experts are hard-isolated at the data level, the design avoids expert collapse and exposes an add_crop interface for local, per-crop updates instead of full retraining. On PlantVillage (5 crops, 26 classes), across a fair five-system comparison, AdapterMoE attains accuracy statistically indistinguishable from the best baselines (Macro-F1 within a 0.24-point band) while cutting training cost to about 9% of full-network baselines, expanding to a new crop in
[CV-41] LASA: Language-and-Source-Anchored Alignment for Domain Generalized Semantic Segmentation
链接: https://arxiv.org/abs/2608.08805
作者: Jinhong Zhu,Weiqi Yan,Shengchuan Zhang,Liujuan Cao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 10 pages, 4 figures
Abstract:Domain Generalization Semantic Segmentation (DGSS) focuses on generalizing knowledge from labeled source domains to unseen target domains where data is unavailable during the training phase. While conventional methods utilize style randomization or feature normalization to mitigate domain shifts, they often impair feature integrity. Specifically, style randomization distorts the underlying feature manifold due to its coarse-grained nature, while feature normalization suppresses discriminative, domain-sensitive semantic details owing to its rigid design. To address these limitations, we propose the Language-and-Source-Anchored Alignment (LASA) framework, which comprises three synergistic components: Text-and-Source-Guided Style Transfer (TSGST), Domain-Aware Query Adapter (DAQA), and Domain-Aware Decoder Optimizer (DADO). Concretely, the TSGST module addresses manifold distortion by utilizing source features as structural anchors and vision-language model (VLM) priors as fine-grained guidance. To restore suppressed discriminative and domain-sensitive details, the DAQA module recalibrates object queries via categorical guidance and domain-aware signatures, while the DADO module aligns the resulting query distributions with a shared classifier to ensure consistent categorical responses across domains. Extensive experiments on challenging benchmarks demonstrate that our method significantly outperforms state-of-the-art approaches.
[CV-42] UPolarSQ: Polar Representation Learning for Optic Disc and Peripapillary Atrophy Segmentation and Quantification in Fundus Photographs
链接: https://arxiv.org/abs/2608.08771
作者: Mengxian He,Yunyun sun,Ziyue Gao,Wengkei Lam,Shunyi Zhang,Wu Yuan
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Myopia-induced posterior-pole remodeling is frequently accompanied by Optic Disc (OD) deformation and Peripapillary Atrophy (PPA), both of which provide clinically relevant structural biomarkers. In Cartesian fundus images, however, PPA often appears as an irregular and partially visible crescent adjacent to the OD, leading to fragmented segmentation and post-processing-dependent quantification. We propose UPolarSQ, a unified polar-domain framework for OD/PPA segmentation and biomarker quantification in myopic fundus images. UPolarSQ first maps an OD-centered region of interest into polar coordinates, where OD and PPA boundaries can be represented as radial profiles. It then employs UPolarSeg, a U-Net-based segmentation network enhanced with a Radial-Angular-Decoupled Module and boundary-aware auxiliary supervision to model anisotropic polar features and radial boundary transitions. Clinical biomarkers, including disc shape and PPA-width-related measurements, are deterministically extracted from the predicted polar masks, aligning segmentation and quantification within a shared geometric representation. Experiments on internal and external cohorts demonstrate that UPolarSQ improves OD/PPA segmentation and supports reliable polar-native biomarker estimation for myopic analysis.
[CV-43] Parcel2Progression: An Anatomy-aware Longitudinal Framework for Alzheimers Disease Diagnosis
链接: https://arxiv.org/abs/2608.08753
作者: Madhumitha Venkatesh,Shanawaj S Madarkar,Konda Reddy Mopuri
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Alzheimer’s disease (AD) progression is a longitudinal process with subtle pathological cues in the early stages. Yet, computational constraints have limited most neuroimaging models to either compromise spatial information or limit the number of longitudinal scans. We aim to overcome this bottleneck and fully leverage high-resolution, variable-length T1w structural MRI (4D sMRI) scan sequences. We introduce Parcel2Progression (P2P), a Longitudinal Transformer Framework which tackles this challenge using an Atlas-guided Parcel Encoder that tokenizes 3D scans into a set of richer anatomically grounded representations. A Longitudinal Transformer then integrates irregular, arbitrary-length longitudinal visits with patient age. This synergy delivers two key advantages: (1) parcel-specific interpretability, and (2) computational tractability for long-term analysis, which scales linearly with the number of scans compared to a naive quadratic 4D ViT cost. P2P outperforms prior works and baselines in both MCI (Mild Cognitive Impairment) to AD conversion prediction and AD vs. CN (Cognitively Normal) classification tasks across ADNI, AIBL, and MIRIAD datasets. Leveraging longitudinal scans boosts performance over single-scan baselines by up to 5% and 7% in balanced accuracy for AD classification and MCI conversion prediction tasks, respectively. Interpretability analysis using parcel saliencies and attention rollouts reveals clinically consistent atrophy patterns in AD and MCI subjects. We also demonstrate the frameworks’ reliability in anomaly detection using a synthetic dataset, and test the model’s generalizability for other neurodegenerative diseases like Frontotemporal Dementia.
[CV-44] IDATA: Scalable Invertible Diffusion for Unrestricted Adversarial Transfer Attack
链接: https://arxiv.org/abs/2608.08734
作者: Yi Pan,Jun-Jie Huang,Tianrui Liu,Zihan Chen,Lin Liu,Zhao Wentao
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: Adversarial attack,Invertible diffusion model,Memory-efficient,Low-frequency
Abstract:Unrestricted adversarial transfer attacks are important for evaluating the black-box robustness of deep visual models. Diffusion-based attacks have shown promising transferability and visual imperceptibility by optimizing adversarial perturbations along denoising trajectories in latent space. However, existing methods are limited by two challenges: memory-intensive multistep backpropagation and frequency-agnostic perturbation over intermediate latents. To address these issues, we propose IDATA, a memory-efficient diffusion framework for unrestricted adversarial transfer attack. IDATA consists of two key components: an Invertible Diffusion Module (IDM) and a Low-Frequency Constraint Module (LFCM). Specifically, IDM reformulates adversarial optimization over diffusion trajectories as an invertible process, enabling constant-memory backpropagation through on-demand reconstruction of intermediate states instead of storing the full denoising chain. Moreover, LFCM leverages Discrete Wavelet Transform (DWT) to decompose latent variables into low- and high-frequency components, restricting perturbations to semantically stable low-frequency subspaces, thereby improving transferability while preserving visual imperceptibility. Extensive experiments on multiple benchmarks and diverse model architectures demonstrate that IDATA consistently outperforms state-of-the-art baselines in attack success rate, memory efficiency, and visual imperceptibility. These results suggest that IDATA is a promising tool for black-box robustness evaluation of deep visual models. Code is available at this https URL.
[CV-45] omaMMU: A Comprehensive Multimodal Understanding Benchmark for Tomato Leaf Diseases ECCV
链接: https://arxiv.org/abs/2608.08727
作者: Gia-Han Truong,Khang Nguyen Quoc,Luyl-Da Quach
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: Accepted at ECCV CVPPA Workshop
Abstract:To address this gap, we introduce TomaMMU, a large-scale Tomato leaf disease MultiModal Understanding dataset, alongside TomaBench, a benchmark for evaluating VLMs on tomato disease understanding. TomaMMU comprises 28,808 high-quality images spanning 15 categories and 213,119 human-annotated visual question-answer pairs, generated through a three-stage pipeline comprising Data Collection, Human Annotation, and Question-Answer Generation. Building on this foundation, TomaBench organizes seven agricultural tasks into a hierarchical three-level taxonomy spanning Basic Perception, Pathology Understanding, and Expert Diagnosis, which together enable systematic evaluation from low-level visual recognition to high-level diagnostic reasoning. The tasks assess visual symptom recognition, taxonomic relationships, and diagnostic reasoning, offering a comprehensive view of how well models grasp plant pathology. Our results pronounced gaps in fine-grained recognition and factually grounded reasoning with 14 state-of-the-art VLMs, consistently underperforming on both challenging MCQs and open-ended questions. These results suggest that current VLMs struggle to translate visual perception into reliable diagnostic knowledge, motivating the need for targeted domain adaptation. Simple fine-tuning on TomaMMU substantially narrows this gap, boosting accuracy on challenging MCQs to 96.09%, outperforming recent VLMs, and pointing toward promising directions for future work. All data and code is available in this https URL.
[CV-46] NewtonGS: Physics-Structured Object-Level Neural Newtonian Dynamics for Gaussian Scene Animation
链接: https://arxiv.org/abs/2608.07598
作者: Lianlei Shan,Feiyang Ye,Yan Chen,Yong Wu
类目: Computer Vision and Pattern Recognition (cs.CV)
备注: 40 pages, 15 figures
Abstract:Animating objects in a static 3D Gaussian scene requires an explicit object-level dynamic state and a controllable model of object motion. Existing dynamic Gaussian methods primarily reconstruct time-varying scenes or simulate deformation, rather than provide compact object states for direct control. To address this gap, we present NewtonGS, a physics-structured framework for object-level state rollout and Gaussian scene animation. NewtonGS represents each object with a 22-dimensional state covering pose, linear and angular velocity, anisotropic scale and its rate, mass, and contact properties. Its Gaussian Neural Newtonian Dynamics (Gaussian-NND) model combines analytic translation, quaternion kinematics, gravity, damping, and scale-restoration dynamics with learned continuous and contact residuals. A discrete event map handles floor contact. Predicted poses and scales define a shared affine transformation that updates the means and covariances of all Gaussians associated with each object. We construct two procedurally generated datasets: State-32 for state-rollout evaluation and Gaussian-32 for state-to-Gaussian transformation. On both the in-distribution and velocity-range-shift splits of State-32, NewtonGS achieves lower trajectory RMSE, final displacement error, and velocity RMSE than five analytic baselines. Experiments on Gaussian-32 further demonstrate effective conversion from predicted states to animated Gaussian objects.
[CV-47] LIRA: Local Cross-Layer Information Routing for Vision-Language-Action Decoding
链接: https://arxiv.org/abs/2608.07596
作者: Zhewei Zhang,Puyue Wang,Guanren Qiao,Yijie Weng,Jiawei Hu,Guo Li,Lujia Wang,Junyan Wang,Tao Gu,Hongliang Lu,Guiliang Liu,Hong Jia,Xinhu Zheng
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 9 pages, 4 figures. Code and model checkpoints will be released upon acceptance of the paper
Abstract:Vision-Language-Action (VLA) models transform representations from pretrained vision-language models (VLMs) into robot actions, yet the interface that routes intermediate VLM features into action decoders remains underexplored. Existing designs either expose only a narrow part of the representation hierarchy or rigidly match each decoder block to one VLM layer, restricting access to complementary task evidence across depths. We introduce LIRA, a local cross-layer action-conditioning mechanism that formulates VLM-to-action conditioning as depth-aware information routing. LIRA operates on task-token features and LIRA Query features derived from intermediate VLM states, then assigns each Parallel Fusion Block a depth-aligned local window centered on its corresponding VLM layer. Parallel Fusion Blocks aggregate neighboring LIRA Query features and integrate them with task-token features and proprioceptive inputs before action prediction. This routing interface leaves the backbone architecture, action decoder, and supervised training recipe unchanged. Across LIBERO, LIBERO-Plus, CALVIN ABC \rightarrow D, and real-world manipulation, LIRA improves the principal aggregate metrics over the VLA-Adapter baseline under the same 0.5B-parameter configuration. In zero-shot transfer to LIBERO-Plus, LIRA increases average success from 59.1% to 78.0%, an 18.9-point gain indicating improved robustness under controlled distribution shifts.
[CV-48] MAGIC-SSCIL: Manifold Anchoring and Geometric Incremental Calibration for Semi-Supervised Class Incremental Learning
链接: https://arxiv.org/abs/2608.07586
作者: Yousef Abdi,Mohammad Asadpour,Yousef Seyfari
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Semi-supervised Class Incremental Learning (SSCIL) is a severe challenge for neural networks, and it is hardest in the exemplar-free setting where no past data may be stored. Existing methods forget catastrophically due to feature drift, and their pseudo-labels become increasingly unreliable as the label space grows. In this paper, we propose MAGIC (Manifold Anchoring and Geometric Incremental Calibration), a framework that stabilizes plasticity without storing exemplars. MAGIC’s design centers on two components. The first is Soft-Weighted Geometry Calibration (SWGC), which uses graph-based label propagation on the learner’s plastic feature space to weight and calibrate class means and variances computed on the frozen backbone; from these calibrated Gaussians, we sample phantom features that stand in for data from previous tasks. The second is a Geometric Structural Alignment (GSA) objective that preserves representation topology by matching the relational structure of student and teacher heads and aligning feature anchors with the fixed classifier prototypes, locking the orientation of the feature space. Together, these constraints keep the adapter from drifting, so geometric relations between classes remain stable as new classes arrive. We implement MAGIC with a frozen ResNet-18 backbone and a learnable plastic adapter. Across CIFAR-100, CUB-200, and ImageNet-R, at label ratios of 1%, 5%, and 10%, MAGIC improves average incremental accuracy over most of the supervised CIL methods equipped with FixMatch and native SSCIL baselines; the largest gains occur in the fine-grained, low-label setting, where confidence thresholding fails most clearly.
[CV-49] ComplexityWorld: Benchmarking Vision-Language Models on Verifiable Visual Decision Making
链接: https://arxiv.org/abs/2608.07584
作者: Ningxin Pan,Hanyu Li,Yehui Tang
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Vision-language models (VLMs) have made rapid progress in visual perception and increasingly support real-world tasks that depend on images. Many such tasks, however, require more than rec- ognizing what an image contains: a model must use visual evidence to make a complete decision whose parts jointly satisfy global constraints. We introduce COMPLEXITYWORLD, a benchmark of 390 tasks across 39 domain-inspired visual worlds and 29 decision categories. Each task is generated from a hidden structured specification, rendered as a visual scene, and scored by an exe- cutable verifier that accepts any feasible solution. Under direct inference, all evaluated models ex- cept GPT-5.6-Sol remain below 40% verifier ac- ceptance rate (VAR), while GPT-5.6-Sol reaches 75.6%. Performance improves substantially when the same decision information is made explicit in structured form, yet varies sharply across equiva- lent visual presentations. Agent scaffolds provide smaller, model-dependent gains. Together, these results reveal a persistent visual-to-decision bot- tleneck that additional inference alone does not remove.
[CV-50] Predictive Failure Detection in Network Hardware Using Thermal Imaging and Deep Learning with Sensor Fusion
链接: https://arxiv.org/abs/2608.07582
作者: Ashly Joseph
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Unplanned network hardware malfunctions can interrupt services and result in expensive downtime in data centers. A deep learning-based predictive maintenance strategy is presented that utilizes thermal imaging and power sensor data to detect early indicators of equipment breakdown in routers, switches, and servers. A simulated dataset was generated comprising annotated thermal pictures and power readings indicative of three operating states: Normal, Warning, and Critical. Three ImageNet-pretrained convolutional neural network (CNN) models ResNet-50, InceptionV3, and VGG16 were assessed together with a multi-modal CNN-LSTM fusion model that integrates visual and sensor time-series information. Experiments were performed with and without pre-processing procedures, including region-of-interest (ROI) extraction and normalization. In the absence of pre-processing, CNNs attained moderate accuracy (e.g., ResNet-50 at 52%), but ROI-based pre-processing significantly enhanced performance (ResNet-50 accuracy reaching 91%). The CNN-LSTM model attained the greatest accuracy of 94%, with precision and recall approaching 95%, illustrating the effectiveness of multi-modal fusion. The results validate that domain-specific pre-processing and sensor fusion substantially improve early failure prediction, providing a potential foundation for proactive maintenance of network hardware through non-intrusive monitoring.
[CV-51] Multi-Branch Policy Optimization for Multimodal Large Language Models
链接: https://arxiv.org/abs/2608.07581
作者: Shuai Lyu,Yuning Gong,Ruiling Gao,Xiaoran Shang,Zhonghong Ou,Ping Zong,Yifan Zhu,Yuan Sun,Yang Qin,Peng Hu
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 10 pages,8 figures
Abstract:Group-based reinforcement learning methods for multimodal large language models typically rely on trajectory-level credit assignment that applies a single advantage to all tokens in a response. However, multimodal reasoning involves substantially higher perceptual uncertainty than text-only settings, where the model must repeatedly re-examine visual information to verify intermediate interpretations, and different visual groundings can lead to divergent reasoning paths, making such uniform credit assignment particularly inadequate and causing relative advantages to progressively degenerate toward zero. To address these challenges, we propose Multi-Branch Policy Optimization (MBPO), a tree-based framework that constructs reasoning trees at vision-language decision boundaries, enabling sibling branches to explore diverse visual hypotheses and assigning segment-level credit through branch-relative advantages. We further introduce a temporal replay buffer to reuse informative segments while controlling policy staleness. Experiments on several multimodal reasoning benchmarks show that MBPO outperforms representative baselines, improving both learning signal quality and optimization efficiency. The code is publicly available at this https URL.
