arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

The Chinese University of Hong Kong(香港中文大学)

共收录 2402
2602.23266 2026-02-27 cs.CL

Discourse-Aware Dual-Track Streaming Response for Low-Latency Spoken Dialogue Systems

面向 discourse 的双轨流式响应用于低延迟语音对话系统

Siyuan Liu, Jiahui Xu, Feng Jiang, Kuang Wang, Zefeng Zhao, Chu-Ren Huang, Jinghang Gu, Changqing Yin, Haizhou Li

机构 * School of Computer Science and Technology, Tongji University(同济大学计算机科学与技术学院) Department of Language Science and Technology, Hong Kong Polytechnic University(香港理工大学语言科学与技术系) Artificial Intelligence Research Institute, Shenzhen University of Advanced Technology(深圳先进技术大学人工智能研究院) School of Data Science, Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)数据科学学院) School of Artificial Intelligence, Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)人工智能学院)

AI总结 DDTSR 提出了一种低延迟语音对话系统框架,通过双轨流式响应机制实现听中思考和说中思考,显著降低响应延迟并保持 discourse 质量。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.23235 2026-02-27 cs.CV cs.AI

Spatio-Temporal Token Pruning for Efficient High-Resolution GUI Agents

时空令牌剪枝用于高效高分辨率GUI代理

Zhou Xu, Bowen Zhou, Qi Wang, Shuwen Feng, Jingyu Xiao

机构 * Tsinghua University(清华大学) Xidian University(西安电子科技大学) The Chinese University of Hong Kong(香港中文大学)

AI总结 GUIPruner通过时空令牌剪枝技术,实现高分辨率GUI导航的高效处理,显著提升性能并降低资源消耗。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21893 2026-02-27 cs.CV

EndoDDC: Learning Sparse to Dense Reconstruction for Endoscopic Robotic Navigation via Diffusion Depth Completion

EndoDDC: 通过扩散深度补全学习稀疏到密集重建以实现内窥镜机器人导航

Yinheng Lin, Yiming Huang, Beilei Cui, Long Bai, Huxin Gao, Hongliang Ren, Jiewen Lai

机构 * Department of Electronic Engineering, The Chinese University of Hong Kong(电子工程系,香港中文大学)

AI总结 EndoDDC通过扩散模型整合图像、稀疏深度信息和深度梯度特征,提升内窥镜环境下深度重建的准确性和鲁棒性。

Comments Accepted by ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.22633 2026-02-27 cs.LG cs.DC

Tackling Privacy Heterogeneity in Differentially Private Federated Learning

应对差分隐私联邦学习中的隐私异质性

Ruichen Xu, Ying-Jun Angela Zhang, Jianwei Huang

机构 * Department of Information Engineering, The Chinese University of Hong Kong(信息工程系,香港中文大学) School of Science and Engineering, Shenzhen Institute of Artificial Intelligence and Robotics for Society(科学与工程学院,深圳人工智能与机器人社会研究院)

AI总结 本文提出了一种隐私感知的客户端选择策略,通过理论分析和实验验证,在异构隐私预算下提升了联邦学习的测试准确率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.22581 2026-02-27 cs.LG

IBCircuit: Towards Holistic Circuit Discovery with Information Bottleneck

IBCircuit: 向信息瓶颈原则下的整体电路发现迈进

Tian Bian, Yifan Niu, Chaohao Yuan, Chengzhi Piao, Bingzhe Wu, Long-Kai Huang, Yu Rong, Tingyang Xu, Hong Cheng, Jia Li

机构 * Department of Systems Engineering Engineering Management, The Chinese University of Hong Kong, Hong Kong, China DAMO Academy, Alibaba Group, Hangzhou, China Hong Kong University of Science Hupan Lab, Hangzhou, China Department of Computer Science, Hong Kong Baptist University, Hong Kong, China School of Artificial Intelligence, Shenzhen University, Shenzhen, China Tencent, Shenzhen, China

