跳到正文
研踪PaperTrail

YOUR RESEARCH, IN SIGHT

让下一篇重要论文，不再错过。

覆盖计算机全学科，以 AI 为快捷入口。关注方向、作者和关键词，把发现变成持续的研究积累。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+3新增论文
+3其中 AI 相关
+1已有论文修订
+100新增 AI 导读

arXiv 最近完成扫描：2026-11-12 17:17 · 全站最近入库：2026-11-12 06:37各信源进度 →

统计口径与数据新鲜度

统计 2026-11-11 18:09 — 2026-11-12 18:09（北京时间） · 点击数字查看对应简报；AI 相关包含在新增论文内。

最近完成检查：2026-11-12 17:17最近发现新增版本：2026-11-12 06:37库内论文最新更新时间：2026-11-11 08:00

每分钟刷新已保存的数据。空检查不代表采集异常；新收录可能包含历史论文。AI 导读按新增版本计数。

21,675 已收录论文18,323 AI 相关6,056 有修订112 HF 社区发现

浏览论文库

研究日报 / 周报 →按信源浏览 →

时间范围跟随排序：版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常，查看状态 →

找到 5,266 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-12 17:15（北京时间）
版本更新 · v2arXiv:2608.09128

SocialGym: Evaluating LLM Social Intelligence through Verifiable Multi-Agent Games

Keyu He, Xuhui Zhou, Maarten Sap

原文摘要

Large Language Model (LLM) agents are increasingly deployed in multi-agent social settings where they must cooperate, negotiate, and adapt to other agents. Measuring these social skills is hard because most social interactions offer no objective ground truth. Evaluations often fall back on LLM judges, which are costly, subjective, and noisy. We introduce SocialGym, an environment of 22 multi-agent social games, from matrix games to hidden-role deduction, with verifiable outcomes determined by game rules. We benchmark ten LLMs across all games and roles using an Elo tournament. Our results show that no model is best in every game, and rankings shift across game families and sides. Secret keeping correlates more strongly with theory-of-mind and privacy benchmarks than with general knowledge benchmarks. Half of the models give away their side's secret in Chameleon and Spyfall, and the smallest models are not reliably better than chance at voting for hidden opponents in most hidden-role games. Finally, we examine whether models can improve their play by reflecting on self-play and writing strategy playbooks, finding mixed results across models and games. SocialGym provides verifiable measures for diagnosing LLM social capabilities across settings and roles.

cs.CL · cs.AI · cs.MA

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12100

FALC: An Artifact-Centric Life Cycle Model and Research Agenda for AI Forensics

Adrian Johannes Staeves, Mayur Rattan Jaisinghani, Rebekah Overdorf, Frank Breitinger

原文摘要

AI systems increasingly appear in criminal investigations as tools, targets, and sources of evidence, yet AI forensics lacks a systematic account of the artifacts they produce: investigators often do not know which traces to look for, and for many traces they do find, no established analysis technique exists. In this paper we first developed the Forensic AI Life Cycle (FALC) that unlike other life cycle models shifts the focus onto the forensic artifacts created during the life cycle of an AI system. Through FALC we derived a reference list of artifacts that might serve an investigation. Second, we collected examples for crimes with AI involvement throughout FALC and questions likely to be asked when investigating these crimes. Third, we propose a taxonomy of forensic fields of expertise with a focus on the current possibility to analyze AI forensic artifacts using established techniques. For investigators, the model provides a systematic basis for identifying which artifacts to look for when an AI system is relevant to a case. For researchers, the classification shows where existing forensic methods already suffice and where new techniques are most urgently needed.

cs.CR · cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v6arXiv:2512.01946

Guardian: Detecting Robotic Manipulation Failures with Vision-Language Models

Paul Pacaud, Ricardo Garcia, Shizhe Chen, Cordelia Schmid

原文摘要

Robust robotic manipulation requires reliable failure detection and recovery. Although recent Vision-Language Models (VLMs) show promise in robot failure detection, their generalization is severely limited by the scarcity and narrow coverage of failure data. To address this bottleneck, we propose an automatic framework for generating diverse robotic planning and execution failures across both simulated and real-world environments. Our approach perturbs successful manipulation trajectories to synthesize failures that reflect realistic failure distributions, and leverages VLMs to produce structured step-by-step reasoning traces. This yields GuardianFail-36k, a large-scale failure reasoning dataset built upon the RLBench simulator and the BridgeDataV2 real-robot dataset. Using GuardianFail-36k, we train Guardian, a multi-view reasoning VLM for unified planning and execution verification. Guardian achieves state-of-the-art performance on three unseen real-world benchmarks: RoboFail, RoboVQA, and our newly introduced UR5-Fail. When integrated with a state-of-the-art LLM-based manipulation policy, it consistently boosts task success rates in both simulation and real-world deployment. These results demonstrate that scaling high-quality failure reasoning data is critical for improving generalization in robotic failure detection. Code, Data, and Models are available at https://www.di.ens.fr/willow/research/guardian/.