[CV-52] Real-time physics inversion for retrieval of sub-pixel wildfire temperatures from VSWIR imaging spectroscopy
链接: https://arxiv.org/abs/2608.07580
作者: William R. Keely,Philip G. Brodrick,Katherine Mistick,Adam Chlus,Robert O. Green,Philip E. Dennison
类目: Computer Vision and Pattern Recognition (cs.CV); Instrumentation and Methods for Astrophysics (astro-ph.IM); Machine Learning (cs.LG)
备注: In Review in Remote Sensing of Environment
Abstract:In this work, we present a wildfire temperature retrieval framework for VSWIR imaging spectroscopy data, employed on data from NASA’s Airborne Visible Infrared Imaging Spectrometer (AVIRIS-3). The retrieval framework utilizes a full-physics approach in which a forward model is employed to resolve both solar and emitted radiance derived from a temperature distribution and utilizes the full spectral range in the residual fit. To optimize the forward model retrieval, we use state-of-the-art nonlinear least squares methods implemented for fast convergence on the on-board GPU, allowing for estimation of effective fire temperature within flight cadence. We verify the forward model assumptions on simulated spectra with an injected thermal signature and find good agreement with an RMSE of 41.8 Kelvin (K). We apply the retrieval over the full 2025 FireSense AVIRIS-3 campaign, totaling 168 overflights with probable active fire spectra, and demonstrate a residual radiance fit of \leq 10% across bands in the short-wave infrared (SWIR). Lastly, we verify the applicability of the retrieved posterior fire temperature parameters to generalize to space-borne imaging spectrometers such as EMIT, by retrieving at coarsened spatial resolution. We find that the posterior distribution exhibits good coverage of the underlying sub-pixel temperature range with an absolute error of 30 K across quantiles and a mean absolute error of 27.16 K between spatial resolutions.
[CV-53] Geometry Beats Estimated Depth: RGB-Only Multi-Camera 3D Tracking under Sim2Real
链接: https://arxiv.org/abs/2608.07579
作者: Abdullah Naeem,Anav Katwal,Ayon Dey,Noman Khan,Md Tamjidul Hoque
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:The AI City Challenge 2026 Track 1 evaluates multi-camera 3D perception in large indoor warehouses under a synthetic-to-real (Sim2Real) setting; depth is available only for training and validation, so inference is RGB-only. We use two RGB-only routes as a controlled test of one hypothesis: that cross-view geometric consistency, not monocular depth accuracy, governs performance under Sim2Real. The first is a geometry-first pipeline: YOLO11x detection, homography lifting to the world frame, class-level 3D size priors, multi-camera fusion, world-coordinate tracking, and offline tracklet stitching. The second is estimated-depth pseudo-LiDAR: monocular depth (D4RT, Metric3D~v2) back-projected into a fused point cloud and passed to a 3D detector (V-DETR), mirroring prior point-cloud winners that used depth. The gap is decisive: geometry-first reaches 13.0 3D HOTA (51.6 LocA), whereas pseudo-LiDAR collapses to 0.12 (9.2 LocA). We trace the collapse to cross-view inconsistency of monocular depth—scale correction is necessary but not sufficient—which domain-adaptation fine-tuning does not repair within budget. Within the geometry pipeline, offline stitching is the only intervention that helps; SAHI detection, appearance Re-ID, learned lifting, RT-DETR ensembling, test-time augmentation, and domain randomization all fail to beat the baseline detector. The bottlenecks are complementary: detection quality bounds the geometry route (DetA), localization consistency bounds pseudo-LiDAR (LocA). We release a complete, reproducible RGB-only pipeline and ablation.
[CV-54] Open-World Hierarchical Perception: Taxonomic Abstraction over Class-Agnostic Proposals for the Safe Handling of Out-of-Vocabulary Road Objects
链接: https://arxiv.org/abs/2608.07577
作者: Felix Schaller
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注: 6 pages, 4 figures, 1 table. Third paper in a series; v1 archived at Zenodo, doi: https://doi.org/10.5281/zenodo.21593472 . Code: this https URL
Abstract:A closed-set detector for autonomous driving must assign every object one of a fixed set of labels. On an object outside that set (a horse-drawn carriage, road debris, livestock on a rural road) it can only force a confident but wrong specific label or drop the object. Prior work in this series replaced the flat label set with a hierarchical taxonomy and a runtime abstraction rule, but evaluated it only on the boxes a closed detector already produces. This paper takes the layer open-world: we place taxonomic abstraction on top of class-agnostic region proposals so objects the closed detector never boxes can still be classified or flagged; we report a feasibility study of three open-world signals (class-agnostic segmentation, appearance-based out-of-distribution scoring, monocular depth) that shows why no single 2D cue suffices and how they compose; and we run the evaluation the earlier papers could not, a ground-truth leave-classes-out benchmark on real annotated objects. Holding out seven COCO classes and classifying their 235 ground-truth crops, a flat closed head emits a confident wrong specific label 100% of the time (37% of them in the wrong super-category, e.g. an animal named as a vehicle), whereas the hierarchical layer emits zero confident wrong specific labels and safely handles 94% of the objects (a correct super-category, or an explicit UNKNOWN OBSTACLE). We are explicit that this is a safety result, not a specificity one: the correct super-category is recovered only 26% of the time and the remaining 69% are conservatively flagged unknown. The contribution is an open-world perception layer that never makes a confident categorical mistake on an out-of-vocabulary object, together with an honest account of its cost.
[CV-55] Beyond Isotropic Assumptions: Continuity-Constrained Segmentation and GPU Morphometry for Nanoscale GBM Analysis
链接: https://arxiv.org/abs/2608.07575
作者: Arash Fatehi,Robin Ebbestad,Linus Butt,Hans Blom,Sigrid Lundberg,Hannes Olauson,Hjalmar Brismar,David Unnersjö-Jess,Thomas Benzing,Katarzyna Bozek
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注:
Abstract:Confocal microscopy of optically cleared and swelled tissue resolves complex biological structures in 3D, but such acquisitions are highly anisotropic: along the under-sampled axial direction the structure can appear discontinuous, hampering reconstruction and automated quantitative analysis. The usual remedy upsamples the axial dimension to an isotropic volume before training a segmentation model, which requires dense annotations in the upsampled space, a prohibitive labeling burden. We present an end-to-end, GPU-accelerated framework that overcomes this without additional annotations. The model is trained on the native acquisition volume; random rotation of training patches leverages the well-resolved lateral plane to supply the missing axial information, and a z-axis continuity loss keeps neighboring slices consistent. We adapt both a convolutional (3D U-Net) and a transformer (SwinUNETR) backbone, aggregate overlapping patches by Gaussian consensus, and compute point-spread-function-corrected membrane thickness by ray-surface intersection on the GPU. We apply the method to the glomerular basement membrane (GBM), a thin, highly convoluted part of the kidney’s filtration barrier that grows more irregular in disease. Segmentation accuracy matches inter-expert agreement. Continuity-aware training improves reconstruction smoothness and suppresses a periodic terracing artifact at minimal accuracy cost. We quantify GBM thickness across the reconstructed 3D surface and capture disease-related thickening, enabling fully automated anisotropic 3D morphometry of biological structures without dense volumetric labels or image restoration. Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) Cite as: arXiv:2608.07575 [cs.CV] (or arXiv:2608.07575v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.07575 Focus to learn more arXiv-issued DOI via DataCite
[CV-56] Multimodal Skin Lesion Classification with Swin Transformer and Clinical Metadata Fusion
链接: https://arxiv.org/abs/2608.07574
作者: Nethmi Pathirana,Isuru Munasinghe,Dileeka Alwis
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注:
Abstract:Skin lesion classification plays an important role in supporting the early diagnosis of skin cancer. However, automated analysis remains challenging due to class imbalance, inter-class similarity, and intra-class variability in dermoscopic images. This paper proposes a multimodal classification framework that combines Swin Transformer-based image features with structured clinical metadata to improve diagnostic performance through integrated visual-context learning. Experiments on a publicly available dataset show that the proposed model achieves a test accuracy of 92.55% and a macro F1-score of 91.33%, with strong performance across minority classes. Temperature scaling is applied as a post-hoc calibration method, resulting in a reduction in expected calibration error and improving prediction reliability, while uncertainty estimation is incorporated to further assess the confidence of model predictions. Qualitative explainability analysis further shows that the model focuses on lesion regions during inference. Therefore, the results demonstrate that multimodal fusion, combined with calibration and interpretability analysis, provides an effective and trustworthy approach for automated skin lesion classification.
[CV-57] BRACE: Taming Sharp Irregularities via Barycentric Rational Forecasting for Fast Diffusion Transformers Inference FAST
链接: https://arxiv.org/abs/2608.07572
作者: Jinlong Yang,Jinke Wu,Lizilin,Yao Zhou
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 10 pages, 6 figures, 5 tables. Project page: this https URL
Abstract:Diffusion Transformers (DiTs) have demonstrated exceptional performance in high-fidelity image and video generation. To alleviate their massive computational overhead, temporal feature caching has been proposed to bypass redundant computations. However, existing cache-then-forecast methods driven by derivative-based polynomials often cause severe quality degradation under high acceleration due to unstable long-step predictions. To address this bottleneck, we propose Barycentric Rational Forecasting with Chebyshev Enhancement (BRACE). Motivated by the observation that DiT feature trajectories are globally smooth yet frequently exhibit sharp irregularities and local non-smoothness, BRACE shifts the paradigm from derivative-driven polynomial extrapolation to feature-driven rational forecasting. Specifically, it maintains a local sliding window to cache sparse historical features and leverages adapted Chebyshev weights to formulate a barycentric rational function, directly aggregating these raw features to ensure numerical stability. Extensive experiments demonstrate that BRACE achieves state-of-the-art quality-efficiency trade-offs across various DiT architectures with negligible computational overhead.
[CV-58] A Review of Vision-Based Vehicle Detection for UAV-Based Traffic Monitoring: Experimental Insights and Future Directions
链接: https://arxiv.org/abs/2608.07571
作者: Jianlin Ye,Christos Kyrkou
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:In Intelligent Transportation System (ITS), unmanned aerial vehicle (UAV)-based surveillance offers an innovative solution to traffic surveillance with wide coverage and real-time data collection capabilities. In comparison to fixed ground-based infrastructure, UAVs are able to respond to dynamic traffic but present challenges such as vehicle detection at varying altitudes, compensation for motion-induced image variations and efficient processing of high-resolution images. Deep learning has been largely beneficial on improving the detection accuracy; however, for practical deployment, a critical assessment of the accuracy, latency, and harmonization with current transportation systems needs to be carefully considered. This survey reviews recent advancements in the UAV-based traffic monitoring, with a primary focus being deep neural network models for traffic analytics in various urban settings. Three main challenges identified in the literature are ensuring compatibility with traffic control systems, achieving real-time processing to optimize traffic flow, and maintaining robust detection in different environmental conditions. Existing solutions often lack comprehensive frameworks for utilizing UAV captured data to respond to incidents and manage traffic effectively. Future research should focus on optimal detection models, edge processing, and adaptive control integration to improve the responsiveness of urban traffic management.
[CV-59] COMEX: A Composition-Grounded Benchmark and Learning Framework for Explainable Aesthetic Image Cropping
链接: https://arxiv.org/abs/2608.07570
作者: Rui Yang,Wei Zhou,Dingyong Gou,Xiaohui Cui,Cong Li,Yinyin Gong,Yipo Huang,Jiliang Zhao
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Explainable aesthetic image cropping requires not only localizing a visually pleasing crop but also explaining why it is preferred. Existing crop-and-explain methods largely treat explanation as post-hoc text generation and overlook composition, a key aesthetic factor that links crop decisions with interpretable reasoning. In this paper, we reformulate explainable aesthetic image cropping as a structured crop-composition-explanation problem. To support this setting, we introduce COMEX, a new benchmark built through image expansion and an IO-reversal pipeline. COMEX contains 33,161 quadruples, each consisting of an expanded image, a crop box, a composition category, and a composition-grounded explanation, enabling joint learning of crop localization, composition understanding, and explanation generation. We further propose a two-stage SFT+GRPO framework, where supervised fine-tuning establishes the structured output protocol and basic cropping ability, and GRPO further improves crop quality, composition prediction, and explanation faithfulness. We benchmark 15 large vision-language models and existing cropping methods on COMEX, establishing a comprehensive testbed for composition-grounded explainable aesthetic cropping. Experiments on both COMEX and prior benchmarks demonstrate the effectiveness and transferability of our framework, with strong performance across evaluation metrics.
[CV-60] Latent-Frequency Validity: Fast Spectral Editing with Screened Video-VAE Transfer Operators
链接: https://arxiv.org/abs/2608.07569
作者: Bowen Xue,Jiafeng Xiong,Xin Quan
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
备注:
Abstract:Direct spectral editing in video-VAE latents can control noise, flicker, smoothness, and frequency content without a decode–filter–reencode pass. However, video VAEs may redistribute pixel-space frequency bands across latent channels, and latent edits can disrupt VAE round-trip dynamics. We introduce \emphlatent-frequency validity (LFV), which learns a compact VAE-specific spectral response and deploys it only when it improves decoded-target fidelity without worsening round-trip drift. LFV follows a validation-selected path from a diagonal per-frequency calibrator (C1) to full channel mixing (CM), making cross-channel capacity a controllable per-edit resource. Across 544 VAE–edit cells spanning six spectral families, LFV emits 423 cheap operators: 277 are handled by C1, while 146 (34.5% of emitted operators) require channel mixing. On the primary 120-cell radial sweep, 99/100 emitted operators pass source-video-grouped held-out evaluation. Across five additional filter families, all 323 emitted operators pass held-out evaluation. Fully frozen OpenVid-fitted operators, including the validation-selected path coefficient, pass all 20 tested CogVideoX and HunyuanVideo generated-domain cells without adaptation. The selected response matches direct latent-filter latency and is about 3\times faster than pixel filter–reencode. The resulting maps reveal distinct VAE regimes, including strongly channel-coupled CogVideoX responses and a sharp Open-Sora high-band stability frontier.
[CV-61] mporal Generalization in fNIRS-Based Autism Classification: A Cross-Time-Window Transfer Benchmark
链接: https://arxiv.org/abs/2608.07567
作者: Marios Petrov,Sahana Vinayak,Targol Bakhtiarvand,Moses Smith Guddah,Adham Atyabi,Frederick Shic,Kevin A. Pelphrey
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Functional near-infrared spectroscopy (fNIRS) is a promising modality for autism spectrum disorder (ASD) classification, yet existing approaches assume temporally aligned evaluation. In practice, the optimal observation window varies across subjects due to differences in hemodynamic delay and neurovascular coupling, creating a temporal distribution shift that degrades performance. We formalize this as a \textitcross-time-window transfer problem, introducing a protocol that varies window length (2.5–10,s) and offset within biological motion trials. Using topographic map representations of fNIRS recordings, we benchmark three vision architectures under two zero-shot baselines and eight adaptation strategies under leave-one-subject-out cross-validation ( N=124 ). Key findings: (1) zero-shot cross-window accuracy is near chance (54–69%); (2) \approx5% subject-specific fine-tuning recovers 90–96%, while a subject-specific upper bound reaches 97–100%, identifying inter-subject variability as the dominant barrier; (3) domain-adversarial and self-supervised strategies achieve 78–90% without target-subject data; and (4) discriminative information is recoverable from windows as short as 2.5,s. These findings provide a practical roadmap for deploying fNIRS-based ASD classifiers under realistic temporal variability.
[CV-62] What to Edit Next: Visually Aligned Image-Editing Follow-Up Suggestions in Conversational Systems
链接: https://arxiv.org/abs/2608.07565
作者: Zhijing Zhang,Jinpeng Yu,Xin Song,Bingnan Li,Chuyue Li,Changhui Du,Xiaolin Fang,Jiaming Liu,Ruihua Huang
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Conversational assistants increasingly recommend follow-up edits to help users continue a task. Existing systems primarily target text-only interactions, leaving image-creation conversations underexplored. In image-creation tasks, useful follow-up edit suggestions must reflect user preferences, offer diverse directions, and remain executable on the current image. We collected 100,000 real multi-turn image-creation conversation samples from Qwen App and found that 80.1% are image-dependent, underscoring the need for multimodal recommendation. We address this setting with a three-stage framework. In Stage 1, we use real online data to build a human-reviewed table of appropriate follow-up editing intents, then create SFT targets and fine-tune a multimodal policy. In Stage 2, to align rule-guided SFT suggestions with actual user choices, we use user click feedback to optimize the policy through multi-objective reinforcement learning. In Stage 3, to reduce visual inconsistencies between suggested edits and the current image, we introduce a visual verifier as additional training supervision. Extensive experiments demonstrate that our framework significantly outperforms baselines on both automatic and human evaluations. In a live user-randomized A/B test with millions of users, our final framework reduces visual inconsistency from 3.7% to 0.9%. Furthermore, it significantly improves recommendation CTR by 32.70%, image take-away rate by 16.32%, and average conversation turns per user by 39.90% (all p0.05).