AI总结 IBCircuit基于信息瓶颈原理,提出端到端方法实现整体电路发现,适用于任意任务,提升电路识别的准确性和效率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.24796 2026-02-27 cs.LO cs.AI cs.FL cs.LG math.CT

LeanCat: A Benchmark Suite for Formal Category Theory in Lean (Part I: 1-Categories)

LeanCat:Lean中的形式范畴论基准测试套件(第一部分:1-范畴)

Rongge Xu, Hui Dai, Yiming Fu, Jiedong Jiang, Tianjiao Nie, Junkai Wang, Holiverse Yang, Zhi-Hao Zhang

机构 * Yau Mathematical Sciences Center, Tsinghua University(清华大学尤洋数学科学中心) Iluvatar CoreX Department of Mathematics, Southern University of Science and Technology(南方科技大学数学系) Westlake Institute for Advanced Study, Westlake University(西湖研究学院) Qiuzhen College, Tsinghua University(清华大学齐臻学院) Department of Physics, The Chinese University of Hong Kong(香港中文大学物理系) Yanqi Lake Beijing Institute of Mathematical Sciences and Applications (BIMSA)(燕琦湖北京应用数学研究所(BIMSA))

AI总结 LeanCat通过100个形式化范畴论任务测试模型的抽象能力,揭示了现有模型在组合泛化上的不足,提出LeanBridge通过迭代优化提升性能至24%。

Comments 22 pages, 9 figures, 5 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.24266 2026-02-27 cs.RO cs.HC

SignBot: Learning Human-to-Humanoid Sign Language Interaction

SignBot: 学习人类-人形机器人手语交互

Guanren Qiao, Sixu Lin, Ronglai Zuo, Zhizheng Wu, Kui Jia, Guiliang Liu

机构 * School of Data Science, the Chinese University of Hong Kong, Shenzhen(数据科学学院,香港中文大学(深圳)) Imperial College London(伦敦帝国理工学院)

AI总结 SignBot通过整合运动重定向、运动控制和生成交互模块,实现人与人形机器人之间自然的手语交流,提升聋人和听力障碍者的沟通可及性。

Comments Accepted by ICRA 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21952 2026-02-26 cs.CV

MindDriver: Introducing Progressive Multimodal Reasoning for Autonomous Driving

MindDriver: 引入渐进多模态推理用于自动驾驶

Lingjun Zhang, Yujian Yuan, Changjie Wu, Xinyuan Chang, Xin Cai, Shuang Zeng, Linzhe Shi, Sijin Wang, Hang Zhang, Mu Xu

机构 * Amap, Alibaba Group(阿里集团蚂巴公司) The Hong Kong University of Science and Technology(香港科学与技术大学) The Chinese University of Hong Kong(香港中文大学) Xi’an Jiaotong University(西安交通大学)

AI总结 MindDriver通过渐进多模态推理框架提升自动驾驶系统的推理能力,实现语义到物理空间的转化与轨迹规划,展现优异的性能表现。

Comments CVPR2026; Yujian Yuan and Lingjun Zhang contributed equally with random order

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21844 2026-02-26 cs.LG cs.DC cs.GT

JSAM: Privacy Straggler-Resilient Joint Client Selection and Incentive Mechanism Design in Differentially Private Federated Learning

JSAM: 一种隐私抗延迟的联合客户端选择和激励机制设计在差分隐私联邦学习中

Ruichen Xu, Ying-Jun Angela Zhang, Jianwei Huang

机构 * Department of Information Engineering, The Chinese University of Hong Kong(信息工程系,香港中文大学) School of Science and Engineering, Shenzhen Institute of Artificial Intelligence and Robotics for Society(科学与工程学院,深圳人工智能与机器人研究院) The Chinese University of Hong Kong(香港中文大学)

AI总结 JSAM通过联合优化客户端选择与隐私补偿,提升差分隐私联邦学习的训练效果和效率,实现隐私保护与参与激励的平衡。

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.10082 2026-02-26 cs.CV

LoRA-Edit: Controllable First-Frame-Guided Video Editing via Mask-Aware LoRA Fine-Tuning