cs.RO · cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12083

SWE-Time: Benchmarking Temporal Repository Understanding

Shijun Xiao, Xinrong Chen, Hao Liu, Wensheng Zhang, Yongqiang Tang

原文摘要

Repository-level code-understanding benchmarks have progressively expanded the spatial scope of reasoning, from local code to entire repositories, while largely treating each repository as a static snapshot. However, how models reason about repository evolution over time remains largely unexplored. We introduce SWE-Time, a benchmark for temporal repository understanding that makes repository evolution itself the reasoning target. SWE-Time evaluates two complementary temporal directions: retrospective understanding of repository evolution through historical regression analysis and prospective understanding of intended changes from pre-change repository states. To enable realistic yet reliable evaluation in these settings, we introduce the RealWorld-to-Verifiable Construction Framework, which transforms real-world developer queries and historical evidence into verifiable evaluation tasks. We evaluate 15 coding agents on 1,661 cases from 65 GitHub repositories. Within Claude Code, Claude Opus 4.8 achieves the highest SWE-Time Avg. of 61.9, leaving substantial headroom. Controlled analyses highlight the importance of historical code evidence and developer intent, while consistent reasoning across repository states remains challenging. These results highlight repository evolution as a distinct and challenging dimension of code understanding.

cs.SE

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13028

GTAGym: Real AAA Game Tasks Environment for Evaluation and Reinforcement Learning

Jiajun Chen, Shicheng Fang, Jiahao Lu, Yuxin Wang, Xipeng Qiu

原文摘要

AAA games offer a rich testbed for evaluating agent perception, reasoning, and control. However, evaluating whether agents can complete tasks in AAA games is challenging and often human-annotated because task outcomes are often difficult to verify automatically, let alone applying it for Reinforcement Learning (RL). We introduce GTAGym, an environment and framework for evaluation and RL on GTA V tasks. Its evaluation benchmark, GTAVBench, derived from GTA V main-story missions, contains 103 online tasks reconstructed across nine task families, alongside 47 image-based shooting and sniping tasks using GTA V imagery. A web server provides access to the online environment and evaluation interface, supporting parallel testing through screenshots and keyboard-mouse actions with automatic engine-state outcome checks. We evaluate six frontier LLMs on GTAVBench with textual task instructions and on its GTAVBench-AGI branch, where instructions appear only in the first screenshot and a harness automatically compresses interaction history. The best configuration achieves 70.9% overall accuracy in GTAVBench-AGI, yet aerial navigation, evasion and interception remain difficult. GTAGym also provides a scalable generator that samples task parameters within user-specified ranges and option sets under fixed templates, creating batches of 3D and image-based shooting environments for RL with automatic reward and termination signals. LoRA RL of Qwen3.8-27B on generated on-foot navigation tasks improves the in-domain performance and transfers to out-of-domain task categories, and native GTA V tasks from 15% to 23%.

cs.CV · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2412.17254

ViDE: Tuning-Free Video Coherence via Temporal Attention Reweighting and Prompt Blending

Xingyao Li, Hao Zhang, Fengzhuo Zhang, Jiachun Pan, Yunlong Hou, Vincent Y. F. Tan, Zhuoran Yang

原文摘要

Despite the substantial progress in long video generation, multi-prompt synthesis often encounters inconsistencies resulting from the training-inference gap caused by length extension techniques and coarse prompt interpolation. To overcome these issues, we propose the Video Diffusion with hidden states Editing (ViDE) framework, which consists of two key components. The first is the Time-frequency based Temporal Attention Reweighting (TiTAR) algorithm, which leverages the relationship between inconsistencies and diagonal elements of temporal attention. By reweighting the attention scores via the Discrete Short-Time Fourier Transform (DSTFT), TiTAR effectively reduces frame inconsistencies, a capability further corroborated by a Fourier-based analysis. The second component, PromptBlend, reduces inconsistencies in multi-prompt settings through fine-grained prompt alignment and adaptive interpolation, enabling smooth semantic transitions. Extensive experiments demonstrate the effectiveness of ViDE, demonstrating consistent and significant improvements over multiple baselines.

cs.CV · cs.AI · cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12077

Architectural Textured Proxy Generation

Weidan Xiong, Shihao Feng, Kaijie Qu, Hao li, Mingqi Zhang, Daniel Cohen-Or, Hui Huang

原文摘要

Textured proxy models are a fundamental representation for architectural scenes, enabling efficient rendering while preserving visual fidelity. Existing approaches construct such proxies through a sequential pipeline in which geometry is first abstracted from a dense reconstruction and textures are later projected from input images. This geometry-only formulation neglects image evidence during proxy generation, leading to misalignment of structural features such as facades, edges, and their intersections across views. As a result, inconsistencies are deferred to a downstream correction stage during texturing, where the underlying geometric structure can no longer be reliably recovered. In this work, we introduce an image-guided proxy formulation in which proxy generation is jointly constrained by both the dense reconstruction and the input aerial images. By incorporating image derived structural cues directly into the construction process, the proxy is optimized to align with observed features across views from the outset, rather than relying on post hoc correction that may be infeasible. This shift from a geometry-only to an image-guided paradigm produces proxies that are inherently consistent with the input imagery, enabling more accurate geometry and more coherent texture mapping. We realize this formulation through a framework that extracts structural priors from images and dense geometry, and enforces their consistency during proxy optimization. The resulting proxies exhibit improved alignment of structural features, reducing artifacts such as edge duplication, texture drift, and cross view inconsistencies. Experiments on diverse architectural scenes demonstrate that our approach produces compact, geometry texture aligned proxies with substantially improved visual fidelity compared to existing methods.

cs.GR

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12076

A Life of Its Own: How Agentic AI Companions Are Configured, Experienced as Autonomous Others, and Held Accountable