[CV-63] Mechanistic Interpretability-Guided Selective Fine-Tuning of Vision-Language Models for Centimeter-Level Flood Depth Estimation ICML
链接: https://arxiv.org/abs/2608.07562
作者: Nafis Fuad,Xiaodong Qian,Dongxiao Zhu
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: This Paper is accepted in International Conference on Machine Learning and Application (ICMLA) 2026
Abstract:Urban flooding poses an escalating threat to transportation infrastructure, yet no operational system provides real-time, street-level flood-depth estimates at centimeter resolution. This paper presents three vision-language models fine-tuned for continuous flood-depth estimation from street-level imagery: FloodLlama-Dense, a fully fine-tuned QLoRA baseline, and FloodLlama-MI5 and FloodLlama-MI6, interpretability-guided sparse variants that fine-tune only the top five and six causally relevant cross-attention layers identified through mechanistic interpretability analysis, respectively. Training uses an approximately 610,000-image subset of a 2.81-million-image synthetic corpus generated in Unreal Engine 5. The dataset combines single-vehicle subsets with 5 cm depth increments and mixed-vehicle subsets with 1 cm depth increments, spanning seven vehicle types, four weather conditions, and flood depths from 0 to 40 cm. FloodLlama-Dense achieves an MAE of 0.40 cm, an RMSE of 1.97 cm, and an Acc@5cm of 97.59%. Mechanistic interpretability analysis combining linear probing, logit lens, centered kernel alignment (CKA), and cross-attention entropy reveals a two-stage adaptation pattern: layers L13-L22 restructure visual representations, while depth first becomes linearly decodable at layer L23. FloodLlama-MI5 and FloodLlama-MI6 leverage this insight by fine-tuning only five or six of the eight cross-attention layers, achieving an 86-88% reduction in trainable parameters (6.55-7.86 million versus 54.4 million) with minimal accuracy loss. On a real-world benchmark, FloodLlama-MI6 achieves 98.62% accuracy, compared with 86.61% for the published STURM-FloodDepth baseline.
[CV-64] XEns-CKD: An Explainable Ensemble-Based Approach for Chronic Kidney Disease Stage Detection
链接: https://arxiv.org/abs/2608.07561
作者: Rehan Ahmad,Gousia Habib,Muhammad Shaban,Ishfaq Ahmad Malik
类目: Computer Vision and Pattern Recognition (cs.CV)
备注:
Abstract:Chronic kidney disease (CKD) is a silent disease. Its progression may not significantly hamper a person’s daily routine. Human kidney function can be classified as normal or as one of the five stages of CKD. Early detection of the CKD stage can help patients understand the functional status of their kidneys and follow medical advice to slow CKD progression. In this paper, we propose XEns-CKD, a novel ensemble vision transformer-based scheme for CKD stage classification using ultrasound images. Three ViTs were trained on a private ultrasound image dataset using different training parameters. The performance of each ViT was evaluated using macro sensitivity, macro specificity, macro precision, macro F1-score, macro Youden index, the Matthews correlation coefficient (MCC), and macro balanced accuracy. The ensemble model achieved an overall classification accuracy of 86.36%. This work also emphasizes identifying and interpreting kidney regions affected by CKD progression. Explainable artificial intelligence techniques, including LIME, LRP, Attention-Min, and Attention-Max, were used to improve model transparency and clinical trust. An attention map combining the Attention-Min and Attention-Max results effectively identified and interpreted kidney regions affected during CKD progression from one stage to another. The attention map also highlighted the effects of CKD progression in these regions. Compared with existing methods, the proposed method classified the five CKD stages and normal kidney status with a 4% improvement in accuracy.
[CV-65] MVMD: A Multi-View Approach for Enhanced Mirror Detection WACV2025
链接: https://arxiv.org/abs/2608.07559
作者: Yidan Shen,Yu Wen,Chen Zhang,Xin Fu,Renjie Hu
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: This work has already published at WACV 2025, just want more accessibility
Abstract:In 3D reconstruction, mirrors introduce significant challenges by creating distorted and fragmented spaces, resulting in inaccurate and unreliable 3D models. As 3D reconstruction typically relies on multi-view images to capture different perspectives of a scene, detecting and labeling mirrors in multi-view images before reconstruction can effectively address this issue. However, existing methods focus solely on single-image detection, overlooking the rich information provided by multi-view setups. To overcome this limitation, we propose MVMD, a novel Multi-View Mirror Detection method, along with the first database specifically designed for mirror detection in multi-view scenes. The design of MVMD is grounded in the inherent associations between objects seen from different views and those reflected inside and outside of mirrors. These relationships are learned through cross- and self-attention mechanisms. MVMD consists of three key blocks: the Inter-Views Block tracks the shifts of objects within mirrors caused by changes in viewpoint; the Intra-View Block detects object reflections inside mirrors; and the Refinement Block sharpens mirror boundaries and enhances detected details. Experimental results show that our method improves accuracy by up to 2.6% and IoU by up to 11.1%, compared to single-image mirror detection techniques. This substantial improvement makes MVMD particularly effective for computer vision tasks, especially in enhancing the accuracy of 3D reconstruction in mirror-dense environments. Comments: This work has already published at WACV 2025, just want more accessibility Subjects: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG) Cite as: arXiv:2608.07559 [cs.CV] (or arXiv:2608.07559v1 [cs.CV] for this version) https://doi.org/10.48550/arXiv.2608.07559 Focus to learn more arXiv-issued DOI via DataCite Journalreference: 2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Related DOI: https://doi.org/10.1109/WACV61041.2025.00904 Focus to learn more DOI(s) linking to related resources
[CV-66] Learning Physical Interaction: A Survey of Tactile- and Force-aware Robot Learning
链接: https://arxiv.org/abs/2608.07558
作者: Shilin Shan,Chuhao Zhou,Ruize Wang,Xinyan Chen,Xiangyu Chen,Xinyu Zhou,Boyu Ma,Iris Yuxuan Hu,Jingliang Li,Celeste Yuxuan Hu,Geng Li,Guohao Chen,Tianrui Zhu,Zhe Li,Yanjie Ze,Haoran Geng,Zhiyang Dou,Jianxin Bi,Yuejiang Liu,Jianshu Zhou,Jiachen Li,Paul Liang,Tatsuya Harada,Robert Katzschmann,Harold Soh,Na Li,Edward Johns,Danica Kragic,Jan Peters,Wojciech Matusik,Masayoshi Tomizuka,Jitendra Malik,Jianfei Yang
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: 53 pages, 7 figures
Abstract:Physically grounded robot intelligence requires robots to perceive, reason about, and regulate their interactions with the physical world. This capability is particularly critical in contact-sensitive manipulation, where successful task execution depends not only on visual perception and motion generation, but also on force regulation and adaptive control. In this context, recent robot learning methods have made substantial progress by integrating force, tactile, vision, language, and proprioceptive sensing into learned manipulation policies. In parallel, many systems adopt multi-phase architectures that combine high-level policies, action-refinement modules, and low-level controllers to bridge semantic task understanding with reactive physical execution. Despite these advances, existing surveys have not explicitly reviewed force- and tactile-aware robot learning from a unified perspective that jointly captures multimodal sensing and multi-phase system design. This survey addresses this gap by proposing TF-ART, a Tactile/Force-Aware Robot learning Taxonomy for multimodal and multi-phase frameworks, which maps individual methods into a unified hierarchical structure. The framework characterizes how recent works organize observation modalities, encode and fuse heterogeneous sensory inputs, generate and refine actions across multiple phases, and connect learned policies to reactive robot-end control. Building on this methodological view, we further examine the task settings and infrastructure requirements of physical interaction, thereby integrating both algorithmic and practical perspectives on force- and tactile-aware robot learning.
[CV-67] AeroDPO: Unleashing Lightweight UAV Navigation with High-Fidelity Perception and Automated Preference Optimization
链接: https://arxiv.org/abs/2608.07557
作者: Peng Xu,Chengcheng Wang,Shaohua Wan
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV)
备注: 7 pages, 3 figures, 4 tables
Abstract:Vision-Language Navigation for Unmanned Aerial Vehicles (UAV-VLN) requires rapid and reactive control in complex 3D environments. Recent minimalist end-to-end paradigms show great promise but typically rely on massive language models containing billions of parameters, incurring prohibitive latency for real-world edge deployment. In this paper, we challenge this parameter-heavy reliance. Comprehensive cross-scale evaluations reveal the critical insight that perception quality fundamentally outweighs language reasoning capacity. We demonstrate that a lightweight 2B model equipped with high-fidelity visual inputs completely matches the overall success rates of massive 7B baselines. However, this minimalist policy exposes a fundamental robustness flaw inherent to pure Behavior Cloning (BC). Lacking explicit negative feedback, the agent fails to internalize robust spatial constraints and exhibits alarming collision rates in out-of-distribution (OOD) scenarios. To overcome this vulnerability without relying on unscalable human annotations, we propose AeroDPO, a zero-cost automated Direct Preference Optimization pipeline driven by deterministic physical simulation state rollback. Upon detecting collisions, the system autonomously rewinds the environment to extract causal reasoning errors as rejected actions, applies decoupled privileged interventions to synthesize collision-avoidance preferred maneuvers, and leverages an offline vision language inspector to filter visual ambiguities. By equipping our 2B model with this automated data flywheel, AeroDPO boosts success rates to 49.16% on unmapped scenarios while drastically suppressing collision rates, establishing a new SOTA for autonomous aerial agents.
[CV-68] Impact of Dataset Composition on Embedded Real-Time UAV Wildfire Detection Using Compact YOLO Models
链接: https://arxiv.org/abs/2608.07554
作者: Eduardo de los Santos,Andre S. Kelbouscas,Ricardo B. Grando,Bruna V. Guterres
类目: Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
备注: Paper accepted at the ICCAS 2026
Abstract:The development of vision-based wildfire detection systems for unmanned aerial vehicles is constrained by the limited availability of diverse real-world training images. This paper investigates the impact of dataset composition on embedded real-time UAV wildfire detection using compact YOLO models as a controlled validation family. Four training configurations were evaluated: real non-augmented, real augmented, hybrid non-augmented, and hybrid augmented, where the hybrid sets combine real wildfire images with AI-generated samples. The objective is to determine whether synthetic data mixing and image augmentation improve practical detection performance under resource-constrained deployment conditions. Experimental results show that the best overall operating point was obtained with the real non-augmented dataset, which achieved the strongest balance between recall and mean average precision for UAV-based wildfire detection. The results also show that neither hybridization with synthetic data nor augmentation produced a better final deployment choice. These findings suggest that, for embedded UAV wildfire detection, dataset realism and domain alignment are more valuable than increasing training set size through synthetic expansion.
[CV-69] Auditing Medical Vision-Language Models on Chest Radiographs: Estimating Reference Agreement Across Institutions
链接: https://arxiv.org/abs/2608.07550
作者: Pengyang Yu,Yiou Wang,Zhongping Dong,Sahraoui Dhelim,Chun-Mei Feng,M. Tahar Kechadi
类目: Computer Vision and Pattern Recognition (cs.CV); Machine Learning (cs.LG)
备注: 10 pages, 3 figures, 3 tables
Abstract:Vision-language models return structured chest-radiograph findings through interfaces exposing no confidence score, so a receiving institution cannot read off how far to trust an individual judgment. Whether agreement with an institution’s reference standard transfers across sites, findings, prediction directions and question formats is largely unmeasured. We evaluated three generative vision-language models on three institutional chest-radiograph corpora and six findings under two elicitation protocols, comprising more than 345,000 finding-level predictions, and estimated finding-by-direction reference agreement at a receiving institution from a small budget of local labels. Estimation strategies were then stress-tested under repeated strict institution-held-out evaluation. Under evaluation excluding the receiving institution from development entirely, adaptive selection among the seven estimators that design admits did not improve on simple fixed alternatives: it achieved a mean Brier score of 0.1083, against 0.0853 for always using a Beta-Binomial empirical-Bayes estimator and 0.0855 for a target-only logistic model. Those two differ by 0.0003, less than this family’s own sensitivity to a change of solver version, and each leads in about half the settings, so no default can be recommended. Their advantage over estimators pooling across institutions was concentrated at one site and not confirmatory once clustered by institution, and a plug-in empirical-Bayes posterior-predictive count interval at a nominal 95% level covered 87.0%, less at the hardest institution. Reference agreement therefore has to be re-evaluated per site and per interface; these results concern agreement with institutional labels, not clinical correctness.
[CV-70] P2Voxel: Pyramid Pivot Voxelization for 3D Mesh Tokenization
链接: https://arxiv.org/abs/2608.07549
作者: Zhenhong Sun,Haozhe Liu,Yifu Wang,Xibin Song,Senbo Wang,Huadong Mo,Daoyi Dong,Hongdong Li,Pan Ji
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Triangle meshes provide explicit and accurate surface geometry, yet their irregular topology connectivity makes 3D mesh tokenization a geometric sampling problem: how to sample and organize geometric evidence into compact, structured and learnable tokens. Beyond field-centric volumetric sampling and edge-intersection surface sampling, we retarget mesh tokenization as \textitlocal surface evidence sampling: identifying the minimal geometric evidence inside each active voxel that is sufficient for deterministic surface recovery. To this end, we introduce \textbfP2Voxel, a pyramid pivot voxelization framework for compact and reconstruction-aware mesh tokenization. P2Voxel is built on three key innovations. Under the \textitLocal Planarity assumption, Pivot Voxelization represents each active voxel with a surface pivot and an orientation sign, providing minimal local evidence that can induce the corner values required for deterministic reconstruction. Under the \textitSpatial Complexity assumption, Pyramid Pivot Voxelization exploits the spatial non-uniformity of real surfaces by allocating finer pivot tokens to geometrically complex regions while keeping smooth regions coarse and compact. Under the \textitBlock Reconstructability assumption, a Pyramid VAE learns compact multi-resolution latent codes over locally reconstructable pivot blocks, avoiding the need to model the entire high-resolution voxelized shape as a dense global field. Together, these designs convert meshes into compact, structured, and learnable pyramid pivot tokens, enabling efficient mesh reconstruction for downstream 3D tasks.
[CV-71] SC2-WM: A Self-Correcting World Model with Closed-Loop Feedback for Vision-and-Language Navigation in Continuous Environments ICML2026
链接: https://arxiv.org/abs/2608.07548
作者: Xuan Yao,Yuze Zhu,Junyu Gao,Zongmeng Wang,Changsheng Xu
类目: Robotics (cs.RO); Computer Vision and Pattern Recognition (cs.CV)
备注: Accepted by ICML 2026
Abstract:Vision-and-Language Navigation in Continuous Environments (VLN-CE) requires agents to make fine-grained navigation decisions under partial observability. However, most existing methods rely on open-loop execution, lacking mechanisms to detect and correct internal state drift during inference. We propose SC ^2 -WM, a self-correcting world model framework that introduces internal feedback for closed-loop decision making in VLN-CE. Our method derives feedback from world-model foresight to perform state-level plan refinement before action execution. To handle challenging scenarios, we further introduce conditional world-aware adaptation, which enables model-level correction by selectively updating the world model at test time when feedback indicates model capacity insufficiency. Experiments on standard VLN-CE benchmarks demonstrate improved navigation robustness and generalization. Our code is available at this https URL.
[CV-72] Learning an Interior Layout Policy in a Domain Specific Language Action Space
链接: https://arxiv.org/abs/2608.07547
作者: Yuhao Lu,Weichen Zhang,Wenyi Xiao,Haohui Chen,Yiyun Fei
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注:
Abstract:Indoor scene layout generation is a challenging task in interior design. Existing methods often oversimplify the task by reducing room conditions to coarse 3D bounding boxes and neglecting structural elements such as doors and windows. More fundamentally, many prior approaches formulate spatial reasoning as direct coordinate prediction, thereby casting interior layout design as continuous regression over raw geometric parameters, which hinders the model from learning the underlying reasoning logic of intelligent layout design. We propose \textbfLayoutDSL, a novel LLM-based framework for learning an interior layout policy in a domain-specific language (DSL) action space. The DSL provides an explicit symbolic representation of layout information and serves as a structured action space for layout reasoning, where each action corresponds to an interpretable design decision. Under this DSL-based policy learning paradigm, we construct 3D-FrontDSL, a dataset of room-structure annotations paired with synthetic DSL action sequences for supervised fine-tuning. To promote a more generalizable and scalable policy with verifiable feedback, we design rewards grounded in interior design principles and physical plausibility, and optimize the policy via reinforcement learning. Extensive experiments demonstrate that LayoutDSL substantially improves spatial plausibility and design logicality over strong baselines and existing methods.
[CV-73] Performance of large language models in the optical diagnosis of colorectal polyps
链接: https://arxiv.org/abs/2608.07543
作者: Joshua C. Vences,William T. Tran,Nikko Gimpaya,Catharine M. Walsh,Rishad J. Khan,Robert Bechara,Asher C. Wiggins,Celine N. Rousan,Kaitlyn V.G.L. Morgado,Angie Ibrahim,Kevin H. M. Kuo,Daniel von Renteln,Alexander Hann,Dennis L. Shung,Michael A. Scaffidi,Charles Ménard,Joshua Landy,Samir C. Grover
类目: Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
备注: 22 pages, 1 figure, 5 tables
Abstract:Background and Study Aims: Accurate optical diagnosis of colorectal polyps guides resection strategy and surveillance, with multimodal large language models (MLLMs) showing potential for image-based diagnosis. We aimed to evaluate the diagnostic accuracy of MLLMs in classifying colorectal polyps and predicting histology. Methods: We conducted a retrospective diagnostic performance study using the PRIME dataset, a curated set of white light and narrow-band imaging (NBI) images. We evaluated Claude Opus 4, Google Gemini 2.5 Pro, GPT-o3, GPT-4o, and GPT-5. For Paris, Narrow-band Imaging Colorectal Endoscopic (NICE), and predicted histology, we calculated F1 scores, percent correct scores, and accuracy of each MLLM compared to expert responses for 132 cases. Cochran’s Q and McNemar’s Test were used to determine differences between predicted values of each MLLM. Results: The F1 scores among MLLMs were 0.9 for all models for neoplastic vs. non-neoplastic polyps. Gemini 2.5 Pro demonstrated the highest F1 scores for invasive vs. non-invasive polyps and low- vs. high-grade adenoma, at 0.560 and 0.492 respectively. Claude Opus 4 and GPT-5 had statistically significantly higher percent correct scores than other MLLMs at 41.7%, using Paris classification. Conclusions: Claude Opus 4 and Gemini 2.5 Pro showed the highest accuracy in differentiating polyp subtypes, performing closest to expert consensus. Sensitivity and specificity, however, did not meet ESGE standards, highlighting the need for prospective multicenter trials and the design of human-in-the-loop workflows before clinical deployment.