LoRA-Edit: 通过掩码感知LoRA微调实现可控的第一帧引导视频编辑

Chenjian Gao, Lihe Ding, Xin Cai, Zhanpeng Huang, Zibin Wang, Tianfan Xue

机构 * Multimedia Laboratory, The Chinese University of Hong Kong(香港中文大学多媒体实验室) SenseTime Research(商汤科技研究院)

AI总结 LoRA-Edit通过时空掩码引导LoRA微调,实现对视频编辑全过程的可控编辑,支持生成动态变化的视频内容。

Comments ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21612 2026-02-26 cs.RO

Jumping Control for a Quadrupedal Wheeled-Legged Robot via NMPC and DE Optimization

四足轮腿机器人通过NMPC和DE优化实现跳跃控制

Xuanqi Zeng, Lingwei Zhang, Linzhu Yue, Zhitao Song, Hongbo Zhang, Tianlin Zhang, Yun-Hui Liu

机构 * Department of Mechanical and Automation Engineering, The Chinese University of Hong Kong(机械与自动化工程系,香港中文大学)

AI总结 本文提出了一种结合NMPC和DE优化的四足轮腿机器人跳跃控制方法,实现了多种跳跃动作并验证了其有效性。

Comments 8 pages, 12 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21589 2026-02-26 cs.CV

SEF-MAP: Subspace-Decomposed Expert Fusion for Robust Multimodal HD Map Prediction

SEF-MAP:子空间分解专家融合用于鲁棒多模态高精度地图预测

Haoxiang Fu, Lingfeng Zhang, Hao Li, Ruibing Hu, Zhengrong Li, Guanjing Liu, Zimu Tan, Long Chen, Hangjun Ye, Xiaoshuai Hao

机构 * National University of Singapore(新加坡国立大学) Xiaomi EV(小米电动车) Chinese University of Hong Kong(香港中文大学) The University of Manchester(曼彻斯特大学) Renmin University of China(中国人民大学)

AI总结 SEF-MAP通过子空间分解和专家融合,提升多模态HD地图预测的鲁棒性和准确性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21552 2026-02-26 cs.CV

Generalizing Visual Geometry Priors to Sparse Gaussian Occupancy Prediction

将视觉几何先验泛化到稀疏高斯占用预测

Changqing Zhou, Yueru Luo, Changhao Chen

机构 * The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州)) The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))

AI总结 GPOcc通过利用可泛化的视觉几何先验进行单目占用预测,在速度和精度上均优于现有方法。

Comments Accepted by CVPR2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21230 2026-02-26 cs.CL

TRACE: Trajectory-Aware Comprehensive Evaluation for Deep Research Agents

TRACE: 基于轨迹的深度研究代理综合评估

Yanyu Chen, Jiyue Jiang, Jiahong Liu, Yifei Zhang, Xiao Guo, Irwin King

机构 * The Chinese University of Hong Kong(香港中文大学)

AI总结 TRACE提出基于轨迹的深度研究代理综合评估框架,通过分层效用函数和阶梯式能力评估协议,解决单一指标评估的不足,揭示代理在准确性、效率和鲁棒性之间的关键权衡。

Comments Accepted by WWW 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21224 2026-02-26 cs.CL cs.AI cs.DC cs.LG

Make Every Draft Count: Hidden State based Speculative Decoding

让每个草稿都值得:基于隐藏状态的推测解码

Yuetao Chen, Xuliang Wang, Xinzhou Zheng, Ming Li, Peng Wang, Hong Xu

机构 * The Chinese University of Hong Kong(香港中文大学) University of Waterloo(滑铁卢大学) University of Science and Technology of China(中国科学技术大学)

AI总结 本文提出基于隐藏状态的推测解码系统,通过自回归预测和延迟整合标记信息,提高隐藏状态重用率,实现3.3倍的推理速度提升。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.21221 2026-02-26 cs.LG cs.AI cs.CL