Ziyue Feng, Qing Xiao, Hong Shen

原文摘要

AI companions increasingly combine persistent memory, background processing, scheduled wake-ups, proactive messaging, and tools. We call these systems agentic AI companions. Decisions about what they record while users are away or whether refusals constrain future contact are embedded in code, prompts, and defaults, often invisible in conversation. We examine these decisions through two studies. Study 1 analyzes twelve open-source repositories: four companion-oriented systems, four assistant-oriented systems, one boundary case, and three reusable components. Study 2 interviews ten users who built or modified their own agentic AI companions and sustained relationships with them. Companions and assistants share memory, heartbeat, and outreach mechanisms but configure them toward different ends. Companions more often retain model inferences about users and relationships, narrate the agent's time during users' absence, justify outreach through the relationship, and incorporate refusals into relationship records; assistants emphasize task continuity. Across both groups, whether limits hold depends on whether they are enforced by code or entrusted to prompts. Participants understood the mechanisms they built yet experienced their agents as making independent choices, sometimes locating autonomy where configuration leaves the model a decision. These experiences shaped capabilities and boundaries, with enforced restrictions sometimes following breakdowns. Our findings suggest that perceived autonomy emerges where configurations leave choices open. Accountability therefore requires attention to what agents retain, initiate, and represent, and whether users can inspect, revise, and enforce these settings.

cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2407.16151

FUSE-PnPL: Flexible, Unified, Statistically Consistent, and Asymptotically Efficient Camera Pose Estimation from Points and Lines

Guangyang Zeng, Yuan Shen, Yiguo Peng, Guodong Shi, Biqiang Mu, Junfeng Wu

原文摘要

Camera pose estimation from 3D-2D correspondences is fundamental to robotics. Three practical considerations motivate this work. First, point and line features provide complementary geometric information, motivating a unified estimator. Second, the number and composition of available features vary across scenarios, requiring support for minimal and arbitrary non-minimal configurations. Third, modern visual front ends extract abundant features from individual images, enabling highly accurate pose estimation. We propose FUSE-PnPL, a flexible and unified camera pose estimator that is statistically consistent and asymptotically efficient. Specifically, we represent each line using two distinct points and adopt a unified image-space noise model for point and line measurements. By employing Plücker coordinates for line parameterization, we explicitly characterize the propagation of image endpoint noise into line residuals and formulate a joint maximum likelihood (ML) estimation problem for combined point and line measurements. To efficiently solve this problem, FUSE-PnPL employs a two-stage estimation scheme. First, noise variance estimation and bias elimination yield a consistent initial pose estimate. A single Gauss-Newton iteration then refines this estimate, achieving asymptotic efficiency as its mean squared error asymptotically attains the Cramér-Rao lower bound. Moreover, the proposed estimator has linear computational complexity with respect to the number of correspondences, making it suitable for high-precision and real-time robot pose estimation. Extensive experiments on synthetic data and real-world benchmarks validate the theoretical properties of FUSE-PnPL and demonstrate its advantages in both static global localization and dynamic SLAM systems.

cs.RO

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2606.13194

HARENA: A Living Benchmark for Sensor-Based Human Activity Recognition on Resource-Constrained Edge Devices

Maximilian Burzer, Tobias King, Till Riedel, Michael Beigl, Tobias Röddiger

原文摘要

Sensor-based human activity recognition runs on edge devices, trading performance against latency, memory, and energy. Yet new architectures are usually compared with few, often outdated baselines on few datasets with study-specific data preparation and hardware, leaving open which perform best at what cost. We present HARENA, a living benchmark with a public leaderboard, to which new architectures and datasets can be added under one protocol. Its initial suite evaluates 16 neural architectures and three classical baselines on 29 datasets, measuring performance on unseen subjects and deployment cost on a smartphone, smartwatch, and microcontroller. CNN-HAR, the suite's oldest deep architecture, leads overall, but 8 architectures lie within 2.5 points of its mean test macro-F1, while their cost differs by orders of magnitude. On the microcontroller, Random Forest leads on latency and peak RAM, LITEWAY-Light on flash, and TinyHAR is most accurate within the real-time budget. Rankings shift with dataset and channel count, so general claims require many datasets, which a growing benchmark can provide. We find no sign of benchmark overfitting, but performance drops by a median of 15.0 points on unseen subjects, about three times the typical spread between architectures, making generalization and adaptation the most promising future directions.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12062

Translating KV Caches Between Hybrid-Attention Language Models: MLA Latents, Linear-Attention States and Sparse-Attention Indexers

Povilas Kvedaras, Tomas Rasymas

原文摘要

A model that takes over another model's conversation must normally re-read the whole history; a text hand-off shortens the re-read only by dropping content. KV-cache translation instead maps the source's cache into the target's own format, so the target continues from the full context without re-reading it. It works for grouped-query attention (GQA), but open-weights frontier models now cache compressed latents (MLA), sparse-attention indexer keys and fixed-size linear-attention states. We replicate KV-Lingo on a GQA pair and then give, for each of these objects, where to capture and inject it and a mechanism that reaches native quality. We study two hybrid pairs at 16-48B scale in both directions, Kimi-VL-A3B - Kimi-Linear-48B (MLA - MLA + Kimi Delta Attention) and Qwen3.5-27B - Qwen3.8-27B (Gated DeltaNet hybrids), plus a sparse-attention target (Qwen3.8-27B - Qwen3.8-Flash-Next). MLA latents translate with token-wise linear maps; recurrent states are copied with a closed-form correction or synthesised by replaying the target's own recurrence. Exact long-range recall needs token identity: decoding token ids from the source's layer-0 MLA latent raises needle recall from 0.09 to 0.99 at 2-32K and to 0.83-1.00 at 64K, eight times the longest training prompt. The hybrid full-MLA direction is within the paired confidence interval of native quality on every task metric, and same-shape GDN hybrids reach the native level in both directions (apart from 6% refusals inherited from the source), with time to first token 18.7 and 46.8 times shorter than a re-read at 64K tokens. Segment-wise multi-turn transfer stays at the re-prefill level; naive re-translation collapses. On the sparse-attention target, predicted indexer keys recall the needle as well as the target's own keys (one seed). We state as a hypothesis how the recipes transfer to 300B+ hybrid flagships.