人工智能
[AI-0] Capability Is Not Propensity: Measuring Pressure-Robust Cooperative Behavior in Civic LLM Agents ICML
链接: https://arxiv.org/abs/2608.09485
作者: Neel Tushar Shah,Manglam Kartik,Akshat Karkar
类目: Artificial Intelligence (cs.AI)
备注: Accepted at ICML AI4GOOD Workshop 2026
Abstract:Cooperative capabilities in language models are dual-use. The same social reasoning that supports civic deliberation can also enable strategic omission, false consensus, and manipulative framing. We argue that Cooperative AI evaluations should separate what models can do under benign instructions from what they tend to do under realistic civic pressure. We introduce DiffCoop-Civic, a 10-scenario pilot evaluation suite spanning preference understanding, evidence and persuasion, commitment design, asymmetric information, and dissent preservation. Across seven models from four model families, subtle omission pressure produces a near-uniform shift: manipulative enablement rises by 1.17 points and dissent preservation falls by 1.67 points on a 5-point scale. Overt false-consensus pressure behaves differently: it triggers refusal or redirection in some aligned API models, but direct compliance in several open-weight models. A lightweight Pareto-Trace prompting intervention improves pressure robustness without simply relying on hard refusal. An anonymous reproducibility package is available at this https URL.
[AI-1] From Prompt to Harness: Coderlet from Scratch
链接: https://arxiv.org/abs/2608.09480
作者: Mengfan Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:A model alone does not determine how a programming agent acts. What the model sees, how actions enter the environment, how feedback returns, and how one run affects the next all depend on how the harness is organized. Minimal examples usually show only the basic interaction between a model and tools, while production systems spread these relationships across complex components and dependencies. This paper studies a compact harness design by following a single request through context formation, model decision, environmental action, observation return, and state continuation. Three boundaries—model, execution, and state—connect the model service, tool environment, and persistent state, while the request lifecycle determines the order in which these transitions occur. Together, they show the harness’s core role: turning model generations into environmental actions, carrying runtime feedback into later decisions, and allowing state to continue across requests. On top of this runtime structure, a harness can also be gradually refined across runs through continued bootstrapping. The design is realized in the executable artifact this https URL.
[AI-2] ActBench: Self-Evolving Benchmark of Behavioral Safety in Cowork Agents
链接: https://arxiv.org/abs/2608.09476
作者: Hongwei Yao,Yiming Liu,Meihui Chen,Jieling Chen,Zikun Chen,Yiling He,Wangze Ni,Cong Wang,Kui Ren
类目: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI)
备注: Benchmark and Code is available this https URL
Abstract:Cowork agents may complete benign tasks while disclosing protected data, manipulating unauthorized state, invocate unauthorized API. We define behavioral safety and introduce ActBench, a self-evolving benchmark that evaluates such behavior risk from execution trajectories rather than final responses. Each case pairs a benign task with an adversarial variant that preserves its instruction, configuration, initial state, rating model, and trusted records while injecting a task-reachable payload. ActBench contains 600 cases from 213 scenarios, spanning 15 risk behaviors, six execution spaces, and 48 web-service this http URL move beyond static payloads, we propose a reward-guided beam search method that jointly optimizes attack effectiveness and task utility, while reflection diagnoses failed execution checkpoint and guides payload revision. Besides, we propose a dual evidence verification mechanism that verifies agent execution safety and utility through log evidence and LLM-based trajectory this http URL evaluate 15 LLMs and 6 open-source cowork agents over 24,000 trajectories. Under a fixed harness, attack success rates ranges from 10.1% to 94.4% across models, while under a fixed base model, they range from 73.7% to 94.4% across this http URL results show greater variation across models than agent harness, while attacks remain highly successful across all tested this http URL benchmark is released at: this https URL.
[AI-3] MixFormer: Linear Transformer with Mixture of Memory Experts
链接: https://arxiv.org/abs/2608.09468
作者: Yu Guo,Lei Duan
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 17 pages, 9 figures, and 4 tables
Abstract:State Space Models (SSMs), as a mainstream research direction of linear Transformers, aim to achieve higher efficiency than standard Transformers in long-context modeling. However, existing SSMs suffer from limited input adaptivity and constrained memory capacity, leading to information loss when modeling ultra-long sequences. To address these limitations, we propose MixFormer, a novel linear Transformer that integrates a Mixture-of-Memory-Experts (MoE) mechanism. Specifically, the model maintains differentiated memory states through multiple collaborating memory experts and employs a novel Time-Aware Linear Attention (TALA) mechanism, which leverages learnable exponential decay functions and positional biases to dynamically update memory. This design enables the model to selectively reinforce important historical information while effectively mitigating memory dilution, substantially improving long-range dependency modeling. Experiments on long-sequence text and image generation tasks demonstrate that MixFormer not only achieves significant performance gains but also provides a more sustainable computational backbone for the next generation of web infrastructure.
[AI-4] Learning to Modulate Not to Cycle: Soft Actor—Critic Recovers Inverter-Style Heat-Pump Control
链接: https://arxiv.org/abs/2608.09453
作者: Faizan Ahmed,Aniket Dixit,James Brusey
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 9 pages, 4 figures, Accepted at UKCI 2026, Will be published in Springer
Abstract:On–off cycling is the main cause of compressor wear in residential heat pumps, yet reinforcement learning (RL) controllers for buildings typically optimise only energy cost and thermal comfort, ignoring how much the learned policy cycles. We add a levelised compressor-wear term to the control reward and study how the resulting behaviour depends on the RL algorithm. Training Soft Actor—Critic (SAC) and Proximal Policy Optimisation (PPO) on an identical Markov decision process for the BOPTEST bestest hydronic heat pump case, we find that SAC learns a continuous modulation policy that keeps the compressor permanently engaged—the operating principle of an inverter-driven heat pump—achieving zero start-ups per day, whereas PPO collapses to bang-bang control that cycles more than the baseline. On the BOPTEST emulator the SAC policy cuts thermal discomfort by up to 90.7% for an 11.5% cost increase, while eliminating all baseline cycling.
[AI-5] WDL-OPD: Weak-Driven On-Policy Distillation via Mixture-Constrained Co-Training
链接: https://arxiv.org/abs/2608.09447
作者: Zehao Chen,Gongxun Li,Tianxiang Ai,Yifei Li,Zixuan Huang,Wang Zhou,Tao Huang,Fuzhen Zhuang,Xianglong Liu,Jianxin Li,Deqing Wang,Yikun Ban
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:On-policy distillation (OPD) aligns a student with a teacher on trajectories sampled from the student itself, reducing the train-test state mismatch of offline distillation. The same feedback loop can nevertheless be unstable: each update changes both the policy and the states on which the next update is computed. We introduce WDL-OPD, a mixture-constrained co-training method with two trainable policies. An anchor policy generates every rollout, an auxiliary policy evaluates the same visited states, and a geometric mixture of their token distributions is matched to a frozen teacher by reverse KL. Both policies receive gradient. We show that freezing the auxiliary recovers an anchor-plus-contrast proxy target closely related to OPD ^2 and W2S-OPD, whereas joint training creates branch-level degrees of freedom that a static delta cannot express. In recorded Qwen3 experiments at 1.7B and 4B scale, WDL-OPD produces the strongest student checkpoint in each of four scale-domain settings. It raises MATH500 accuracy from 0.630 to 0.685 at 4B and from 0.521 to 0.585 at 1.7B. In code generation, seven single-policy OPD configurations exhibit entropy growth or trajectory degradation, while co-training reaches independently re-evaluated development scores of 0.637 and 0.375. Because several comparisons differ in curriculum or initialization, these results support a stabilization hypothesis rather than a universal causal claim. We provide the exact training algorithm, failure evidence, and the controlled comparison matrix needed to test that hypothesis.
[AI-6] Coupled Graph–Policy Distillation for Personalized Medication Safety in Older Adults with Multimorbidity
链接: https://arxiv.org/abs/2608.09443
作者: Zihan Wang,Anglin Liu,Rongyi Wang,Dantong Li,Yi Lu,Siqing Yuan,Hongxia Xu,Zhongtian Long,Jintai Chen
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language model (LLM) agents can support medication review between clinical visits, but safe choices for older adults with multimorbidity depend on conditions, medications, and geriatric risks that users may omit. We introduce ATLAS, a coupled graph–policy distillation framework for patient-adaptive medication safety. ATLAS structures guideline evidence as a medication-safety graph. Targeted questions update the patient state and distill relevant relations into a patient-specific medication conflict graph (PMCG). A risk-first multi-agent policy uses the PMCG to screen contraindications, assess cautions and monitoring needs, identify safer alternatives, and verify the final medication plan. We also introduce GeriMedBench, an interactive benchmark that tests safety-critical information acquisition and evidence-based decision revision. Across a European non-interactive multimorbidity benchmark, an Asian interactive multimorbidity benchmark, and an Asian non-interactive cross-guideline benchmark, ATLAS achieves the strongest complete-decision performance among the compared systems. On the European non-interactive multimorbidity benchmark, it exceeds the strongest proprietary LLM baseline by 53.73 points in Strict Success Rate and 14.63 points in overall safety reasoning score (OSRS), with no unsafe recommendations under the automated evaluator. A blinded clinician evaluation gives ATLAS higher mean ratings across all five criteria and flags potentially unsafe recommendations in one ATLAS case and two Gemini cases.
[AI-7] Listen See and Track: Spatio-Temporal Audio-Visual Sound Event Reasoning for Omni-Modal Language Models
链接: https://arxiv.org/abs/2608.09435
作者: Zhi Zeng,Cheng Zhang,Zesheng Yang,Rendong Pi,Jiaying Wu,Di Zhang,Zihan Ma,Guodong Li,Zhou Yang,Yu Xiang,Yifei Zheng,Minnan Luo
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Understanding dynamic sound sources requires jointly determining what produces a sound, where the source is located, and how it moves over time. Yet existing audio-language models often represent clips as global acoustic events, while vision-language models lack the spatial audio cues needed to localize and track individual sources. To evaluate this missing capability, we introduce ST-OmniQA, a spatio-temporal audio-visual question-answering benchmark built from panoramic videos paired with synchronized first-order Ambisonics (FOA) audio of moving sound sources. It contains 40K videos and 400K question-answer pairs organized into four capability levels covering sound-event recognition, direction of arrival, source distance, motion trajectories, and temporally grounded audio-visual reasoning. Building on this benchmark, we propose ST-Omni-R1, which integrates FOA-derived semantic and trajectory representations with panoramic visual context and is trained through progressive curriculum learning and reasoning-tree reinforcement learning. ST-Omni-R1 achieves 77.83% average semantic accuracy across the four levels, compared with 37.28% for the best evaluated baseline. Results on three public spatial-audio benchmarks further indicate that its learned spatial and motion representations transfer beyond ST-OmniQA.
[AI-8] How Simple Can It Get? From Interpretable Equations to Readable Rules for Financial Decision Making
链接: https://arxiv.org/abs/2608.09433
作者: Adia Lumadjeng,Ilker Birbil,Erman Acar
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注: 8 pages, 3 figures. Submitted to ICAIF 2026
Abstract:In regulated domains such as finance, a model that cannot be explained cannot be deployed, yet many interpretable classifiers defeat their own purpose by producing formulas with dozens of features that no regulator could read. We take the reverse direction. Starting from an interpretable classifier expressed as a single equation over the input features, we progressively simplify it into more readable forms, including a pruned monomial, a directional if–then rule, and the integer scorecards and tallies that finance already deploys. Because the equation is itself the predictive model rather than a post-hoc explanation we can directly quantify what is lost under each simplification. Across four financial datasets, we find that pruning is nearly free and that fidelity can erode faster than predictive performance, allowing simpler rules to remain effective classifiers without faithfully reproducing the original model. A human assessment shows that simplification improves perceived readability, while preferences for different representations vary by professional background. Beyond measuring these losses empirically, we show that some can be anticipated from the original model: we derive a bound on the change caused by pruning and predict how faithfully a rule retaining only the direction of each feature’s effect preserves the original ranking.
[AI-9] KVDiagnosis: A Diagnostic Benchmark for KV-Cache Compression in Long-Context Language Models
链接: https://arxiv.org/abs/2608.09412
作者: Chen Qiu,Ziwu Liu,Chao Fei,Guozhong Li,Panos Kalnis
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:KV-cache compression reduces long-context memory, but aggregate task scores reveal neither which correct executions fail nor why. We present KVDiagnosis, a diagnostic dataset and benchmark with three contributions. First, a 25-method taxonomy groups methods into five mechanism families and links them to eight verified implementations and their valid diagnostic measurements. Second, for every supported method setting, we evaluate all sources in each fixed split against a per-source FullCache control before selecting FullCache-correct/compressed-wrong (C-to-W) rows separately for each method-setting, so no compressor defines another’s test set. Third, a common record format links paired outputs and run metadata to cache, likelihood, attention, and decoding measurements with explicit applicability states. On Qwen3-8B, four evidence-aware workloads yield 59 800 supported compressed runs over 2600 sources and 12 520 C-to-W rows. Under fixed diagnostic rules, 63.2% have low or partial measured/projected coverage. Only 19 rows (0.2%) combine high measured/projected coverage with strong likelihood drift; another 2,126 (17.0%) preserve structural position addressability, for which representation fidelity remains unknown, while showing the same drift. Against C-to-C success controls, all ten diagnostics separate failed from successful compression (stratified AUROC 0.684-0.871). Among 96 reproducible low-EAR failures, a controlled 4x evidence-attention boost repairs 29.2%, versus 6.3% under a count-matched sham intervention and 3.3% degradation on matched C-to-C controls. Code and data are available at this https URL.
[AI-10] Monotonicity-Guided Bottom-Up Petri Net Discovery: The SPECpp Framework
链接: https://arxiv.org/abs/2608.09398
作者: Leah Tacke genannt Unterberg,Lisa L. Mannel,Wil M. P. van der Aalst
类目: Databases (cs.DB); Artificial Intelligence (cs.AI)
备注: 59 pages, 15 figures
Abstract:Process discovery is one of the central challenges in process mining. Petri nets are particularly attractive because simple local constructs can express complex behavior, including concurrency. While their global behavior may be difficult to analyze, individual places can be efficiently characterized using monotonic properties, enabling bottom-up discovery. Unlike top-down approaches such as the Inductive Miner, which rely on predefined constructs for sequences, choices, loops, and concurrency, our approach allows such structures to emerge organically and can exploit the full expressive power of Petri nets, including free-choice constructs and long-term dependencies. The main challenge is the exponential number of candidate places and their combinations. We present the SPECpp framework which implements strategies to obtain high-quality models under time and resource constraints. SPECpp supports rapid experimentation and is used to evaluate these strategies using both synthetic and real-life event data.
[AI-11] OpenLoopEvolve: A Verifiable Self-Evolution Framework for Loop Policies in Long-Horizon Complex Tasks
链接: https://arxiv.org/abs/2608.09380
作者: Siqi Wang,Xinlin Li,Zhenglin Li,Li Li
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Long-horizon complex tasks require agents to repeatedly observe states, formulate plans, invoke tools, verify results, and recover from failures in continuously changing environments. However, such control experience often remains confined to a single context or a fixed prompt, and is difficult to accumulate and reuse across historical traces. This paper presents OpenLoopEvolve (OLE), a self-evolution framework centered on the Loop Policy. OLE represents an agent’s observation, planning, memory, action, verification, recovery, stopping, and budget control as portable policy assets with versions and lineages, and provides online and offline evolution modes that can be selected according to practical needs: the online mode triggers candidate generation based on feedback from continuous operation, whereas the offline mode searches for candidate policies from archived traces and failure evidence. Both modes share an evolution mechanism consisting of autonomous proposals by a large language model, Champion–Challenger paired evaluation, and robust release. Policies released online are activated at a subsequent task boundary, monitored using subsequent feedback, and rolled back to their parent versions when degradation conditions are met. On the simulated business benchmark YC-Bench, both modes improve aggregate task performance, task success rate, and risk metrics relative to a fixed initial Loop Policy. The results indicate that treating the Loop Policy as a governable asset can support the accumulation, comparison, release, and reuse of control experience and improve agent performance on long-horizon complex tasks.