Latent Context Compilation: Distilling Long Context into Compact Portable Memory

潜在上下文编译:将长上下文编译为紧凑便携内存

Zeju Li, Yizhou Zhou, Qiang Xu

机构 * The Chinese University of Hong Kong(香港中文大学)

AI总结 本文提出潜在上下文编译方法,通过LoRA模块将长上下文压缩为紧凑内存,无需合成数据即可提升LLM部署效率。

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.00981 2026-02-26 cs.SD

FlexiCodec: A Dynamic Neural Audio Codec for Low Frame Rates

FlexiCodec: 一种用于低帧率的动态神经音频编解码器

Jiaqi Li, Yao Qian, Yuxuan Hu, Leying Zhang, Xiaofei Wang, Heng Lu, Manthan Thakker, Jinyu Li, Sheng Zhao, Zhizheng Wu

机构 * The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳)) Microsoft(微软公司) Shanghai Jiao Tong University(上海交通大学) Shenzhen Loop Area Institute(深圳河套学院) City University of Macau(澳门城市大学) Amphion Technology Co., Ltd.(Amphion科技有限公司)

AI总结 FlexiCodec通过动态帧率方法和双流编码架构,提升低帧率下的语义保留和音频重建质量。

Comments Accepted to ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20159 2026-02-25 cs.CV cs.AI cs.LG cs.MM cs.RO

A Very Big Video Reasoning Suite

一个非常大的视频推理套件

Maijunxian Wang, Ruisi Wang, Juyi Lin, Ran Ji, Thaddäus Wiedemer, Qingying Gao, Dezhi Luo, Yaoyao Qian, Lianyu Huang, Zelong Hong, Jiahui Ge, Qianli Ma, Hang He, Yifan Zhou, Lingzi Guo, Lantao Mei, Jiachen Li, Hanwen Xing, Tianqi Zhao, Fengyuan Yu, Weihang Xiao, Yizheng Jiao, Jianheng Hou, Danyang Zhang, Pengcheng Xu, Boyang Zhong, Zehong Zhao, Gaoyun Fang, John Kitaoka, Yile Xu, Hua Xu, Kenton Blacutt, Tin Nguyen, Siyuan Song, Haoran Sun, Shaoyue Wen, Linyang He, Runming Wang, Yanzhi Wang, Mengyue Yang, Ziqiao Ma, Raphaël Millière, Freda Shi, Nuno Vasconcelos, Daniel Khashabi, Alan Yuille, Yilun Du, Ziming Liu, Bo Li, Dahua Lin, Ziwei Liu, Vikash Kumar, Yijiang Li, Lei Yang, Zhongang Cai, Hokin Deng

机构 * University of California, Berkeley(加州大学伯克利分校) Nanyang Technological University(南洋理工大学) Northeastern University(东北大学) University of Tübingen(图宾根大学) Johns Hopkins University(约翰霍普金斯大学) University of Michigan(密歇根大学) University of Southern California(南加州大学) Washington University in St. Louis(圣路易斯华盛顿大学) Shanghai Jiao Tong University(上海交通大学) East China Normal University(华东师范大学) Stanford University(斯坦福大学) University of Texas at Austin(得克萨斯大学奥斯汀分校) University of California, Los Angeles(加州大学洛杉矶分校) Cornell University(康奈尔大学) San Jose State University(圣何塞州立大学) University of California, Irvine(加州大学尔湾分校) Technical University of Munich(慕尼黑技术大学) University of California, San Diego(加州大学圣地亚哥分校) Imperial College London(伦敦帝国学院) University of Wisconsin--Madison(威斯康星大学麦迪逊分校) University of Edinburgh(爱丁堡大学) Hong Kong University of Science(香港科学大学) New York University(纽约大学) Auburn University(阿伯丁大学) Columbia University(哥伦比亚大学) University of Bristol(布里斯托大学) University of Waterloo(滑铁卢大学) The Chinese University of Hong Kong(香港中文大学) Carnegie Mellon University(卡内基梅隆大学) University of Oxford(牛津大学) University of North Carolina at Chapel Hill(北卡罗来纳大学教堂山分校)