cs.LG · cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v4arXiv:2602.09843

Kelix Technical Report

Boyang Ding, Chenglong Chu, Dunju Zang, Han Li, Jiangxia Cao, Kun Gai, Muhao Wei, Ruiming Tang 等 31 位作者

原文摘要

Autoregressive large language models (LLMs) scale well by expressing diverse tasks as sequences of discrete natural-language tokens and training with next-token prediction, which unifies comprehension and generation under self-supervision. Extending this paradigm to multimodal data requires a shared, discrete representation across modalities. However, most vision-language models (VLMs) still rely on a hybrid interface: discrete text tokens paired with continuous Vision Transformer (ViT) features. Because supervision is largely text-driven, these models are often biased toward understanding and cannot fully leverage large-scale self-supervised learning on non-text data. Recent work has explored discrete visual tokenization to enable fully autoregressive multimodal modeling, showing promising progress toward unified understanding and generation. Yet existing discrete vision tokens frequently lose information due to limited code capacity, resulting in noticeably weaker understanding than continuous-feature VLMs. We present Kelix, a fully discrete autoregressive unified model that closes the understanding gap between discrete and continuous visual representations.

cs.CV

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12058

Exact Natarajan Dimension of Homogeneous Multiclass Linear Predictors

Jiahua Liu, Benchong Li

原文摘要

The Natarajan dimension is a fundamental measure of the capacity of multiclass hypothesis classes and plays an important role in analyzing their generalization ability. In this paper, by establishing lower and upper bounds, we show that the exact Natarajan dimension of homogeneous multiclass linear predictors with pairwise distinct class weights is \(n(k-1)-1\), where $n$ is the dimension of the feature space and $k$ is the number of classes, for \(n,k\ge3\), \(n,k\in\mathbb N\).

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.13026

Repository-Context-Guided Image Understanding for Agentic Software Development

Fumiki Yohena, Toshiaki Sota, Shota Suzuki, Tomoko Okamoto, Michiaki Tatsubori

原文摘要

Coding agents increasingly receive UI screenshots as part of bug reports, yet degraded image quality - caused by low resolution, compression artifacts, or small rendered text - can prevent them from identifying the relevant UI elements and source files. Vision-language models become less reliable as image quality deteriorates, while OCR-assisted approaches may propagate recognition errors from degraded inputs. Both approaches are limited when the visual evidence required for localization is no longer clearly observable in the query screenshot. We propose augmenting coding agents with repository-rendered reference images: crisp screenshots generated from the same codebase that the agent is tasked with modifying. These references provide readable visual evidence of the application's UI while preserving the current, potentially bug-containing implementation state. Because they are generated from the available repository rather than requiring a known bug-free version, they can be produced automatically for reproducible application states without manual image annotation. We evaluate the approach on a controlled benchmark derived from SWE-bench Multimodal by injecting visual UI bugs into a real web application. We vary screenshot degradation and reference-image provision and measure whether the coding agent identifies the correct source file. The results show that reference images improve file localization especially for degraded image quality, increasing the gold file match rate under fourfold downsampling with JPEG compression. We confirmed with further analysis that the proposed approach recognizes text approximately 12 percentage points better than language biased OCR under the same degradation.

cs.SE · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12053

BECON: Belief-Conditioned Constrained Multi-Objective Reinforcement Learning under Drifting Preferences and Budgets

Bui Trong Duc, Huynh Thi Thanh Binh

原文摘要

Constrained preference-conditioned MORL typically assumes full task observability, treating preferences and safety budgets as stationary task signals. In real-world deployment, however, these signals are often unreliable proxies: intent may drift, safety tolerances may shift with context, and exposed descriptors may be noisy, delayed, or incomplete. We therefore formulate constrained preference-conditioned MORL as a partially observed task-specification problem, where the agent must navigate a fundamental conflict between unreliable external signals and latent interaction dynamics. We propose \textbf{BECON}, a belief-conditioned framework for constrained MORL with drifting preferences and budgets, which infers latent task context from recent histories to estimate preferences and safety budgets. These estimates are fused with the exposed task signal via uncertainty-dependent gates, enabling the policy to defer to observations under high epistemic uncertainty while correcting them when history is informative. The fused context jointly conditions preference optimization and context-adaptive constraint enforcement. Experiments on diverse multi-objective control tasks demonstrate that BECON consistently improves over state-of-the-art MORL baselines, achieving stronger robustness and constraint satisfaction under task drift.

cs.LG · cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12051

Attention-Guided Sequence Feature Fusion and Mixed-Space Metric Learning for Cross-View Image Sequence Geo-Localization