[AI-12] st-Time Augmentation for LLM s: When Input Diversity Beats Output Diversity at Matched Compute
链接: https://arxiv.org/abs/2608.09351
作者: Nikita Kozodoi,Zainab Afolabi,Jack Butler
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Machine Learning (stat.ML)
备注: Published at the COLM 2026 Workshop on Efficient Reasoning
Abstract:Test-time scaling improves LLM accuracy but multiplies inference cost, making the accuracy gained per unit of compute the metric that matters in deployment. Self-consistency is one of the established approaches, which spends this budget entirely on the output side by sampling repeated reasoning paths. We study Test-Time Augmentation (TTA), which extends self-consistency by also perturbing the input, aggregating predictions across transformed versions of the input, and ask whether input-side diversity converts compute into accuracy more efficiently than output-side diversity. We perform a systematic, matched-compute comparison: we evaluate three simple input-side strategies (semantic rephrasing, lexical perturbations, and visual transformations) across six datasets covering general and multilingual knowledge, mathematical reasoning, multi-modal question answering, and sentiment classification, against chain-of-thought prompting and self-consistency. Semantic rephrasing delivers consistent and statistically significant accuracy gains while Pareto-dominating self-consistency on cost-effectiveness, delivering roughly 1.8X more accuracy per dollar and outperforming it on five of six tasks. We further analyze the number of augmentations, multi-modal strategies, and base model scaling, finding that TTA is most cost-effective for mid-tier models where a stronger model is unavailable or too expensive. Our findings indicate that for current mid-tier LLMs, varying the input converts inference compute into accuracy more efficiently than varying the reasoning path alone. The TTA implementation is available at this https URL.
[AI-13] LLM -Guided Heuristic Design from Simulation Traces: A Case Study in Dynamic Production and AGV Scheduling
链接: https://arxiv.org/abs/2608.09343
作者: Jinbo Li,Chuanhao Li
类目: Artificial Intelligence (cs.AI)
备注: 33 pages, 8 figures
Abstract:Simulation-based optimization (SBO) evaluates executable policies under stochastic dynamics, but most methods treat the simulator as a black box: aggregate scores rank candidates without revealing why they fail or which policy logic should change. We present an LLM-guided heuristic design framework that uses repeated simulation for selection and event-level traces for diagnosis. Each incumbent is assessed through multiple replications, while replaying its lowest-scoring one produces a queryable trace. A manager agent formulates bottleneck hypotheses from this evidence, and editing agents implement parallel code-level revisions. After execution checks and repeated evaluation, best-so-far selection retains only improvements. LLM revision occurs between evaluation batches, while a fixed policy controls each simulation run. We evaluate the framework in a discrete-event simulation of dynamic production and automated guided vehicle (AGV) scheduling. Across five independent optimization runs with Gemini-3.1-Pro, final mean scores averaged 77.51 on the simulator’s 0-100 scale. In the highest-scoring run, trace-based diagnoses motivated proactive charging, distance-aware AGV assignment, and rebalanced dispatch priorities, raising the best-so-far mean score from 62.49 to 78.61. On 100 matched seeds, the best final policy outscored representative rolling-MILP, rule-based, and metaheuristic policies on every seed and retained its advantage under random faults without re-optimization. After separate re-optimization for a longer horizon and variable order interarrival times, the resulting policies again outscored all baselines. Ablations with two LLM backbones showed that removing either parallel candidate generation or trace-database access reduced final mean scores. These results show that simulation traces can guide targeted code-level policy improvement in complex simulation-based scheduling. Comments: 33 pages, 8 figures Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2608.09343 [cs.AI] (or arXiv:2608.09343v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2608.09343 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-14] Control-Oriented Scenario Tree Construction through Reinforcement Learning
链接: https://arxiv.org/abs/2608.09335
作者: Fabio Pavirani,Bert Claessens,Pierre Pinson,Chris Develder
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Systems and Control (eess.SY)
备注:
Abstract:Multistage stochastic model predictive control (MPC) handles uncertainty by optimizing over a scenario tree, a finite branching approximation of future outcomes constructed from sampled forecasts. To build such a tree, conventional methods focus on matching the underlying probability distribution—e.g., via Wasserstein-based scenario reduction—but improved distributional accuracy does not necessarily yield better control performance. We propose a control-oriented approach that learns scenario tree construction directly from its impact on downstream decisions. Fixing the tree topology, we formulate tree construction as a sequential assignment of sampled scenarios to leaves. This assignment is parameterized by an attention-based policy over the scenario set and trained using reinforcement learning, with closed-loop control profit as the objective. Training is stabilized by an asymmetric critic that leverages realized future trajectories. We evaluate the method on a risk-averse battery arbitrage problem. Across a range of forecast set sizes, the learned construction consistently achieves the highest profit, outperforming classical forward and backward reduction methods and certainty-equivalent (single-trajectory forecast) control. The learned policy also exhibits greater robustness on challenging instances, consistently demonstrating better tail-risk characteristics. Analysis of the resulting trees indicates that our method constructs compact, selectively branching structures that capture high-impact events while keeping most trajectories nearly deterministic. These findings highlight that the value of a scenario tree depends critically on the decisions it supports, and provide an effective framework to train scenario tree constructors merely based on the closed-loop control optimization signal.
[AI-15] Who Bridges Safety? Identifying and Targeting Cross-Lingual Shared Safety Pathways
链接: https://arxiv.org/abs/2608.09095
作者: Shuyi Miao,Wangjie Qiu,Pengyang Shao,Canran Xiao,Fei Shen,Zhiming Zheng,Tat-Seng Chua
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Uncovering the internal mechanisms underlying the safety capabilities of large language models (LLMs) is crucial for developing trustworthy artificial intelligence. Currently, mechanistic interpretability studies on multilingual safety are largely confined to local components, such as isolated neurons. However, this static and fragmented perspective overlooks the synergy among components and fails to elucidate how safety signals dynamically propagate within the model to drive safety decisions ultimately. In this work, we move beyond isolated neurons to identify and target the cross-layer functional pathways formed during safety signal propagation, thereby uncovering the mechanisms driving the cross-lingual safety gap. Specifically, we first identify monolingual safety pathways and validate their impact on refusing harmful requests. Subsequent cross-lingual analyses reveal a sparse subset of cross-lingual shared safety pathways, confirming that this intersection acts as the internal bridge transferring safety capabilities from high-resource (HR) languages to non-high-resource (NHR) languages. Building on these mechanistic findings, we propose a pathways-targeted alignment method based on the cross-lingual shared safety pathways. Experimental results show that updating only a small fraction of pathway parameters significantly improves safety in NHR languages while largely preserving the model’s general capabilities.
[AI-16] A Multi-Scale Temporal Framework with Dynamic Fusion for EEG-Based Emotion Recognition
链接: https://arxiv.org/abs/2608.09088
作者: Stefanos Gkikas,Yang Guo,Guangliang Li,Raul Fernandez Rojas,Giorgos Giannakakis,Randy Gomez
类目: Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Sound (cs.SD)
备注: The paper has been accepted at: IEEE | 2026 9th International Conference on Pattern Recognition and Artificial Intelligence (PRAI 2026)
Abstract:Mixed emotions represent a clinically relevant but still underexplored target for automatic emotion recognition. EEG provides millisecond-level access to neural activity, yet most EEG pipelines analyze the signal through a single temporal window, thereby fixing the temporal structure available to the model. This study introduces a multi-scale temporal framework for EEG-based emotion recognition. The EEG waveform is decomposed into windows of one or several durations, processed by a shared attention-based encoder, and integrated through a dynamic fusion module that assigns sample-specific weights across temporal scales. The framework is evaluated under a subject-independent protocol in binary and three-class settings, with the three-class task including the mixed affective category. The best results are 65.22% for the two-class task and 45.43% for the three-class task. Both are obtained with three-scale dynamic-fusion configurations and remain substantially above the full-signal baseline. The best-performing temporal scales differ between the two tasks. Dynamic fusion outperforms concatenation in the highest-scoring two-class configuration and slightly exceeds it in the highest-scoring three-class configuration, although these multi-scale settings require substantially more computation than the full-signal baseline.
[AI-17] A Unified Issue Resolution Benchmark for Requirement Clarification Planning and Code Generation for Coding Agents
链接: https://arxiv.org/abs/2608.09072
作者: Xin Zhou,Chun Yong Chong,Kisub Kim,Yun Peng,Rui Shu,Zihan Wu,Xu Han,Guowen Yuan,Zeyang Zhuang,Jounghoon Kim,Jeongjin Ju,Seongmin Ju,Taein Yoon,David Lo
类目: oftware Engineering (cs.SE); Artificial Intelligence (cs.AI)
备注: 9 pages
Abstract:Large language model-powered coding agents are increasingly used to modify existing code repositories, for example, by adding features or fixing bugs. Yet existing repository-level benchmarks typically evaluate only whether the final patch passes tests. Satisfying a user request requires a long chain of interdependent reasoning and decisions: an agent must recover explicit and implicit requirements, formulate a repository-grounded implementation plan, and translate it into correct code. A pass/fail outcome cannot characterize how an unsuccessful trajectory diverges from the requirements and implementation process needed for a correct patch. To address this gap, we introduce SWE-RPG, a repository-level benchmark that combines executable patch evaluation with validated ground-truth references (GTs) for (1) Requirement Clarification and (2) Implementation Planning. These intermediate GTs support retrospective, GT-aligned diagnosis of complete coding-agent trajectories across clarification, planning, code generation, and artifact submission. SWE-RPG comprises 163 tasks from 31 Python and Java repositories, including 113 bug fixes and 50 feature additions. We evaluate 3 coding agents, including Claude Code, Codex, and OpenCode, with 6 large language model backends, including Claude-Sonnet-5 and GPT-5.6-Terra. Results show that the evaluated popular coding agents still struggle to implement user requests in existing repositories, achieving an average resolved rate of only 31.5% on SWE-RPG. Intermediate-GT diagnosis further identifies implicit requirement recovery as the main bottleneck, accounting for 24.5%–46.0% of agent runs. This result suggests implicit-requirement recovery as a key candidate direction for improving coding agents. The benchmark data and evaluation code are available at this https URL.
[AI-18] wo-Step MV-DeepONet: Probabilistic Operator Learning for Uncertainty Propagation Driven by Random Input Fields
链接: https://arxiv.org/abs/2608.09071
作者: Yupei Nie,Lei Wang,Jiasen Liu
类目: Numerical Analysis (math.NA); Artificial Intelligence (cs.AI)
备注:
Abstract:Forward uncertainty propagation in complex physical systems can induce structured covariance across field-valued outputs. For a probabilistic surrogate, the total predictive covariance comprises the covariance of conditional means across input realizations and the average conditional predictive covariance. Probabilistic DeepONet (Prob-DeepONet) provides lightweight uncertainty quantification by predicting pointwise Gaussian means and variances in a single forward pass, but its conditional predictive covariance is restricted to a diagonal form. To represent cross-location conditional dependence without explicitly parameterizing a full high-dimensional covariance matrix, we develop a two-step mean-variance DeepONet (two-step MV-DeepONet) through two principal modifications. First, two-step training is used to decouple output-basis learning from the input-to-coefficient mapping, together with basis orthogonalization and subspace rotation. Second, Gaussian probabilistic modeling is transferred from the high-dimensional physical output space to the low-dimensional rotated coefficient space. Mapping these probabilistic coefficients through the shared basis induces a generally non-diagonal conditional predictive covariance in the physical output space while retaining single-pass inference. A Frobenius-norm error decomposition and corresponding upper bound identify low-rank covariance compressibility, trunk-subspace approximation, finite-sample statistical error, and coefficient-space covariance estimation as the principal factors governing covariance recovery. Numerical experiments on three representative problems governed by partial differential equations (PDEs) and a hypersonic blunt-body aerothermal problem show improved generalization, more structured uncertainty bands, and accurate recovery of off-diagonal correlation patterns compared with Prob-DeepONet.
[AI-19] DualCert: A Solver for the Traveling Salesman Problem with Constraint-Coupled Learning
链接: https://arxiv.org/abs/2608.09042
作者: Yancheng Song,Yongzhi Qi,Wei Qi,Zuo-Jun Max Shen
类目: Artificial Intelligence (cs.AI)
备注: 12 pages, 1 figure
Abstract:Large traveling salesman problem (TSP) instances require a solver to allocate limited computation while preserving the validity of its outputs. Existing neural–operations-research (OR) hybrids predict guidance without requiring learned transitions to satisfy constraints discovered during search. DualCert introduces \emphconstraint-coupled learning, in which current degree equations and dynamically separated subtour-elimination constraints (SECs) define each learned transition. At each refinement, the degree equations and selected, strictly satisfied SEC equations, with positive slacks, define an iterate-dependent primal-slack Karush–Kuhn–Tucker (KKT) manifold. Repaired dual variables and violated SEC rows define a local cost field. An exact constrained mirror-descent step maps each finite state to a positive state on the same manifold. Where selected rows and deterministic ties remain fixed, implicit differentiation maps parameter perturbations into the manifold tangent space and reuses the forward constraint operator for the local-cost-field derivative. The terminal edge state allocates computation across Held–Karp ascent, candidate-graph edge tests, and tour construction under a fixed budget. Deterministic verification recomputes original costs and accepts only verified candidate-graph lower bounds and edge decisions. On 1,000 held-out TSP1000 instances, DualCert attains a mean tour-cost gap of (0.0573%) from Lin–Kernighan–Helsgaun version 3 (LKH-3) reference tours in (9.55) batch-amortized seconds per instance. It returns a verified candidate-graph lower bound for every instance and achieves (81.46%) edge-decision coverage. The mean gap is (67.1%) smaller than the reported NeuroLKH mean gap. Thus, optimization constraints govern learning, while deterministic verification preserves output validity.
[AI-20] MusicLayout: Explicit Structural Planning for Controllable Text-to-Music Generation
链接: https://arxiv.org/abs/2608.09035
作者: Shuyu Li,Kejun Zhang,Jiahe Lei,Shulei Ji,Zihao Wang,Jiaxing Yu,Wanying Wu,Lei Wang
类目: ound (cs.SD); Artificial Intelligence (cs.AI); Multimedia (cs.MM)
备注:
Abstract:Text-to-music generation has advanced rapidly, but current systems still rely primarily on global text prompts, leaving the structural organization of generated music implicit and difficult to inspect, control, or revise before audio generation. To address this issue, we introduce MusicLayout, an explicit intermediate representation for controlling musical structure in text-to-music generation. MusicLayout describes a musical piece as a time-aligned layout of sections, textures, repetitions, variations, and instrument-level arrangements, serving as an interpretable planning layer between textual intent and the generated music. We integrate MusicLayout into a text-to-music framework built on a unified autoregressive formulation, where the model first generates a MusicLayout representation and subsequently predicts audio tokens conditioned on this representation within a single sequence. The resulting MusicLayout can be inspected and modified prior to audio generation, providing a mechanism for layout-level structural control. We evaluate MusicLayout through layout-conditioned generation, layout manipulation experiments, and matched-data ablations, providing evidence that explicit layout planning can improve long-range structural organization and support layout-level control.
[AI-21] Context Is Not Authority: Structured Runtime Governance for Financial Market Agents
链接: https://arxiv.org/abs/2608.09025
作者: Rui Tang,Qiangqiang Liu,Yichi Zhang,Youwei Wang,Xi Chen,Chen Dong
类目: Artificial Intelligence (cs.AI); Cryptography and Security (cs.CR); Machine Learning (stat.ML)
备注: 15 pages, 1 figure, 9 tables. Qiangqiang Liu and Yichi Zhang are corresponding authors
Abstract:Financial agents can turn correct context into an unauthorized effect: a customer-facing commitment, trade, or deployed policy. We present SAGE-Fin, a finance-specific authority-handoff contract that makes the proposed effect, not merely its text, the object of runtime control. SAGE-Fin compiles proposals into typed, adapter-bound candidates; records missing or stale institutional obligations as coverage debt; contracts authority under current market, account, policy, and dialogue state; and requires an exact-artifact receipt whose nominal type matches the consuming response, execution, or policy adapter. Evidence and workflow progress cannot substitute for effect authority, and prior authorization is rechecked after state changes. Across an authored 616-case catalog, five deterministic specifications yield 3,080 outputs; a label-isolated harness obtains 616/616 binary reference-prototype parity, including 3/3 named response-gate fixtures, while 22 tests cover selected paths. These results establish executable conformance, not independent safety accuracy. Separately, SAGE-Fin’s response gate processed real customer-facing production requests at a confidential digital-asset platform. An operational team independent of the implementation team reached a strongly positive post-deployment conclusion on practical usefulness and workflow fit, and end-user feedback was also strongly positive. Disclosure permits only the review’s independence, stakeholder classes, assessed dimensions, and directional conclusion, so this is qualitative field corroboration rather than an aggregate effect estimate. Three distinct de-identified predecessor failures, with independently confirmed 0/3 interception, ground repeated-emission drift, stale account evidence, and missing escalation state without estimating prevalence or treatment effect.
[AI-22] Hybrid Neural-Classical Correction for Frozen Time Series Foundation Models: A Comprehensive Ablation Study on High-Frequency Stock Prediction IJCNN2026
链接: https://arxiv.org/abs/2608.08825
作者: Kasun Dewage,Suranadi De Silva,Shankhadeep Mondal
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Statistical Finance (q-fin.ST)
备注: Accepted and presented at IJCNN 2026, part of the IEEE World Congress on Computational Intelligence (WCCI 2026)
Abstract:Foundation models for time series forecasting demonstrate impressive zero-shot generalization but often underperform on specialized domains such as high-frequency finance. We present a comprehensive study of hybrid neural-classical correction for adapting frozen TimesFM (200M parameters) to stock return prediction during the volatile opening trading hour. We compare two neural correction architectures - AttnCorrect (multi-head self-attention, approximately 471K parameters) and GatedLinear (low-rank bilinear projection with gating, approximately 49K parameters) - each augmented with Random Forest residual learning. Through systematic ablation across 10 major technology stocks (NVDA, MSFT, AAPL, GOOG, GOOGL, AMZN, META, AVGO, TSLA, NFLX) spanning 2 million data points, we reveal critical insights: (1) The hybrid neural-classical approach achieves 0.597 pooled correlation and 6.4x mean per-day correlation improvement over frozen TimesFM; (2) Classical residual learning (Random Forest) provides the largest single-component contribution, matching or exceeding the neural correction component; (3) Simpler neural architectures surprisingly outperform complex ones when classical residual learning is removed; (4) Self-attention provides the largest neural-only contribution. GatedLinear+RF achieves best overall performance with 9x fewer neural parameters than AttnCorrect+RF. We report three complementary correlation metrics - mean per-day, cross-day cumulative, and pooled - to provide a complete picture of predictive quality. Our results provide practical guidance: effective foundation model adaptation requires careful integration of neural and classical components, with classical methods playing a crucial complementary role.