AI总结 VBVR数据集和评估框架旨在解决视频推理能力研究中的大规模数据缺乏问题,通过大规模实验观察到对未见任务的泛化能力。

Comments Homepage: https://video-reason.com/

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.03356 2026-02-25 cs.RO

Effective Reinforcement Learning Control using Conservative Soft Actor-Critic

通过保守的软演员-评论家实现有效的强化学习控制

Zhiwei Shang, Xinyi Yuan, Wenjun Huang, Yunduan Cui, Di Chen, Meixin Zhu

机构 * School of Data Science, The Chinese University of Hong Kong, Shenzhen, China(香港中文大学(深圳)数据科学学院) Graduate School of Engineering Science, Osaka University, Japan(大阪大学工学研究科) Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences, China(中国科学院深圳先进技术研究院) Department of Electrical and Electronic Engineering, The Hong Kong Polytechnic University, Hong Kong SAR, China(香港理工大学电子与电气工程系) School of Transportation, Southeast University, China(东南大学交通运输学院)

AI总结 本文提出保守的软演员-评论家算法,通过熵和相对熵正则化提升强化学习控制的稳定性与效率。

Comments 8 pages, 9 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20859 2026-02-25 cs.CL

FinAnchor: Aligned Multi-Model Representations for Financial Prediction

FinAnchor:用于金融预测的对齐多模型表示

Zirui He, Huopu Zhang, Yanguang Liu, Sirui Wu, Mengnan Du

机构 * New Jersey Institute of Technology(新泽西理工学院) Georgia Institute of Technology(佐治亚理工学院) Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))

AI总结 FinAnchor通过整合多个LLM的嵌入并对齐到锚定空间,提升金融预测的鲁棒性与性能。

Comments 11 pages, 4 figures, 5 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20818 2026-02-25 cs.CV

GatedCLIP: Gated Multimodal Fusion for Hateful Memes Detection

GatedCLIP:用于仇恨表情包检测的门控多模态融合

Yingying Guo, Ke Zhang, Zirong Zeng

机构 * The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))

AI总结 GatedCLIP通过门控融合和对比学习提升多模态仇恨表情包检测性能,实现0.66 AUROC优于CLIP基线。

Comments Preprint

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20636 2026-02-25 cs.CV cs.AI

SurgAtt-Tracker: Online Surgical Attention Tracking via Temporal Proposal Reranking and Motion-Aware Refinement

SurgAtt-Tracker: 通过时间提案重排和运动感知细化实现在线手术注意力跟踪

Rulin Zhou, Guankun Wang, An Wang, Yujie Ma, Lixin Ouyang, Bolin Cui, Junyan Li, Chaowei Zhu, Mingyang Li, Ming Chen, Xiaopin Zhong, Peng Lu, Jiankun Wang, Xianming Liu, Hongliang Ren

机构 * Department of Electronic Engineering, The Chinese University of Hong Kong, Hong Kong SAR, China(香港中文大学电子工程系) Division of Gastrointestinal Surgery, Shenzhen People's Hospital, China(深圳人民医院胃肠外科) College of Mechatronics and Control Engineering, Shenzhen University, China(深圳大学机电与控制工程学院) Department of Mechanical Engineering, The University of Hong Kong, Hong Kong SAR, China(香港大学机械工程系) Department of Electronic and Electrical Engineering, Southern University of Science and Technology(南方科技大学电子与电气工程系)

AI总结 SurgAtt-Tracker通过时间提案重排和运动感知细化实现手术注意力跟踪,提供帧级视野指导以支持机器人规划和自动摄像头控制。

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20548 2026-02-25 cs.CV

Robust Spiking Neural Networks Against Adversarial Attacks

对抗攻击下的鲁棒脉冲神经网络

Shuai Wang, Malu Zhang, Yulin Jiang, Dehao Zhang, Ammar Belatreche, Yu Liang, Yimeng Shan, Zijian Zhou, Yang Yang, Haizhou Li