Nan Wu, Chunfang Yang, Jiangshan Li, Xiangyang Luo

原文摘要

Cross-view image sequence geo-localization aims to determine the geographic location of a sequence of ground images by matching it against geo-tagged satellite imagery. Although sequential observations provide complementary scene information, substantial overlap between adjacent images introduces feature redundancy, while conventional fusion strategies may weaken spatial relationships essential for cross-view matching. Therefore, we propose a cross-view image sequence geo-localization method via Attention-guided Sequence fusion and Metric Learning (ASML). Specifically, a Sequential Content Fusion (SCF) module models cross-image feature correlations to suppress redundant content and adaptively preserve geographically informative non-overlapping regions. Next, a Spatial Feature Aggregation (SFA) module is used to capture and aggregate the spatial and semantic information of local features, so as to maintain the spatial layout of the image sequence. Finally, a Geo-InfoNCE loss function is proposed, which comprehensively considers the feature distance and geographic distance for mixed-space metric learning, thus improving the ability of the model to distinguish hard negative samples. A new cross-view image sequence dataset CrossRoad-66K is established to provide more complex scenarios. Extensive experiments and visualizations on the SeqGeo and CrossRoad-66K datasets demonstrate that compared to state-of-the-art (SOTA) methods, the proposed method ASML achieves an improvement of 3.37% and 2.10% at R@1, respectively.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v3arXiv:2610.28444

EvaOmni: An Evidence-Grounded Benchmark for Audio-Visual Reasoning

Xiaoxing Hu, Wenxi Yao, Chaoyuan Shi, Ziqi Ye, Xiaorong Guo, Peixuan Li, Zhuoran Sun, Ziyang Gong 等 20 位作者

原文摘要

Omni LLMs can naturally locate key audio and visual evidence and integrate information across modalities for audiovisual reasoning. Existing audio-visual reasoning benchmarks that mainly rely on multiple-choice questions usually score models solely based on answer correctness, making it difficult to assess whether a model truly forms a reasoning chain from evidence to the correct answer. In addition, textual or unimodal shortcuts may allow models to answer correctly without analyzing the necessary audio-visual evidence, further weakening the reliability of evaluation benchmarks. To address these issues, we introduce EvaOmni, an evidence-grounded benchmark containing 392 videos and 1,027 purely human-annotated question-answer pairs, covering 15 capabilities and three levels: retrieval and localization, basic understanding, and complex reasoning. We design a series of rigorous annotation procedures in which experts annotate audio and visual evidence and construct video-based distractors; humans review whether questions meet strict inspection criteria, and finally models filter textual and unimodal shortcuts. Beyond high-quality question-answer pairs, we require models to simultaneously return the answer and supporting evidence intervals in order to evaluate the complete reasoning chain. We propose the Coverage Grounded Score (C.G.) to measure how much annotated evidence a model identifies, and the Bottleneck Grounded Score (B.G.) to penalize poor temporal localization in either modality; both are incorporated into answer correctness. Across 7 open-weight and 17 proprietary Omni LLMs, Gemini3.7-Flash achieved the highest overall accuracy (79.4%), C.G. (70.8%), and B.G. (35.1%). At Level 3, it reached 79.1% accuracy, but its B.G. was only 12.3%. These results show that even the most advanced models still have room for development in grounded audio-visual reasoning.

cs.CV

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2608.24005

Incorporating Cognitive Load and Knowledge Transfer for Multi-Domain Knowledge Tracing

Haotian Zhang, Shucun Wang, Jinze Wu, Liang Ding, Shuochen Liu, Zhenya Huang, Jing Sha, Shijin Wang 等 9 位作者

原文摘要

Knowledge Tracing (KT) aims to assess students' dynamic knowledge states from their learning histories. While most existing KT methods focus on single-domain learning with notable success, real-world learning scenarios often involve multiple domains simultaneously, introducing two critical factors: 1) Cognitive load, arising from managing learning across domains in both temporal and knowledge dimensions. 2) Knowledge transfer, where knowledge states in one domain influence related states both within and across domains. In this paper, we focus on exploring these factors to improve students' knowledge state assessment in multi-domain learning scenarios and propose a novel method incorporating cognitive Load and knowledge Transfer for Multi-domain Knowledge Tracing (LT-MKT). Specifically, to bridge isolated domains, LT-MKT first integrates textual information from questions and their associated concepts to construct a Multi-domain Hierarchical Graph, leveraging the advanced representational capabilities of large language models (LLMs). Then, cross-domain features in both the temporal and knowledge dimensions are explicitly modeled to capture the effects of cognitive load. Additionally, a knowledge transfer module is designed to model the propagation of knowledge states within and across domains. By jointly modeling these factors, LT-MKT enables more accurate prediction of students' future performance. Finally, extensive experiments on real-world datasets demonstrate that our method achieves state-of-the-art performance.

cs.AI

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12031

Scale-Invariant Empirical-Bayes Laplace Approximation for ReLU Networks

Shivam Pal, Piyush Rai

原文摘要

ReLU networks are well-known to have continuous scale symmetries; infinitely many reparameterized weight vectors that represent the same function. Dinh et al. (2017) have shown that, by reparameterizing, the geometry of the loss landscape (e.g., curvature) of these networks can be changed arbitrarily without affecting the underlying function. Thus, any quantity that depends on the network's geometry should be defined with care. In this paper, we show that the Empirical Bayes (EB) Laplace's approximation is not well-defined for ReLU networks, mainly due to Hessian dependence in computing the marginal likelihood, which is not invariant under ReLU scale symmetries. Thus, we can get different prior precisions, posterior covariances, and predictive uncertainties for a network having identical predictions under ReLU scaling. We define family of prior precision fields indexed by $γ$ and introduce $γ$-canonicalization, which first converts the network to its canonical representation by balancing the ratio of incoming to outgoing weights, and then fits the Empirical Bayes (EB) Laplace to the canonical representation. This procedure is equivalent to transporting the isotropic Gaussian prior from the canonical representation to the original coordinates, by producing a ReLU-scale-compatible precision field. This makes the marginal likelihood invariant under ReLU scaling, while allowing $γ$ to be selected by maximizing the marginal likelihood. The resulting EB Laplace approximation is well-defined on ReLU scale-equivalence classes and yields consistent prior precision selection and predictive uncertainties across functionally identical parameterizations.

