arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Chinese Academy of Sciences(中国科学院大学)

2026-08-21 至 2026-08-21 共收录 4
2608.15242 2026-08-21 cs.AI cs.SE 版本更新

LongRCA Bench: Diagnosing Responsible Roles and Root Causes in Long-Horizon Agent Failures

LongRCA Bench:诊断长 horizon 智能体失败中的责任角色与根本原因

Yunfei Zhang, Boyu Feng, Changhua Pei, Zexin Wang, Zhihuang Peng, Xinlong Liu, Hengyue Jiang, Difeng Ma, Jiayi Zhang, Yongzhou Yao, Yanan Zhao, Fei Sun, Yintong Huo, Zhaoyang Liu, Jingjing Li, Gaogang Xie, Dan Pei

机构 * Computer Network Information Center, Chinese Academy of Sciences(中国科学院计算机网络信息中心) Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences(中国科学院大学杭州高等研究院) Chongqing University(重庆大学) Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所) Singapore Management University(新加坡管理大学) Tongyi Lab, Alibaba Group(阿里巴巴集团通义实验室) Tsinghua University(清华大学)

AI总结 LongRCA Bench 是含1140条失败轨迹的长 horizon 智能体失败诊断基准,本文提出无需训练的 RCTA 方法,在责任角色归因和根本步骤定位上优于现有基线,表明需将二者作为独立评估目标。

Comments 18 pages, 6 figures. Yunfei Zhang and Boyu Feng contributed equally. Changhua Pei is the corresponding author

详情

展开后加载摘要…

URL PDF HTML 收藏
2608.11907 2026-08-21 cs.CV cs.AI 版本更新

A Model-Internal Protocol for Assessing Multimodal Models as Integrated Systems

你能看到你所绘制的内容吗?面向统一多模态模型整体评估的语义闭环框架

Hao Zhang, Jiaxin Qi, Zhijiang Tang, Jianqiang Huang

机构 * Hangzhou Institute for Advanced Study(杭州高等研究院) University of Chinese Academy of Sciences(中国科学院大学) Computer Network Information Center(计算机网络信息中心) Chinese Academy of Sciences(中国科学院)

AI总结 本研究针对统一多模态模型的系统级评估缺口,提出无标注的SGU语义闭环框架,通过感知-生成-推理流程评估模型综合能力,发现高性能模型存在自生成上下文推理局限,为下一代模型开发提供基准基础。

Comments 21 pages, 8 figures, 12 tables. Title updated; author contribution and corresponding-author notes added. Hao Zhang and Jiaxin Qi contributed equally; Jianqiang Huang is the corresponding author

详情

展开后加载摘要…

URL PDF HTML 收藏
2607.29011 2026-08-21 cs.RO 版本更新

DART: Dual-Axis Airborne Reachability-Gated Torque-Reaction for Off-Road Vehicle Jumps

DART:用于越野车辆跳跃的双轴机载可达性门控扭矩反应

Yu Hu, Fangzhou Zhao, Mingyuan Sang, Chen Min, Liang Chen, Wei Li, Wenyu Kuang, Shican Chen, Jinwei Li, Baolei Chen

机构 * Research Center for Intelligent Computing Systems, Institute of Computing Technology, CAS(中国科学院计算技术研究所智能计算系统研究中心) School of Computer Science and Technology, University of Chinese Academy of Sciences(中国科学院大学计算机科学与技术学院) Dong Feng Off-Road Vehicle Co., Ltd(东风越野车有限公司)

AI总结 该研究针对越野车辆腾空着陆的碰撞风险,提出DART双轴机载控制方法,通过起飞前速度整形和飞行中扭矩调节提升着陆成功率,仿真中其性能优于RW-PD和TOBB方法。

Comments 20 pages, 9 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2603.21875 2026-08-21 eess.AS cs.CL cs.SD 版本更新

Disentangling Speaker Traits for Deepfake Source Verification via Chebyshev Polynomial and Riemannian Metric Learning

通过切比雪夫多项式和黎曼度量学习实现深度伪造源验证的说话人特性解耦

Xi Xuan, Wenxin Zhang, Zhiyu Li, Jennifer Williams, Ville Hautamäki, Tomi H. Kinnunen

机构 * University of Eastern Finland(东芬兰大学) City University of Hong Kong(香港城市大学) University of Southampton(南安普顿大学) University of Chinese Academy of Sciences(中国科学院大学) University of Science and Technology of China(中国科学技术大学)

AI总结 本文提出SDML框架,通过切比雪夫多项式和黎曼度量学习解耦说话人特性,提升深度伪造源验证的准确性。

Comments Accepted to Interspeech 2026; The code, evaluation protocols and demo website are available at this https URL (https://github.com/xxuan-acoustics/RiemannSD-Net)

详情

展开后加载摘要…

URL PDF HTML 收藏