机构 * University of Electronic Science and Technology of China(电子科技大学) Northumbria University(诺丁汉大学) Shenzhen Loop Area Institute(深圳河套学院) The Chinese University of Hong Kong, Shenzhen(香港中文大学(深圳))

AI总结 本文提出阈值保护优化方法,通过增加梯度稀疏性和引入噪声脉冲神经元,提升直接训练SNNs在对抗攻击下的鲁棒性。

Comments Published as a conference paper at ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20344 2026-02-25 cs.LG cs.AI q-bio.QM

Hierarchical Molecular Representation Learning via Fragment-Based Self-Supervised Embedding Prediction

通过基于片段的自监督嵌入预测进行层次分子表示学习

Jiele Wu, Haozhe Ma, Zhihan Guo, Thanh Vinh Vo, Tze Yun Leong

机构 * School of Computing, National University of Singapore, Singapore(新加坡国立大学计算机学院) Engineering, The Chinese University of Hong Kong, China(香港中文大学工程学院)

AI总结 本文提出GraSPNet,通过层次自监督框架学习分子的原子和片段级语义,以提升分子性质预测的性能。

Comments 15 pages (8 pages main text),8 figures

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20216 2026-02-25 cs.RO

Sample-Efficient Learning with Online Expert Correction for Autonomous Catheter Steering in Endovascular Bifurcation Navigation

基于在线专家纠正的高效学习用于内血管分叉导航的自主导管操控

Hao Wang, Tianliang Yao, Bo Lu, Zhiqiang Pei, Liu Dong, Lei Ma, Peng Qi

机构 * Tongji University(同济大学) College of Electronics and Information Engineering(电子信息工程学院) Shanghai Institute of Intelligent Science and Technology(上海智能科学与技术研究院) The Chinese University of Hong Kong(香港中文大学) School of Mechanical and Electric Engineering(机械电子工程学院) Soochow University(苏州大学) Shanghai Operation Robot Co., Ltd.(上海操作机器人有限公司) State Key Laboratory of Cardiovascular Diseases(心血管疾病国家重点实验室)

AI总结 本文提出一种结合在线专家纠正的样本高效强化学习框架,用于内血管分叉导航中的自主导管操控,提升了在复杂血管结构中的操控精度和鲁棒性。

Comments This paper has been accepted by IEEE ICRA 2026. 8 pages, 5 figures, 1 table

详情

展开后加载摘要…

URL PDF HTML 收藏
2602.20215 2026-02-25 cs.RO

Vision-Based Reasoning with Topology-Encoded Graphs for Anatomical Path Disambiguation in Robot-Assisted Endovascular Navigation

基于拓扑编码图的视觉推理用于机器人辅助血管内导航中的解歧

Jiyuan Zhao, Zhengyu Shi, Wentong Tian, Tianliang Yao, Dong Liu, Tao Liu, Yizhe Wu, Peng Qi

机构 * Department of Control Science and Engineering, College of Electronics and Information Engineering, and Shanghai Institute of Intelligent Science and Technology, Tongji University(控制科学与工程系,电子与信息工程学院和智能科学与技术上海研究院,同济大学) Department of Electronic Engineering, Faculty of Engineering, The Chinese University of Hong Kong(电子工程系,工程学院,香港中文大学) Shanghai Operation Robot Co., Ltd.(上海操作机器人有限公司) Department of Cardiology, Zhongshan Hospital, Fudan University, Shanghai Institute of Cardiovascular Diseases, National Clinical Research Center for Interventional Medicine(心内科,复旦大学中山医院,上海心血管疾病研究所,介入医学临床研究中心) State Key Laboratory of Cardiovascular Diseases and Medical Innovation Center, Shanghai East Hospital, School of Medicine, Tongji University(心血管疾病国家重点实验室和医学创新中心,上海东医院,同济大学医学院)