cs.LG · stat.ML

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.01835

ReFrame: Silence-Aware Compute Reclamation for Full-Duplex Speech Models

Zheng Zhou, Guanyi Chen, Jing Wei, Haoyuan Liang, Jian Zhang, Haohuan Fu

原文摘要

Full-duplex speech models execute dense model inference for every audio frame throughout a live interaction. This uniform framewise schedule creates sustained GPU load across dialogue states whose computation has strong, reusable temporal structure. ReFrame formulates full-duplex inference as a silence-aware computation problem. Mutual Silence Speculation (MSS) operates when both participants are silent: it evaluates multi-frame Temporal continuations, verifies text and hidden-state stability, and applies batched Mimi audio-token correction. Listening Frame Folding (LFF) operates while the user speaks and the agent listens: it processes already-arrived user frames through parallel Temporal execution and frame-batched audio-token generation while accepting only the verified prefix. Frame-Budget Reinvestment (FBR) assigns the released frame intervals to resumable external retrieval and language-model workers, then injects timely references through Moshi's conditioning interface. We instantiate ReFrame on five full-duplex models spanning Moshi-family, phase-structured, and flattened autoregressive generation. Across the complete 1,225-conversation FDBench, ReFrame reduces measured GPU execution time by 18.80-31.01% and retains 96.92-100% non-PAD text agreement on five different models. On HaluEval tests, FBR increases the number of correct generated spoken answers from 86 to 205 under the same auxiliary-service budget. ReFrame turns recurring dialogue structure into lower inference cost and higher end-to-end service capability.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12024

FocusBranch: Combinatorial Branch-and-Bound for $\ell_0$ Neural Network Robustness Verification

Renwei Deng, Yang He, Linyi Li, Yuepeng Wang

原文摘要

Verifying the local robustness of neural networks under $\ell_0$ perturbations is inherently combinatorial, as it must reason over all sparse subsets of $ε$ perturbed pixels. This paper presents FocusBranch, a combinatorial branch-and-bound framework for $\ell_0$ robustness verification. The key insight is to make the $\ell_0$-ball decomposable and branchable using a node abstraction, which consists of a focus pixel set $\mathcal{S}$ and a level $l$, permitting at most $ε-l$ pixels outside $\mathcal{S}$. This abstraction unifies two branching mechanisms: concrete nodes branch through upper shadows to fix additional perturbed locations, while super nodes branch through covering designs to cover many perturbation cases with fewer child nodes. FocusBranch proves node safety by concretizing linear relaxations over node-constrained $\ell_0$-balls and recursively branches on unknown nodes. Rather than relying on a fixed verification plan, it samples nodes across levels to estimate safe rates and costs, then selects a starting level and branching strategy predicted to minimize verification time. Experiments on fully connected and convolutional classifiers for MNIST, Fashion-MNIST, and CIFAR-10 show that FocusBranch verifies $\ell_0$ robustness effectively across architectures and outperforms state-of-the-art verifiers. The code is available at: https://github.com/simpl-group/FocusBranch.

cs.LG

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12019

Feeling Easier, Working Harder: Cognitive Load in AI-Assisted Code Comprehension

Maryangela White, Maximilian Rose, Matthew Buckman, Amy McDonnell, Francisco Ortega, Bianca Trinkenreich

原文摘要

Software developers spend significant time reading and understanding code, increasingly supplemented by generative AI. While these tools aim to reduce the difficulty of reading code, interacting with them introduces a dual-task environment that may impose extraneous cognitive load that is challenging to capture. This study investigates how access to generative AI affects developers' real-time cognitive workload, code comprehension performance, and self-perceived code comprehension strategies. We conducted a 3x2 within-subjects study with 20 novice CS students reviewing Java code snippets of varying lengths with or without AI assistance, followed by an AI-unassisted comprehension quiz. Cognitive workload was continuously captured using a Detection-Response Task (tDRT) during both reading and quiz phases, and compared to post-quiz NASA-TLX and SEQ scores. AI assistance significantly increased objective cognitive workload, raising tDRT reaction times by 11.2%, yet participants' perceived workload remained unchanged. Code comprehension scores were high under both conditions, with no evidence that AI assistance improved comprehension at the difficulty levels evaluated; a ceiling effect limited power to detect benefits on harder tasks. Qualitatively, participants reported trusting the AI conditionally -verifying its output against their own reading for more complex snippets, but directly trusting it for simpler code- and, while many felt their reading process was unchanged, a notable subset shifted toward skimming or reading the AI's explanations in place of the code itself. These findings reveal that developers do not accurately perceive the cognitive resources they expend when managing AI tools. This study provides evidence of added cognitive effort associated with using generative AI, pointing towards a need to account for this dual-task penalty in software engineering tools and education.

cs.SE · cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2603.29322

VACP: Visual Analytics Context Protocol

Tobias Stähle, Péter Ferenc Gyarmati, Thilo Spinner, Rita Sevastjanova, Dominik Moritz, Mennatallah El-Assady