[AI-23] hree Generations of Healthcare IT: From the Digital Record to the Computable Care Process
链接: https://arxiv.org/abs/2608.08806
作者: Alexander Apartsin,Yehudit Aperstein
类目: Artificial Intelligence (cs.AI)
备注: 11 pages, 3 figures
Abstract:Objective. Healthcare IT is usually organized by the technologies it adopts. We instead organize it by the unit of information a system makes computable, and describe a computational layer whose object is patient-specific clinical intent. Approach. We give criteria for a computational layer, derive three (record, clinical state, and a proposed layer of intent), and formalize the Actionable Clinical Record (ACR) as the atomic object of the third layer. Discussion. The framework distinguishes prescribed, observed, and intended process; existing standards represent intent once it is structured but do not recover it from natural communication, the capability we localize. The ACR is complementary to FHIR workflow resources, guidelines, and process mining; a companion feasibility study illustrates tractability for one narrow subproblem. Conclusion. Computable clinical intent is a coherent research direction; the ACR, its readiness ladder, and an executable-correctness evaluation framework are reusable constructs for subsequent work to extend, evaluate, or falsify.
[AI-24] Improving Generalization Robustness of Multimodal RLVR
链接: https://arxiv.org/abs/2608.08802
作者: Pengfei Zhou,Zhiwei Tang,Xiaopeng Peng,Chenrui Zhou,Lama Moukheiber,Yixing Ma,Bin Xu,Jiajun Song,Zhenglin Wan,Wangbo Zhao,Jiasheng Tang,Bohan Zhuang,Fan Wang,Yang You
类目: Artificial Intelligence (cs.AI)
备注: 32 pages, 5 figures
Abstract:Reinforcement Learning with Verifiable Rewards (RLVR) makes Multimodal Large Language Models more accurate, but the gains are brittle: simply paraphrasing a question or changing the prompt template can degrade them, which challenges reliable deployment in high-stakes scenarios like medical VQA. We trace this to two issues of the standard RL objective. First, the binary verifier conflates format with content, so the reward signal cannot tell a wrong answer apart from a misformatted one. Second, the training distribution covers only a thin slice of the real-world prompts that the model might meet at deployment, so policies that perform well on the training distribution can behave differently under unseen prompts during test. Both failures call for a robust post-training method that helps the policy cover a broader distribution of semantically equivalent prompts, and we identify two measures that help achieve this objective: separating format from semantics in the reward, and applying policy invariance across perturbed prompts with equivalent semantics. We therefore propose Prompt-Invariant RLVR (PIRL), consisting of a dynamic trinary reward and a consistency regularizer based on an embedding-space adversary. Under stress testing, PIRL’s average accuracy on benchmarks drops by only \le 1% , where GRPO drops ~3%. On dynamic evaluation, PIRL also achieves the smallest performance drop.
[AI-25] Deferred Audio Pruning with Local Audio-Visual Dynamics for Omni-LLM s
链接: https://arxiv.org/abs/2608.08794
作者: Kyeongyoon Lee,Hongyeob Kim,Youngeun Kim,Sungeun Hong
类目: Artificial Intelligence (cs.AI); Multimedia (cs.MM); Sound (cs.SD)
备注:
Abstract:Omni-modal LLMs jointly process audio, video, and text, but long multimodal sequences incur substantial prefill and KV-cache costs. Existing omni-modal compression methods primarily focus on pre-LLM token reduction, leaving modality-specific compression across the LLM boundary underexplored. We propose A-PACK, a two-stage framework that defers audio pruning until query-conditioned multimodal interactions emerge. Our analysis shows that audio exhibits higher task-relevant information density and representational diversity per token than video. We further find that local audio-visual dynamics provide a more effective cue for visual selection than token-wise matching. We therefore preserve audio and compress video with local dynamics before the LLM, then progressively prune low-relevance audio and visual tokens and their KV-cache entries inside the LLM. Across four benchmarks on Qwen2.5-Omni-7B/3B, A-PACK achieves the strongest average performance among the evaluated prior methods while reducing prefill FLOPs by up to 78% and improving decoding throughput by up to 2.21x.
[AI-26] SymDiag: Explainable Diagnosis for LLM Reasoning via Neuro-Symbolic Verification
链接: https://arxiv.org/abs/2608.08786
作者: Wenyao Cui,Huaping Zhang,Yongyi Huang,Qiuchi Li,Jian Xu,Cheng-Lin Liu,Chunxiao Gao,Juan Wang,Baohua Zhang
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Large language models (LLMs) increasingly serve as data-driven reasoners, yet their chains-of-thought (CoT) can be unfaithful even when final answers are correct. Most existing verification'' signals are not diagnostic: answer matching observes only the outcome, LLM-as-judge provides subjective and non-verifiable critiques, and scalar rewards (e.g., PRMs/RMs) offer little insight into where a multi-step derivation this http URL propose \textbfSymDiag, a neuro-symbolic framework that \textbfreframes reasoning verification as structured failure diagnosis. SymDiag translates natural-language CoT into symbolic constraints and performs step-level satisfiability/entailment checks to (i) localize failing steps and (ii) produce verifiable diagnostic evidence, including counterexamples, inconsistency witnesses, and missing-premise indicators. A central challenge is that apparent logic violations’’ can be caused either by genuine reasoning defects or by neural-to-symbolic translation noise. SymDiag therefore incorporates a Self-Auditor that disentangles TranslationError from ReasoningError via dual symbolic encodings consistency checks, enabling robust diagnosis under partial observability. Across diverse mathematical, logical, scientific, and general reasoning benchmarks, SymDiag improves detection of unfaithful reasoning and provides substantially more effective feedback for multi-round reasoning repair than outcome-only verification and LLM-based judging, offering a principled foundation for trustworthy and scalable reasoning diagnosis.
[AI-27] Learning from Consensus and Disagreement: Unsupervised On-Policy Self-Distillation with Minority-Trajectory Contrast
链接: https://arxiv.org/abs/2608.08764
作者: Jiaxin Guo,Yanwei Yue,Xuanbo Fan,Chunyu Yang,Yan Zhang
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:On-policy self-distillation improves language-model reasoning by querying a teacher on states actually visited by the student. Recent methods create a powerful information asymmetry by exposing the teacher to privileged context, yet they fundamentally rely on external supervision—such as gold solutions or verifiers—to construct this advantage. We introduce CoDA (Consensus and Disagreement Alignment), a fully unsupervised framework that creates reliable privileged information entirely from the latent uncertainty structure of a model’s own unlabeled rollouts. CoDA extracts two complementary signals. In the positive branch, answer-level consensus identifies a stable reasoning mode, which conditions a frozen self-teacher to provide dense distributional guidance on fresh student trajectories. However, because agreement does not guarantee correctness, positive-only distillation risks amplifying correlated errors into a false consensus. To break this harmful feedback loop, CoDA incorporates a negative branch that exploits disagreement: minority trajectories are treated as unstable alternatives and gently penalized via a reference-anchored, KTO-style calibration objective. This unpaired binary feedback provides robust regularization without requiring the strong assumption that the consensus is the absolute ground truth. Empirical evaluations on competition-level mathematical benchmarks demonstrate that CoDA significantly improves reasoning, outperforming self-generated baselines and effectively stabilizing training against erroneous consensus.
[AI-28] Eco-SoC: A Sustainable VLSI Architecture for Energy-Proportional Artificial Intelligence
链接: https://arxiv.org/abs/2608.08761
作者: Jatin Chopra
类目: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI)
备注:
Abstract:In an era defined by escalating climate change and the pervasive deployment of edge intelligence, the environmental cost of semiconductor manufacturing and operation has reached a critical threshold. As Deep Learning (DL) accelerators dominate System-on-Chip (SoC) die area, achieving true sustainability requires a paradigm shift from static worst-case efficiency to dynamic energy-proportionality. This paper introduces Eco-SoC, a highly scalable VLSI architecture co-designed specifically for sustainable artificial intelligence. We propose a hardware-level Dynamic Precision-Scaling Logic (DPSL) framework that adaptively modulates bit-width precision based on real-time activation sparsity, successfully reducing switching activity by up to 42% on a commercial 7nm FinFET process node. Furthermore, we transcend traditional Power-Performance-Area (PPA) metrics by providing a comprehensive Life Cycle Assessment (LCA) using the Architectural Carbon footprint Tool (ACT). Our synthesis demonstrates that Eco-SoC offsets its increased embodied carbon footprint (a marginal 4.8% area overhead) within 1.1 years of edge deployment. Finally, by introducing a thermal-aware power gating mechanism that mitigates localized hotspots, Eco-SoC doubles the projected Mean Time To Failure (MTTF) of the silicon, providing a tangible, scalable strategy for electronic waste (e-waste) mitigation in next-generation computing systems. Subjects: Hardware Architecture (cs.AR); Artificial Intelligence (cs.AI) ACMclasses: B.7.1; C.1.3 Cite as: arXiv:2608.08761 [cs.AR] (or arXiv:2608.08761v1 [cs.AR] for this version) https://doi.org/10.48550/arXiv.2608.08761 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[AI-29] Scale-to-Dialogue: Low-Burden Elicitation of Daily Premenstrual Symptom Ratings with Small Language Models
链接: https://arxiv.org/abs/2608.08746
作者: Yifan Wang
类目: Artificial Intelligence (cs.AI)
备注: 10 pages, 4 figures, 1 table
Abstract:Prospective daily symptom tracking is central to premenstrual health assessment, but repeated ordinal forms impose substantial response burden. We formulate conversational administration as an ordinal label-recovery problem: the system actively elicits a small set of symptom clusters and maps each response to the original severity labels. We used 3,320 complete participant-days from the mcPHASES dataset, covering cramps, mood swing, fatigue, sleep issues, stress, and bloating on a six-level scale. Six participants were reserved for development and 36 for a frozen evaluation comprising 360 participant-days and 2,160 item labels. A ModernBERT evidence gate detected whether a symptom was expressed, and Qwen2.5-1.5B-Instruct produced deterministic structured severity scores. Fixed six-item questioning achieved a quadratic weighted kappa of 0.976, whereas three joint symptom-cluster questions achieved 0.913, 97.45% agreement within one severity level, and 80.94% recall for moderate-or-higher symptoms while reducing questions by 50%. Open-first adaptive policies required 3.92-5.98 questions and produced lower agreement than the corresponding fixed policies. Participant-cluster bootstrap analysis estimated a kappa difference of -0.062 (95% CI -0.076 to -0.048) between the three-cluster and six-item strategies. Active cluster-level elicitation provides a direct, local-model route from natural conversation to reusable daily symptom labels.
[AI-30] FitAQA: A Benchmark of Fitness Action Quality Assessment for Multimodal Large Language Models
链接: https://arxiv.org/abs/2608.08736
作者: Kaili Zheng,Kaiwen Wang,Xun Zhu,Qingyuan Yang,Chenyi Guo,Ji Wu
类目: Artificial Intelligence (cs.AI)
备注:
Abstract:Fitness Action Quality Assessment (AQA) is important for intelligent sports training, yet the capabilities of Multimodal Large Language Models (MLLMs) in this setting remain underexplored. Existing benchmarks rely on action-specific annotation schemes and focus primarily on final assessment outputs, offering limited insight into how models assess exercise quality. We introduce FitAQA, a systematic benchmark for evaluating MLLMs in fitness AQA, containing 2,219 videos and 5,512 QA instances across 30 bodyweight exercises. In collaboration with experts in sports science, we develop a unified form error taxonomy that defines 38 recurring form errors within six complementary quality dimensions: alignment, symmetry, stability, coordination, tempo, and completeness. This taxonomy provides a shared assessment framework across different exercises. FitAQA further formulates three evaluation tasks: perception for recognizing relevant visual evidence, judgement for combining that evidence with domain knowledge to assess execution correctness, and temporal grounding for localizing form errors over time. Extensive evaluation shows that current MLLMs still struggle to assess exercise quality comprehensively and localize form errors precisely. Controlled experiments further indicate that visual perception is a key bottleneck, as judgement performance improves substantially when ground-truth perceptual evidence is provided. The dataset and evaluation code will be made publicly available.
[AI-31] PAST: Privileged Adaptation from Complete Student Trajectories for On-Policy Self-Distillation
链接: https://arxiv.org/abs/2608.08726
作者: Yangyang Feng,Zhuoyan Feng,Junlan Chen
类目: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
备注:
Abstract:On-policy self-distillation (OPSD) uses a privileged teacher to supervise a reasoning model on prefixes sampled from its own rollouts. Yet each rollout also reveals how the student’s response unfolds and whether it succeeds, student-specific hindsight that standard OPSD does not use to form the teacher. We introduce Privileged Adaptation from Student Trajectories (PAST), which treats each completed student trajectory as additional privileged information for the OPSD teacher while leaving the student’s distillation prefixes unchanged. PAST preserves the student’s next-token distribution on correct trajectories and uses failed trajectories to adapt the teacher toward verified success under student-proximity regularization. We characterize what such a trajectory-conditioned teacher can transfer to a prefix-only student. Forward-KL distillation projects the teacher distributions to their conditional arithmetic mean given the prefix. This projection separates trajectory-specific variation that remains privileged from the mean policy shift available to the student. For correct trajectories, the unclipped population objective also has the frozen student as an ideal distributional fixed point. Across three mathematical reasoning benchmarks, PAST improves the Avg@12 macro average over Vanilla OPSD by 5.6 percentage points. A 2\times2 factorial study shows gains from both complete-trajectory access and teacher adaptation, while trajectory removal and shuffling confirm that the adapted teacher uses the matching hindsight context.
[AI-32] he Field Knows: Cross-Dimensional Geometry from Navigation to Black Holes
链接: https://arxiv.org/abs/2608.07566
作者: Chenghao Xu
类目: Artificial Intelligence (cs.AI); Robotics (cs.RO)
备注:
Abstract:We introduce a continuous metric field framework trained by a single causal contrastive loss. The framework encodes a scene into coefficients of a fixed symmetric matrix basis, assembles them into a Lie algebra element, and exponentiates the result to a Riemannian or Lorentzian metric. Across dimensions, this field discovers the full spectrum of geometric structures: from obstacle-avoiding geodesics in robot navigation across planar and manipulator configuration spaces, to event horizons of black holes in Lorentzian spacetime. Extensive zero-shot generalization studies demonstrate that the field captures transferable geometric structure rather than memorizing specific configurations. In the black hole setting, the causal loss spontaneously evolves genuine black-hole-like structures with the correct Lorentzian signature. The same loss, the same architecture, and the same training protocol produce the full range of geometric phenomena across dimensions. The field knows geometry, and geometry knows physics.
[AI-33] Generalizing deep reinforcement learning across cable-driven parallel robot configurations with actuator-level policies
链接: https://arxiv.org/abs/2608.07546
作者: Abir Bouaouda,Mohamed Boutayeb(CRAN, UIR),François Charpillet(LARSEN),Dominique Martinez(LORIA, ISM),Rémi Pannequin(CRAN)
类目: Robotics (cs.RO); Artificial Intelligence (cs.AI)
备注:
Abstract:Cable-driven parallel robots (CDPRs) present diverse configurations and complex control challenges, which can be addressed by deep reinforcement learning (DRL) by learning their nonlinear dynamics. However, DRL methods often require extensive training time, and the resulting policies do not generalize well to different robot configurations or varying numbers of actuators. In this article, we introduce a novel DRL approach for controlling CDPRs that does not depend on the specific robot configuration. Our method trains an actuator-level policy that controls each motor to achieve its target cable length, in contrast to conventional DRL approaches that learn to control the entire robot to reach a desired end-effector position. To the best of our knowledge, this is the first work to apply DRL to control CDPRs using an actuator-level policy. This approach offers two main advantages: (i) a single shared policy can be applied to any CDPR configuration, regardless of actuator count, and (ii) reliance on inverse kinematics, avoiding the more challenging forward kinematics problem. Training is performed in simulation, and the learned policy is successfully transferred to a real CDPR. Experimental results show that the actuator-level policy (ALP) surpasses traditional reinforcement learning methods in both robustness and precision. We further control a real 8-motor CDPR with 3D motion using a policy trained on a simulated 4-motor planar CDPR operating in 2D. This illustrates that the proposed method is applicable to any CDPR configuration, independent of actuator number or placement.