AI总结 本文提出SCAR-UNet-GAT框架,通过拓扑编码图实现机器人辅助血管内导航中的路径解歧,显著提升路径规划的准确性和鲁棒性。

Comments This paper has been accepted by IEEE ICRA 2026. 8 pages, 3 figures, 3 tables

详情

展开后加载摘要…

URL PDF HTML 收藏
2511.16156 2026-02-25 cs.CV

Pluggable Pruning with Contiguous Layer Distillation for Diffusion Transformers

可插拔剪枝与连续层蒸馏用于扩散变换器

Jian Ma, Qirong Peng, Xujie Zhu, Peixing Xie, Chen Chen, Haonan Lu

机构 * OPPO AI Center(OPPO人工智能中心) Sun Yat-sen University(中山大学) The Chinese University of Hong Kong(香港中文大学)

AI总结 本研究提出PPCL方法,通过连续层蒸馏和灵活剪枝技术,在保持图像生成质量的同时实现50%的参数压缩,适用于资源受限环境。

Comments Accepted to CVPR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.19975 2026-02-25 cs.LG

From Samples to Scenarios: A New Paradigm for Probabilistic Forecasting

从样本到场景:概率预测的新范式

Xilin Dai, Zhijian Xu, Wanxu Cai, Qiang Xu

机构 * ZJU-UIUC Institute, Zhejiang University(浙江大学ZJU-UIUC研究院) Department of Computer Science and Engineering, The Chinese University of Hong Kong(香港中文大学计算机科学与工程系) School of Software, Tsinghua University(清华大学软件学院)

AI总结 本文提出概率场景范式,通过直接生成场景与概率对,改进概率预测模型,实现更高效和准确的预测效果。

Comments Accepted by ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.15148 2026-02-25 cs.CL

ATTS: Asynchronous Test-Time Scaling via Conformal Prediction

ATTS: 通过置信预测实现异步测试时间扩展

Jing Xiong, Qiujiang Chen, Fanghua Ye, Zhongwei Wan, Chuanyang Zheng, Chenyang Zhao, Hui Shen, Hanbo Li, Chaofan Tao, Haochen Tan, Haoli Bai, Lifeng Shang, Lingpeng Kong, Ngai Wong

机构 * The University of Hong Kong(香港大学) Huawei Technologies Co., Ltd(华为技术有限公司) University College London(伦敦大学学院) The Chinese University of Hong Kong(香港中文大学)

AI总结 ATTS通过异步测试时间扩展方法,在保持精度的同时实现测试时间扩展的显著加速和吞吐量提升。

Comments Accepted by ICLR 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.12600 2026-02-25 cs.CV

Revisiting the Generalization Problem of Low-level Vision Models Through the Lens of Image Deraining

通过图像去雨的视角重新审视低级视觉模型的泛化问题

Jinfan Hu, Zhiyuan You, Jinjin Gu, Kaiwen Zhu, Tianfan Xue, Chao Dong

机构 * Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(深圳先进技术研究院,中国科学院) University of Chinese Academy of Sciences(中国科学院大学) The Chinese University of Hong Kong(香港中文大学) Shanghai Jiao Tong University(上海交通大学) Shanghai Artificial Intelligence Laboratory(上海人工智能实验室) Shenzhen Key Lab of Computer Vision and Pattern Recognition, Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences(深圳计算机视觉与模式识别重点实验室,深圳先进技术研究院,中国科学院) Shenzhen University of Advanced Technology(深圳大学)

AI总结 本文通过图像去雨研究,揭示低级视觉模型泛化问题源于捷径学习,提出平衡数据复杂性和利用生成模型先验以提升模型鲁棒性。

Comments arXiv admin note: substantial text overlap with arXiv:2305.15134

详情

展开后加载摘要…

URL PDF HTML 收藏