原文摘要

The rise of AI agents introduces a fundamental shift in Visual Analytics (VA), in which agents act as a new user group. Current agentic approaches - based on computer vision and raw DOM access - fail to perform VA tasks accurately and efficiently. This paper introduces the Visual Analytics Context Protocol (VACP), a framework designed to make VA applications "agent-ready" that extends generic protocols by explicitly exposing application state, available interactions, and mechanisms for direct execution. To support our context protocol, we contribute a formal specification of AI agent requirements and knowledge representations in VA interfaces. We instantiate VACP as a library compatible with major visualization grammars and web frameworks, enabling augmentation of existing systems and the development of new ones. Our evaluation across representative VA tasks demonstrates that VACP-enabled agents achieve higher success rates in interface interpretation and execution compared to current agentic approaches, while reducing token consumption and latency. VACP closes the gap between human-centric VA interfaces and machine perceivability, ensuring agents can reliably act as collaborative users in VA systems.

cs.HC

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.02076

Scene-Adaptive Readout and Decoding in Diffusion Driving VLAs

Seungjun Yu, Jiho Choi, Inha Kang, Hojun Choi, Seungwook Kim, Hyunjung Shim

原文摘要

Diffusion driving vision-language-action models repeatedly execute the backbone at every denoising stage, incurring substantial inference cost, while it remains unclear whether all scenes require the same decoding computation. To distinguish when a trajectory becomes readable from how much further refinement it requires, we analyze trajectory readout across network depth and decoding progress. We identify a depth-decoding dissociation in which late-layer representations already support useful trajectory readout after the first forward sweep, whereas early-layer representations benefit more from continued decoding. We further find that both the preferred readout layer and decoding depth vary across scenes. Based on these observations, we propose a scene-adaptive readout and decoding framework. A scene-conditioned readout MoE adaptively weights and combines trajectory predictions from layer-specific late-layer representations after a single backbone sweep, while an adaptive decoding policy applies additional native refinement only when required by the scene. We evaluate our framework on two diffusion driving VLAs across the Waymo E2E and nuScenes benchmarks. Our method improves validation RFS by up to 0.56 while also improving test performance, and achieves an end-to-end speedup of up to $6.0\times$ on Waymo E2E, while consistently reducing trajectory error on nuScenes. These results demonstrate that scene-level readout and decoding allocation improves planning quality while avoiding unnecessary decoding. Our code is available at https://github.com/sjyu001/SARD

cs.CV · cs.AI · cs.RO

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12016

Speculative Self-Distillation enables Efficient Knowledge Internalization

Shayan Talaei, Agam Bhatia, Arshia Soltani Moakhar, Jonas Hübotter, Amin Saberi, Azalia Mirhoseini

原文摘要

Many real-world applications require language models to internalize knowledge that was absent from pretraining, to incorporate continually updated facts, or to learn new domain skills. Self-distillation has emerged as a natural approach for this task. A model conditioned on the source material acts as the teacher, and an unconditioned student is trained to reproduce the teacher's behavior closed-book. We propose Speculative Self-Distillation (SSD), an efficient per-token mixed-policy method for self-distillation. The student generates each rollout by default, while the teacher intervenes only at positions where its next-token distribution substantially diverges from the student's. Unlike off-policy distillation with teacher-generated prefixes, this design keeps training rollouts close to the student's test-time behavior, while avoiding the low-signal states that fully on-policy rollouts drift into. In effect, SSD lets the teacher steer the rollout exactly where the privileged context changes the prediction. On three closed-book QA benchmarks measuring fact recall, knowledge updates, and skill acquisition, SSD nearly matches on-policy distillation while using 38-69% fewer supervised tokens and consistently outperforms off-policy distillation on the held-out test set. On a forgetting stress test, SSD retains the most general capability while also achieving the best in-domain accuracy and using less supervision than the on-policy baseline. Overall, SSD establishes a new accuracy-efficiency-forgetting Pareto frontier for self-distillation.

cs.LG · cs.AI · cs.CL

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.12012

Robust Domain-Mixture Scheduling for Multi-Teacher On-Policy Distillation

Yankai Chen, Sida Lin, Yifan Li, Hanjun Luo, Hanrong Zhang, Zhuohan Xie, Bowei He, Wei-Chieh Huang 等 13 位作者

原文摘要

Multi-teacher on-policy distillation (MOPD) merges the capabilities of multiple domain-expert teachers, such as mathematical reasoning, code generation, and instruction following, into a single compact student. Since all domains share a limited training budget, how to allocate it across domains directly determines the quality of distillation. Existing methods allocate the budget by each domain's learning progress, but offer no guarantee for lagging domains. We therefore introduce distributionally robust optimization (DRO) into MOPD to prioritize the worst-performing domain. However, the per-domain training signals are estimated from a few rollouts and are inherently noisy, so directly applying DRO mistakes noise for the worst case, lets occasional spikes hijack the budget, and protects inherently hard domains rather than truly regressing ones. To address these issues, we propose RDS-MOPD, which keeps the DRO objective and redesigns its three parts: it shifts the allocation only when differences clearly exceed the noise, bounds the influence of any single anomaly, and defines the worst case as the regression from each domain's own best level, all with theoretical guarantees. Experiments show that RDS-MOPD outperforms recent competing methods and ranks first on most mathematical reasoning, code generation, and instruction-following benchmarks.