[AI-34] DarwinX: Evolving Agent Harnesses Through Natural Selection
链接: https://arxiv.org/abs/2608.07545
作者: Yifan Zhang,Yutong Dai,Juntao Tan,Luyu Yang,Rishi Mullur,Thai Hoang,Zhiyuan Hu,James Zhu,Phil Mui,Silvio Savarese,Ran Xu,Zeyuan Chen
类目: Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI); Machine Learning (cs.LG); Software Engineering (cs.SE)
备注:
Abstract:An LLM agent’s capability depends not only on model weights but on its harness: prompts, tools, skills, and control flow. Self-improvement loops already edit harnesses, yet single-lineage search is path-dependent and local wins often regress other tasks. We introduce DarwinX, which treats self-evolution as selection over a population of harnesses with the model frozen: a preserve-and-extend contract admits only variants that extend coverage without regressing, an archive keeps alternative lineages for recombination, and failure-, teacher-, and self-derived evidence share one edit interface. Fitness comes from each benchmark’s own verifier: no gold solutions, no hand-picked winners. Across four benchmarks that progressively separate the evolution signal from the test, one loop adds about 17 points on average: Terminal-Bench 2.1 rises +7.7 to 83.2% on a matched base and to the verified frontier at 84.7% on a stronger one; TerminalWorld’s held-out split reaches 68.3%, ahead of every off-the-shelf agent; WebArena-Infinity real-task pass@1 rises from 43.5% to 93.0% audit-clean; and a Terminal-Bench 2.1 harness transfers unchanged to SWE-bench Verified. What evolves is general agent competence, not benchmark-specific patches, so it survives changes of task, verifier, and base model. A frozen model need not be a fixed agent: harness selection turns evaluation compute into durable capability.
[AI-35] MOSAIC: Adversarial Co-evolution of Specialist Heuristics and Problem Instances for LLM -based Automated Heuristic Design
链接: https://arxiv.org/abs/2608.07544
作者: Oguzhan Gungordu,Siheng Xiong,Faramarz Fekri
类目: Neural and Evolutionary Computing (cs.NE); Artificial Intelligence (cs.AI)
备注:
Abstract:Automated heuristic design (AHD) with large language models (LLMs) has produced strong heuristics for combinatorial optimization problems (COPs). Yet existing frameworks optimize for average performance on a small fixed dataset and steer the search with “verbal gradients” distilled from scalar better/worse feedback. No single heuristic dominates across instance distributions, and scalar feedback tells the LLM whether a heuristic improved, but not where in the instance space or why. We propose MOSAIC, a grid-based framework that adversarially co-evolves problem instances and specialist heuristics inside a Quality-Diversity (QD) archive indexed by structural instance features. Instances evolve to expose weaknesses of the current heuristics, and heuristics evolve to eliminate them by specializing to the newly exposed regions. Each archive cell keeps a specialist heuristic, representative instances, and insights explaining what works in its region, forming a persistent memory that accumulates over the evolutionary search. For each heuristic pair sampled from distant grid regions, an LLM-guided evolutionary loop generates discriminative instances, and a decision tree identifies the feature-space regions where each heuristic wins. A reflection LLM then contrasts the two heuristics to produce multi-directional insights that persist in those regions and guide crossover and mutation. The archive is simultaneously a co-evolved benchmark of discriminative instances and a pool of region specialist heuristics, from which greedy selection extracts a compact complementary portfolio. Across COPs, test sizes, and LLM backbones, the portfolio consistently outperforms state-of-the-art LLM-based AHD methods, and the co-evolved instances attain higher feature-space coverage and stronger heuristic discrimination than evolutionary instance-generation baselines.
[AI-36] REAT: Evaluating Access to Formal Knowledge across Equivalent Mathematical Representations
链接: https://arxiv.org/abs/2608.07540
作者: Fateme Mazdarani,Carlos Toxtli
类目: Artificial Intelligence (cs.AI)
备注: Accepted at 28th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2026)
Abstract:AI systems increasingly operate between flexible input representations and formal objects used by downstream tools. A key challenge is recognizing when an unfamiliar formulation denotes a known formal object. We study this challenge through theorem recognition: given an equivalence-preserving transformation of a theorem condition, a model must recover the theorem identity associated with the standard statement. We introduce TREAT, a benchmark for evaluating whether large language models can recover known theorem identities from equivalence-preserving formula-level transformations. Rather than paraphrasing theorem text, TREAT changes the mathematical form of theorem conditions themselves, expressing known results through residual equations, witness statements, optimization identities, set relations, operator forms, and proof-intermediate characterizations. Starting from scraped theorem pages, we filter for entries with usable mathematical expression forms, extract canonical theorem conditions, and generate transformed variants with recorded assumptions and inverse mappings. The final corpus contains 737 theorem identities and 29,480 transformed rows. On a test panel, the best model retrieves the correct theorem identity in only 60.73% of cases. Other systems reveal different failure modes, including abstention, wrong detection, and malformed outputs. These suggest that theorem knowledge can be fragile under equivalent changes in representation. TREAT therefore provides a controlled testbed for evaluating representation-robust access to formal knowledge, with broader relevance to domains that require stable target objects, explicit equivalence relations, validation procedures, and auditable scoring.
[AI-37] When LLM Agents Negotiate: Private Information and Dynamic Bargaining in Supply Chains
链接: https://arxiv.org/abs/2608.07538
作者: Chen Liang,Fasheng Xu
类目: Artificial Intelligence (cs.AI); Computer Science and Game Theory (cs.GT); General Economics (econ.GN)
备注:
Abstract:As LLM agents move from decision support to autonomous procurement, firms need to know whether delegated negotiators create value, divide it predictably, and avoid money-losing contracts. We study this in a canonical supply chain bargaining problem: a buyer with private demand information negotiates a quantity-payment contract with an uninformed seller. We benchmark nine LLMs from OpenAI, Google, and Alibaba against a validated Perfect Bayesian Equilibrium across 9,840 LLM-to-LLM negotiations. First, capability governs value creation. Agents agree in 98.9% of negotiations and capture 95.4% of first-best surplus undiscounted, but average 2.98 rounds against the benchmark’s 1.25, and this delay erodes 21-34% of surplus. Capability also governs reliability: baseline models accept individually irrational contracts in 19.2% of cases, versus 0.0-0.6% at mid-tier and flagship, making automated profit verification the binding guardrail below that threshold. Second, surplus capture is relational. Provider identity predicts who captures surplus better than capability rank: self-play buyer shares average 40% for OpenAI, 50% for Google, and 70% for Alibaba’s Qwen, an ordering that survives restricted communication and no discounting. Reversing which provider sells moves the division by 7-18 percentage points, and the capable Qwen flagship is the weakest cross-family seller: vendor choice is a first-order distributional decision. Third, the prompt is a strategic lever. Delegation separates the principal’s economic patience from the agent’s prompted strategic patience, a free deployment choice that is the single strongest driver of surplus division (90% of explained variance). Together these establish an equilibrium-referenced audit of AI agents along three dimensions: discounted efficiency, distributional profile, and operational reliability.
机器学习
[LG-0] racking the Best Strategy in an Extensive-Form Game
链接: https://arxiv.org/abs/2608.09501
作者: Stephen Pasteris,Rahul Savani,Theodore Turocy
类目: Machine Learning (cs.LG)
*备注:
Abstract:We consider the extensive-form bandit problem where on each trial the learner plays an extensive-form game against an oblivious adversary. We focus on the notion of switching regret, which measures the expected performance of the learner against that of any switching sequence of mixed strategies in retrospect. Our algorithm takes a parameter \rho0 and achieves a switching regret of \tilde\mathcalO((1/\rho+\rho K)\sqrtH A T) where K is the number of switches in the comparator sequence, H is the maximum number of the learner’s information sets that can be traversed during a play of the game and A is the number of actions that the learner can possibly take. Our algorithm is extremely efficient, taking a per trial time of only \mathcalO(H B) where B is the maximum number of actions available to the learner at any of its information sets.
[LG-1] Walk-on-Spheres Monte Carlo and deep neural network approximations of elliptic PDEs with drift and killing
链接: https://arxiv.org/abs/2608.09494
作者: Konrad Kleinberg,Thomas Kruse
类目: Numerical Analysis (math.NA); Machine Learning (cs.LG); Analysis of PDEs (math.AP); Probability (math.PR)
*备注:
Abstract:In this paper we provide Monte Carlo and deep neural network approximations for stochastic representations of solutions to linear elliptic partial differential equations with constant diffusion, drift and killing. Building on the modified Walk-on-Spheres algorithm of Beznea et al. (arXiv:2209.01432), we introduce Monte Carlo estimators that explicitly incorporate sampled random times arising in the analyzed stochastic representations. We establish uniform error bounds for these estimators and show that, under suitable assumptions, a prescribed approximation accuracy is achieved with sample complexities growing at most polynomially in both the inverse accuracy and the problem dimension. Furthermore, we prove a deep neural network approximation result for the stochastic representations. Assuming suitable neural network representations of the boundary data and the distance function to the boundary, we use the constructed Monte Carlo to design deep neural networks that approximate the representation uniformly with a number of parameters growing at most polynomially in the inverse accuracy and the problem dimension. These results extend previous complexity analyses to a broader class of elliptic equations involving drift and killing.
[LG-2] When Do Task Vectors Interfere? Mapping the Validity Boundaries of Weight-Space Composition
链接: https://arxiv.org/abs/2608.09490
作者: Chencheng Zhu,Xiaoyang Li,Taotao Cai
类目: Machine Learning (cs.LG)
*备注: 12 pages, 3 figures; includes references and an expanded appendix
Abstract:Task arithmetic treats fine-tuning displacements as composable directions in weight space, yet it remains unclear when parameter addition reflects predictable changes in model function. We separate parameter geometry from functional geometry and measure pairwise functional non-additivity over a two-dimensional task-vector surface, using a first-token predictive-distribution interaction ratio conditioned on an input distribution and evaluated with norm-matched controls, three training seeds, and response-only fine-tuning. On Qwen2.5-1.5B, code+safety is more non-additive than the matched code+math control on code and instruction prompts, but not on math prompts. In a prospectively specified six-task expansion, all eight high-versus-low comparisons of unseen task pairs have the predicted sign. The primary ordering further persists under full-parameter fine-tuning at 0.5B, Qwen2.5 LoRA scale tests up to 7B, and a Llama-3.1-8B cross-architecture audit. External validation exposes a sharper boundary: raw public code, instruction, and safety prompts preserve the continuous contrast, whereas an instruction-style wrapper collapses it on the identical public-code prompts, and EvalPlus pass@1 interactions do not robustly reproduce it. Weight-space composition therefore supports coarse, input- and format-conditioned functional statements across adaptation methods, scales, and one additional model family, not a universal merging-performance predictor.
[LG-3] Hierarchical rank-evolving representation for physics-informed neural networks
链接: https://arxiv.org/abs/2608.09483
作者: Ruoyang Su,Xi-Le Zhao,Kun Li,Liang Li
类目: Machine Learning (cs.LG); Numerical Analysis (math.NA); Computational Physics (physics.comp-ph)
*备注:
Abstract:Recently, tensor-based physics-informed neural networks (T-PINNs) have received increasing attention. However, existing T-PINNs still face a fundamental challenge: they mainly rely on pre-specified low-rank tensor decompositions with manually tuned ranks, which limits their ability to capture the underlying structures of multivariate solution functions and hinders their practical deployment. To address this challenge, we propose a hierarchical rank-evolving (abbreviated as HRE) representation for multivariate functions, which endows us to faithfully capture the underlying structure of the targeted multivariate function accompanying with automatic rank determination. Concretely, in the hierarchical design of HRE representation, the target multivariate function is decomposed as a small-scale inner tensor with a set of univariate functions along each mode, where a customized tensor network decomposition can be readily deployed to capture the underlying structure of the small-scale inner tensor. In HRE representation, the crucial hyperparameters, ranks, can be adaptively revealed during the decomposition, freeing us from manual rank tuning and making HRE practically applicable to real-world problems. Besides, we build the HRE-PINNs correspondingly. Extensive numerical experiments, including high-dimensional static problems (Helmholtz equation and Poisson equation), nonlinear time-dependent problems (Klein-Gordon equation), and complex fluid-dynamics problems (flow mixing equation and Navier-Stokes equation), demonstrate that HRE-PINNs consistently outperform existing state-of-the-art approaches in terms of accuracy.
[LG-4] From Approachability Residuals to Anytime-Valid Evidence: The Online Convex Geometry of Testing by Betting
链接: https://arxiv.org/abs/2608.09450
作者: Jinze Zhao
类目: Machine Learning (cs.LG)
*备注:
Abstract:Betting-based sequential tests and Blackwell approachability are linked by a rate-explicit reduction through support-function residuals. For a compact convex target S and vector observations r_t , an OCO learner selects a predictable normal w_t and produces q_t=\langle w_t,r_t\rangle-h_S(w_t) . We prove the exact pathwise identity \dist(\bar r_T,S) =\frac1T\sum_t=1^Tq_t+\frac\Reg_TT. When |q_t|\leq B , composing this identity with one-sided betting yields a finite-time transfer: if the OCO and log-wealth regrets are at most a_T and \ell_T , respectively, then a target gap exceeding [ \fraca_TT +2B\sqrt\frac\log(1/\alpha)+\ell_TT ] forces rejection by time T , while non-rejection certifies the converse radius. We then formulate a controlled stochastic experiment in which an action selected after w_t satisfies Blackwell’s supporting-halfspace condition for every null mean payoff. The resulting wealth is an e-process under adaptive nulls; sublinear OCO regret gives stochastic approachability, whereas persistent mean separation under an alternative gives exponential wealth at rate at least \delta^2/(4B^2) . Deterministic Blackwell games and passive tests are, respectively, the noise-free and singleton-action cases of this protocol. Bounded two-sample means, kernel MMD, and active heterogeneous data sources instantiate the reduction. The resulting connection is exact algebraically, quantitative at finite time, and operational when experiments are controlled. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.09450 [cs.LG] (or arXiv:2608.09450v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.09450 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Jinze Zhao [view email] [v1] Mon, 10 Aug 2026 11:23:30 UTC (18 KB) Full-text links: Access Paper: View a PDF of the paper titled From Approachability Residuals to Anytime-Valid Evidence: The Online Convex Geometry of Testing by Betting, by Jinze ZhaoView PDFHTML (experimental)TeX Source view license Current browse context: cs.LG prev | next new | recent | 2026-08 Change to browse by: cs References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[LG-5] A Mechanistic Diagnostic of Rank Collapse in Post-Norm Decoder Transformers
链接: https://arxiv.org/abs/2608.09417
作者: Xingjian Wang,Qingyu Han,Xiaodong Luo,Yin Zhang
类目: Machine Learning (cs.LG)
*备注:
Abstract:Deep decoder-only Transformers often replace the original Post-Norm architecture with Pre-Norm variants because Post-Norm training is highly sensitive to warmup and learning rate under conventional initialization schemes. Although prior work has identified rank collapse and gradient vanishing as related symptoms, it remains poorly understood how causal attention creates high-similarity representations and why training dynamics fail to repair them. We give a two-stage analysis of Post-Norm rank collapse using token similarity as a scalar state variable. First, at initialization, causal attention acts approximately as a prefix-averaging operator that increases token similarity across depth, while the SwiGLU branch contributes only a smaller damping effect. Second, once training enters a high-similarity regime, growth of pre-normalization residual norms makes the RMSNorm backward factor contractive; under mild conditions, gradients to earlier layers decay geometrically. As a complementary result, we characterize the properties of a collapsed network: its best predictor is frequency distribution with relatively high loss floor, and gradients in collapsed layers vanish at frequency distribution. Experiments on 48-layer decoder-only Transformers trained on C4 dataset match the predicted initialization-time similarity growth and collapse-time gradient contraction, and show that collapsed runs stay near the predicted frequency loss. Together, these results distinguish the forward similarity amplification and backward repair incapacity in Post-Norm collapse, while also characterizing the behavior of collapsed networks.
[LG-6] From Objectives to What Models Learn: A Landau Theory of Invariant Learning
链接: https://arxiv.org/abs/2608.09396
作者: Pinli Wang,Yue He,Peng Cui
类目: Machine Learning (cs.LG)
*备注:
Abstract:Invariant learning seeks representations that remain predictive across environments, yet the behavior of its objectives along the regularization path is often opaque. We address this objective-behavior gap by viewing representation learning as multimode magnetization and deriving, from concrete invariant-learning objectives, a Landau-type effective free energy whose low-order coefficients form objective signatures and induce distinct regularization phenotypes. Effective quadratic corrections move the phase boundary and enable finite-strength mode elimination; quartic corrections regulate post-onset amplitude and typically leave residual loading at finite strength; higher-order structure governs non-monotone tails, instability, and collapse at large regularization. In a canonical bilinear model, the theory yields closed-form phase boundaries and steady-state loadings, as well as distinct critical strengths for shortcut and stable modes that define a selective-retention window. Controlled experiments confirm the predicted phase boundaries, loadings, and regularization phenotypes. In one- and two-hidden-layer ReLU networks, the same signatures remain predictive of qualitative regularization-path behavior despite depth-dependent shifts in scale. A matrix extension generalizes the framework to coupled collective modes and yields a spectral phase-boundary criterion. Together, the framework turns low-order objective signatures into predictions of regularization phenotypes and, ultimately, of what models learn as regularization varies.