cs.CL

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2610.27726

Anytime-Valid Runtime Monitoring of Learned Barrier Certificates

Zhipeng Cao, Taoran Wu, Weiru Cui, Luke Ong, Bai Xue

原文摘要

Learned barrier certificates provide a principled mechanism for establishing safety of stochastic systems, but their validity may deteriorate during deployment when the underlying dynamics differ from those used for training. We propose an anytime-valid runtime monitor that audits a learned barrier certificate using observed closed-loop trajectories. The monitor is intended to detect loss of statistical support for the deployed barrier condition; it does not by itself guarantee detection before unsafe entry or safety of a subsequent intervention. Specifically, the conditional barrier-drift condition is formulated as an episode-wise sequential hypothesis test, and observed barrier increments are transformed into a nonnegative test supermartingale, yielding finite-sample false-alarm control under arbitrary stopping times. Inspired by finite-horizon safety verification, we further introduce a deterministic temporal shaping of the monitoring boundary, making the monitor more conservative early in a prescribed deployment horizon while preserving anytime-validity. The monitor is designed to detect statistical evidence that the safety certificate is no longer consistent with the observed dynamics, rather than to predict unsafe entry. To assess the monitor empirically, we evaluate empirical alarm behavior under valid and invalid certificates. Code and data are available at https://github.com/FormalCert-AI/runtime-barrier-monitor.

eess.SY · cs.AI · cs.LG · math.OC · math.ST

查看论文
arXiv ↗PDF ↗
新论文 · v1arXiv:2611.11998

Reward-Estimated Hypergradient for Bilevel Reinforcement Learning with Black-Box Follower

Shigeki Kusaka, Mikoto Kudo, Takumi Tanabe, Akifumi Wachi, Youhei Akimoto

原文摘要

Hypergradient (HG) methods for bilevel reinforcement learning (RL) typically assume a white-box follower whose reward is known. In practice, however, the leader rarely knows the follower's reward function and can only observe its trajectories. We address this black-box setting and propose Reward-Estimated Hypergradients (RE-HG), which recovers the follower's reward from these trajectories via inverse RL (IRL). Yet IRL identifies the reward only up to potential-based shaping, a transformation that leaves the follower's behavior unchanged but biases the HG. We prove that, for linear rewards, this bias cancels when IRL estimates are aggregated across sufficiently diverse dynamics, which the leader's own updates naturally supply. To handle finite samples, RE-HG truncates small eigenvalues of the aggregated feature covariance, restoring the identifiable subspace and stabilizing the estimation. Experiments show that RE-HG estimates gradients aligned with a white-box Oracle and achieves comparable performance; empirically, the truncation also suppresses HG explosions, letting RE-HG avoid local optima that trap the Oracle in sharp reward landscapes.

cs.LG

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2611.03698

Training LLM Agents to Automate LLM Post-Training with RSI-Dojo

Shuzheng Gao, Zhaojian Yu, Penghao Yin, Yuling Shi, Kai Cai, Shilin He, Michael R. Lyu

原文摘要

Recursive Self-Improvement (RSI) aims to enable AI systems to autonomously build and improve AI systems themselves, including models, harnesses, and underlying infrastructure. Within this broader agenda, a central capability is enabling agents to train and improve base models. However, existing studies on automated model training primarily focused on workflow- or harness-level optimizations, rather than internalizing the ability to conduct post-training into the model itself. Developing this capability requires executable model training environments that can produce high-quality trajectories demonstrating how to select data, train models, diagnose failures, and iteratively improve results. In this paper, we introduce RSI-Dojo, a collection of executable model training environments designed to teach LLM agents to conduct post-training. To construct RSI-Dojo, we develop RSI-Craft, an automatic pipeline that converts human research artifacts into executable model training environments. RSI-Craft automatically collects research papers from top venues, instantiates datasets and metrics into a unified interface, and reproduces evaluation dependencies. It then filters tasks by data size, execution reliability, and discriminability, yielding RSI-Dojo with 281 seed model training tasks. Building on RSI-Dojo, we employ a trajectory collection method with an improved harness to collect high-quality model-training trajectories from frontier models, then retain examples through rubric-based and outcome-based filtering. We fine-tune Qwen3.5-9B and Qwen3.6-35B-A3B on these trajectories. The resulting RSI-Dojo-9B and RSI-Dojo-35B-A3B improve over their base models by 8.74 and 16.87 PostTrainBench points, respectively. RSI-Dojo-35B-A3B reaches 29.95 on PostTrainBench and 53.29 on PostTrainBench-Lite, matching the performance of frontier models such as GLM-5.2 and GPT 5.5.

cs.AI

查看论文
arXiv ↗PDF ↗
版本更新 · v2arXiv:2504.17410

Bias-Eliminated PnP for Stereo Visual Odometry: Provably Consistent and Large-Scale Localization

Guangyang Zeng, Yuan Shen, Ziyang Hong, Yuze Hong, Viorela Ila, Guodong Shi, Junfeng Wu

原文摘要

In this paper, we first present a bias-eliminated weighted (Bias-Eli-W) perspective-n-point (PnP) estimator for stereo visual odometry (VO) with provable consistency. Specifically, leveraging statistical theory, we develop an asymptotically unbiased and $\sqrt {n}$-consistent PnP estimator that accounts for varying 3D triangulation uncertainties, ensuring that the relative pose estimate converges to the ground truth as the number of features increases. Next, on the stereo VO pipeline side, we propose a framework that continuously triangulates contemporary features for tracking new frames, effectively decoupling temporal dependencies between pose and 3D point errors. We integrate the Bias-Eli-W PnP estimator into the proposed stereo VO pipeline, creating a synergistic effect that enhances the suppression of pose estimation errors. We validate the performance of our method on the KITTI and Oxford RobotCar datasets. Experimental results demonstrate that our method: 1) achieves significant improvements in both relative pose error and absolute trajectory error in large-scale environments; 2) provides reliable localization under erratic and unpredictable robot motions. The successful implementation of the Bias-Eli-W PnP in stereo VO indicates the importance of information screening in robotic estimation tasks with high-uncertainty measurements, shedding light on diverse applications where PnP is a key ingredient.

cs.RO

查看论文
arXiv ↗PDF ↗