[LG-7] Regret equilibrium and learning in games: A guided tour
链接: https://arxiv.org/abs/2608.09389
作者: Panayotis Mertikopoulos
类目: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 44 pages, 3 figures; to appear as a chapter in “Equilibria in Games: Existence, Selection, and Dynamics”, edited by Sylvain Sorin and Bernhard von Stengel
Abstract:This note aims to serve as an entry point to the literature on learning in games, a topic with significant theoretical appeal and a wide range of applications – from machine learning and data science to economics and beyond. Our presentation is structured around two complementary viewpoints: We first consider a single agent – the learner – engaged in a sequential decision process in an unknown, non-stationary, and possibly adversarial environment. We then examine what happens when the environment is shaped by the decisions of several interacting agents, not necessarily aware of each other’s actions or goals, and all seeking to improve their individual rewards. In this general context, we examine a family of regularized learning policies based on best-responding to the past history of play, up to a regularization penalty intended to encourage exploration and prevent over-commitment to suboptimal choices. In the single-agent setting, we present some basic regret bounds for regularized learning in adversarial multi-armed bandits; in the multi-agent setting, we describe an ergodic equilibrium convergence result for zero-sum games in the spirit of classical results on fictitious play, as well as a “folk theorem” linking strategic and dynamic notions of stability – Nash equilibria and attracting points of regularized learning, respectively. We pay special attention to the information available to the players and, through a unified analysis framework, we study both oracle- and payoff-based (bandit) methods. Our goal is to provide a coherent and comprehensible – albeit, by necessity, not comprehensive – account of some recent ideas in the field, and to discuss their implications for the study of rationality.
[LG-8] Beyond Binary: Continuous State Optimization with Graph-Structured Objectives
链接: https://arxiv.org/abs/2608.09366
作者: Corinna Cortes,Yishay Mansour,Mehryar Mohri
类目: Machine Learning (cs.LG)
*备注:
Abstract:Large-scale learning systems often face the challenge of balancing multiple, potentially competing objectives, such as fairness, accuracy, and latency. While recent work has formalized this as an optimization problem over binary states, many real-world control parameters, such as fairness thresholds, diversity mixing rates, or resource budgets, are continuous. In this work, we extend the framework to \emphcontinuous state spaces. We model the problem as minimizing a sum of linear objectives subject to \emphmovement costs that penalize system instability. We capture the local structure of the objectives using a \emphdependency graph (or factor graph), where each objective is determined by a subset of the state attributes. To address the tension between exploration and stability, we propose \emphLazy Graph-LinUCB, an algorithm that performs lazy updates to minimize switching costs while maintaining near-optimal regret. Beyond stability, we introduce three advanced mechanisms to exploit the underlying graph structure: (1) an \emphasynchronous update schedule that eliminates synchronization overhead in sparse graphs; (2) an \emphadaptive algorithm that learns the graph structure from data; and (3) a \emphjoint estimator that leverages data sharing among correlated objectives to significantly tighten regret bounds. Empirically, we demonstrate that these structural exploitations reduce movement costs by more than a factor of three in heterogeneous systems while maintaining similar cumulative losses. Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.09366 [cs.LG] (or arXiv:2608.09366v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.09366 Focus to learn more arXiv-issued DOI via DataCite (pending registration) Submission history From: Mehryar Mohri [view email] [v1] Mon, 10 Aug 2026 09:46:28 UTC (153 KB) Full-text links: Access Paper: View a PDF of the paper titled Beyond Binary: Continuous State Optimization with Graph-Structured Objectives, by Corinna Cortes and 2 other authorsView PDFTeX Source view license Current browse context: cs.LG prev | next new | recent | 2026-08 Change to browse by: cs References Citations NASA ADSGoogle Scholar Semantic Scholar export BibTeX citation Loading… BibTeX formatted citation loading… Data provided by: Bookmark checked="checked"class=“labs-tab-input”> Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) IArxiv recommender toggle IArxiv Recommender (What is IArxiv?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv’s community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?) mathjaxToggle(); We gratefully acknowledge support from our major funders, member institutions, , and all contributors. About Help Contact Subscribe Copyright Privacy Accessibility Operational Status (opens in new tab) Major funding support from
[LG-9] In-Context Density Estimation for Tabular Data
链接: https://arxiv.org/abs/2608.09348
作者: Patryk Marszałek,Jacek Tabor,Marek Śmieja
类目: Machine Learning (cs.LG)
*备注:
Abstract:Density estimation underlies many unsupervised tasks on tabular data such as anomaly detection, out-of-distribution detection, and data augmentation. Although all these problems reduce to questions about where probability mass lies, they are typically solved individually by fitting a separate model to each dataset, with its own hyperparameters and tuning budget. We introduce ICED, an in-context, energy-based density estimator that removes this per-dataset cost. ICED is a transformer-based model pretrained once on a synthetic prior built specifically for density estimation under an objective that fits log-density where it is informative and preserves its ordering elsewhere. In the inference, it reads a dataset as context and returns an unnormalized log-density for any query point in a single forward pass, with no fitting, sampling, or hyperparameter selection. A single frozen ICED model then drives four tasks usually handled by four specialized pipelines: density estimation, out-of-distribution detection, unsupervised anomaly detection, and generative augmentation. Across all four, it is competitive with the strongest task-specific method, while being the only approach that needs no retraining, no tuning, and no labels to move between them. The code is available at this https URL.
[LG-10] F2STNet: Fair and Federated Spectral-Temporal Modeling for Graph Forecasting
链接: https://arxiv.org/abs/2608.09082
作者: Jiayi Zhang,Jinfeng Xu,Hewei Wang,Siyuan Cen,Haidong Huang,Yiyao Zhan,Zheyu Chen,Jinjiang You,Ai Jian,Edith C. H. Ngai
类目: Machine Learning (cs.LG)
*备注: 13 pages
Abstract:Spatiotemporal prediction on graph-structured data is central to traffic forecasting and environmental monitoring, yet decentralized and heterogeneous data complicate both sequence modeling and collaborative training. We propose F ^2 STNet, a federated forecasting framework that combines truncated graph-Fourier features, a lightweight diagonal state-space temporal encoder, graph convolution, and Fairness-aware Federated Aggregation (FFA). The spectral branch exposes graph-frequency structure, while the state-space layer models long temporal dependencies with linear complexity in the sequence length. FFA adjusts the FedAvg prior using client validation losses and an increasing fairness schedule. Experiments on PeMS04, HZMetro, and KnowAir show favorable forecasting accuracy relative to the evaluated baselines; federated experiments on PeMS04 additionally improve worst-client and client-dispersion metrics.
[LG-11] Decision-Focused Learning in Network Interdiction Games
链接: https://arxiv.org/abs/2608.09036
作者: Luca M. Hartmann,Parinaz Naghizadeh
类目: Computer Science and Game Theory (cs.GT); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注: 20 pages, 6 figures, accepted at GameSec2026 Conference on Game Theory and AI for Security
Abstract:We study decision-focused learning (DFL) in shortest-path network interdiction (SPNI) games, a Stackelberg game where an interdictor (leader) strengthens the networks’ arcs against attacks, while an evader (follower) who is uncertain about costs of attacking network arcs relies on a machine-learned predictor to identify the shortest path. While DFL is highly effective as an end-to-end optimization framework, we show that it faces a fundamental structural failure when employed in this game setting: its training objective admits a broad decision-equivalence class of cost estimators that achieve zero nominal loss yet fail under interdiction, reversing DFL’s usual advantage over a naive prediction-focused learning (PFL) approach. To address this, we propose Adversarial DFL (A-DFL), which replaces nominal training samples with interdicted scenarios to collapse the harmful equivalence class. Experiments on synthetic and real-world networks confirm that A-DFL restores DFL’s advantage in this game setting, enabling effective end-to-end optimization.
[LG-12] HOPPER: Learnable Hop Extraction for Linearized Graph Sequence Models
链接: https://arxiv.org/abs/2608.09031
作者: Isuru Herath,Arin Gopakumar,Sharan Sahu
类目: Machine Learning (cs.LG)
*备注: 24 pages, 1 figure, 4 tables
Abstract:Graph neural networks typically propagate information through repeated message-passing layers, coupling the distance over which information travels with the number of nonlinear transformations applied. This coupling can make deep architectures difficult to optimize and can lead to over-smoothing, over-squashing, and the loss of long-range information. Linearized Graph Sequence Models (LGSMs) address this issue by separating information depth from processing depth and treating the successive propagation states of each node as a sequence. However, existing LGSMs construct these sequences using fixed graph operators, limiting their ability to adapt propagation to the input graph, node features, and downstream task. We introduce HOPPER, an end-to-end learnable extension of LGSM that learns how hop sequences should be extracted before they are processed by a modern state-space model. Our framework supports feature-conditioned, structure-aware, graph- and hop-adaptive propagation mechanisms while preserving permutation equivariance. Standard adjacency-based and non-backtracking LGSM sequences arise as special cases of our proposed extractor family. We show that HOPPER is state-of-the-art or competitive across the ECHO-Synth benchmark, and that varying the maximum neighborhood size of message backtracking cancellation (i.e. structural memory window) can optimize accuracy on the LRIM physics-based long-range dependency benchmark. These results demonstrate that learnable sequence extraction provides a flexible and effective approach to long-range graph representation learning.
[LG-13] Closing the loop in learning with missing data
链接: https://arxiv.org/abs/2608.09030
作者: Dimitrios Pylorof,Humberto E. Garcia
类目: ystems and Control (eess.SY); Machine Learning (cs.LG); Optimization and Control (math.OC)
*备注:
Abstract:What should a machine learning model learn when data is missing during training? We look at the learning process from a dynamical systems perspective, cast data missingness as a structured loss of actuation that limits controllability of the parameter error dynamics, and ultimately derive adaptation mechanisms with Lyapunov stability characteristics that throttle model updates in ways that preserve learning coherence under partial, intermittent observability. Under recurrent excitation, our analysis provides ISS-type residual-to-state bounds with respect to a bounded closed-loop mismatch between the loss residual and the preconditioned update geometry. We evaluate the efficacy of our directional observability-aware adaptive learning approach on multimodal contexts, reinforcing its premise in promoting learning coherence and stability even in pathologically sparse domains and problems.
[LG-14] Distilling Vision-Language Models for Robust Traffic Sign Perception in Autonomous Vehicles IROS2026
链接: https://arxiv.org/abs/2608.08815
作者: Pedram MohajerAnsari,Amir Salarpour,Mert D. Pesé
类目: Machine Learning (cs.LG)
*备注: Accepted to the 2026 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2026)
Abstract:Traffic sign recognition (TSR) models based on deep neural networks achieve strong clean-data performance but remain vulnerable to physically realizable adversarial attacks, including shadow perturbations, natural-light interference, and printed patches. Existing defenses often improve robustness against one attack type while degrading performance on others, and can reduce clean accuracy. We propose LAMDA (Language-Anchored Model for Direction Alignment), a training framework that transfers language-grounded structure into TSR models without using adversarial examples or adding inference-time overhead. LAMDA builds two fixed prototype banks from VLM-generated sign descriptions and class names using a frozen OpenCLIP text encoder, and uses them to supervise visual features through two complementary auxiliary losses during training. At inference, the adapter and prototype banks are discarded, leaving a standard backbone and classifier. Evaluated on GTSRB and LISA across four backbones and three physical attack types, LAMDA is the only method among ten evaluated that consistently improves robustness across all attack-backbone-dataset combinations, with gains of up to +12.5 pp under shadow attacks and +13.2 pp under natural-light attacks, while preserving or improving clean accuracy in nearly all cases.
[LG-15] ML-Based Hierarchical Prediction for Practical Energy Scheduling in Dynamic NTN-WPT Systems
链接: https://arxiv.org/abs/2608.08804
作者: Zhanyu Ju,Wenchi Cheng
类目: ystems and Control (eess.SY); Machine Learning (cs.LG)
*备注: 21 pages, 11 figures
Abstract:With advancements in long-distance wireless power transfer (WPT) and space-based energy technologies, integrating WPT into non-terrestrial networks (NTNs), referred to as NTN-WPT, is emerging as a promising approach for next-generation wireless networks. This paper proposes an energy-scheduling approach that jointly optimizes energy efficiency, task completion rate, and task waiting time for power transfer from low Earth orbit satellites to terrestrial mobile user devices (UDs). To address scheduling challenges caused by satellite and UD mobility and channel uncertainty from stochastic propagation effects, we decompose the problem into three subproblems within a three-layer predictive framework: 1) a state prediction layer forecasts UD and satellite states; 2) an interaction mapping layer uses a graph neural network (GNN) to model energy transfer efficiency; and 3) a decision-making layer determines the energy allocation plan. Distinct machine learning (ML) methods are tailored to each layer. To balance the competing objectives, we adopt a multi-objective reinforcement learning (MORL) technique that scalarizes them into a weighted-sum reward, transforming the multi-objective problem into a tractable single-objective problem. We further introduce a multi-agent deep learning model integrating self-attention with multi-agent proximal policy optimization (MAPPO) to improve objective balancing. Simulation results show that the proposed approach achieves a better overall trade-off than baseline methods, maintaining competitive task completion rates and energy efficiency while reducing task waiting times, and remains robust under highly variable conditions.
[LG-16] Quantum-Classical Physics-Informed Kolmogorov-Arnold Networks for Solving Fuzzy Differential Equations
链接: https://arxiv.org/abs/2608.08782
作者: Xiang Rao,Yuxuan Shen
类目: Machine Learning (cs.LG)
*备注:
Abstract:In this study, we propose a quantum-classical physics-informed Kolmogorov-Arnold network (QCPIKAN) dedicated to the solution of fuzzy differential equations. The network takes the spatiotemporal coordinates and membership level as joint inputs and employs ChebyKAN modules and a parameterized quantum circuit to construct a hybrid function approximator. It simultaneously approximates the lower and upper endpoint functions associated with the \alpha-cuts and incorporates the governing equations, initial-boundary conditions, and fuzzy-structural constraints into the training objective. Theoretically, a unified error-analysis framework is established for QCPIKAN and PIKAN, in which the endpoint-solution error is decomposed into approximation, sampling, optimization, and fuzzy-structure constraint errors. Under the assumptions of well-posedness and residual stability, it is proved that QCPIKAN has a smaller a priori error bound when the representational gain introduced by quantum entanglement features exceeds the additional computational error. Numerical experiments are conducted for elliptic, parabolic, and hyperbolic equations in an ideal quantum-simulation environment. The results show that QCPIKAN captures the overall contraction of the solution interval as increases. At most tested membership levels, the mean relative L2 error of PIKAN is approximately 1.1-2.7 times that of QCPIKAN. In the fuzzy convection example, the mean wavefront-position error of PIKAN is approximately 1.77 times that of QCPIKAN. Nevertheless, both models still exhibit local fuzzy-structure violations near boundaries, in high-gradient regions, and around the wavefront. These results indicate that QCPIKAN provides a quantum-classical hybrid physics-informed computational framework with comparatively high predictive accuracy for solving fuzzy partial differential equations represented by \alpha-cuts.
[LG-17] Memory-Efficient Activation Checkpointing with Sliding Window and Hirschbergs Algorithm for 0/1 Knapsack Solving in PyTorch
链接: https://arxiv.org/abs/2608.08740
作者: Jędrzej Maczan
类目: Machine Learning (cs.LG)
*备注: Accepted to COLM 2026 Workshop on Efficient Reasoning
Abstract:Activation checkpointing minimizes the runtime of neural networks under a given memory budget, by selecting which intermediate tensors to store and which to recompute. PyTorch solves this as a 0/1 knapsack problem, where operations from a joint forward-backward computation graph are items with a memory cost (weight) and a runtime saving (value). The default solver, dp_knapsack, allocates a full dynamic programming (DP) table of shape (n+1) \times (W+1) , where n is the number of operations and W is the quantized memory budget. This method is resource-hungry and crashes at n = 100 items on a machine with 64 GB RAM. In this paper, we introduce dp_knapsack_sliding_hirschberg, which combines the sliding window trick and Hirschberg’s algorithm to reduce peak memory from O(nW) to O(W) while preserving the exact optimal solution. Our experiments show successful knapsack execution at n = 2000 , where dp_knapsack fails at n = 100 , a 20 \times increase in computable problem size. In addition, our benchmarks show a consistent 25-28% runtime speedup over dp_knapsack. The implementation is merged into PyTorch and released in version 2.10. Comments: Accepted to COLM 2026 Workshop on Efficient Reasoning Subjects: Machine Learning (cs.LG) Cite as: arXiv:2608.08740 [cs.LG] (or arXiv:2608.08740v1 [cs.LG] for this version) https://doi.org/10.48550/arXiv.2608.08740 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
[LG-18] Measuring and Reducing WebGPU Dispatch Overhead for LLM Inference
链接: https://arxiv.org/abs/2608.08730
作者: Jędrzej Maczan
类目: Machine Learning (cs.LG); Distributed, Parallel, and Cluster Computing (cs.DC); Performance (cs.PF)
*备注: Accepted to The Fourth UK AI Conference 2026 as a full paper, to be published in Proceedings of Machine Learning Research (PMLR)
Abstract:Large Language Models are deployed to multiple types of environments, from internet browsers to edge devices, and WebGPU serves as a modern cross-platform standard. The engines for browser-based LLM inference have proliferated, yet the overhead of WebGPU per-operation dispatch remains poorly characterized. In this work, we introduce a sequential-dispatch measurement method and show that naive single-operation measurements overestimate per-dispatch cost by conflating dispatch with synchronization. Using our method, we measure the per-dispatch cost and show that it is independent of data type used. We show that the dispatch overhead, not kernel quality, is the bottleneck at batch size 1, and isolate the dispatch count as the cause. Therefore, we conclude that at batch size 1, the effective approach to LLM inference optimization in WebGPU is reducing dispatch count. Our findings point to dispatch amortization, in the inference engines and in the WebGPU specification, as a path to practical browser-based inference.
附件下载
