arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

Harbin Institute of Technology(哈尔滨工业大学)

2026-01-21 至 2026-01-21 共收录 10
2601.13722 2026-01-21 cs.CL cs.AI

OP-Bench: Benchmarking Over-Personalization for Memory-Augmented Personalized Conversational Agents

OP-Bench:用于内存增强个性化对话代理的过个性化基准测试

Yulin Hu, Zimo Long, Jiahe Guo, Xingyu Sui, Xing Fu, Weixiang Zhao, Yanyan Zhao, Bing Qin

机构 * Harbin Institute of Technology(哈尔滨工业大学)

AI总结 OP-Bench通过评估多种模型和方法,揭示了内存增强对话代理中过个性化问题的普遍性,并提出Self-ReCheck机制以缓解该问题。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.13243 2026-01-21 cs.LG

A Comprehensive Evaluation of LLM Reasoning: From Single-Model to Multi-Agent Paradigms

大语言模型推理的全面评估:从单模型到多智能体范式

Yapeng Li, Jiakuo Yu, Zhixin Liu, Xinnan Liu, Jing Yu, Songze Li, Tonghua Su

机构 * Harbin Institute of Technology(哈尔滨工业大学)

AI总结 本研究全面评估了LLM推理范式,包括单模型和多智能体系统,通过新基准测试揭示了不同范式在成本与准确率之间的权衡。

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.23770 2026-01-21 cs.CV

GenView++: Unifying Adaptive Generative Augmentation and Quality-Driven Supervision for Contrastive Representation Learning

GenView++:统一自适应生成增强和质量驱动监督以实现对比表征学习

Xiaojie Li, Bei Wang, Wei Liu, Jianlong Wu, Yue Yu, Liqiang Nie, Min Zhang

机构 * Harbin Institute of Technology, Shenzhen(哈尔滨工业大学(深圳)) Peng Cheng Laboratory(鹏城实验室)

AI总结 GenView++通过自适应生成增强和质量驱动监督,提升对比学习在视觉和视觉-语言任务中的性能。

Comments The code is available at \url{https://github.com/xiaojieli0903/GenViewPlusPlus}

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12995 2026-01-21 cs.CL

Graph Reasoning Paradigm: Structured and Symbolic Reasoning with Topology-Aware Reinforcement Learning for Large Language Models

图推理范式:基于拓扑感知强化学习的结构化和符号推理用于大语言模型

Runxuan Liu, Xianhao Ou, Xinyan Ma, Jiyuan Wang, Jiafeng Liang, Jiaqi Li, Tao He, Zheng Chu, Rongchuan Mu, Zekun Wang, Baoxin Wang, Dayong Wu, Ming Liu, Shijin Wang, Guoping Hu, Bing Qin

机构 * Harbin Institute of Technology(哈尔滨工业大学) State Key Laboratory of Cognitive Intelligence(认知智能国家重点实验室) Tianjin Normal University(天津师范大学) Pengcheng Laboratory(鹏城实验室)

AI总结 图推理范式通过拓扑感知强化学习实现结构化和符号推理,提升大语言模型的数学推理和代码生成能力。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12952 2026-01-21 cs.RO cs.SY eess.SY

Imitation learning-based spacecraft rendezvous and docking method with Expert Demonstration

基于模仿学习的航天器对接与 docking 方法与专家示范

Shibo Shao, Dong Zhou, Guanghui Sun, Liwen Zhang, Mingxuan Jiang

机构 * Department of Control Science and Engineering, Harbin Institute of Technology(控制科学与工程系,哈尔滨工业大学) 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 本文提出基于模仿学习的航天器对接与 docking 控制框架,通过专家示范学习控制策略,提升鲁棒性和稳定性,实现准确且节能的无模型控制。

Comments 6 figures, 4 tables. Focus on 6-DOF spacecraft rendezvous and docking control using imitation learning-based control method

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12766 2026-01-21 cs.CV cs.SY eess.SY

Spatial-VLN: Zero-Shot Vision-and-Language Navigation With Explicit Spatial Perception and Exploration

空间视觉导航:具有显式空间感知和探索的零样本视觉-语言导航

Lu Yue, Yue Fan, Shiwei Lian, Yu Zhao, Jiaxin Yu, Liang Xie, Feitian Zhang

机构 * Robotics and Control Laboratory, School of Advanced Manufacturing and Robotics, and the State Key Laboratory of Turbulence and Complex Systems, Peking University(机器人与控制实验室、先进制造与机器人学院,以及湍流与复杂系统国家重点实验室,北京大学) Defense Innovation Institute, Academy of Military Sciences, Beijing(国防科技创新研究院,军事科学院,北京) Tianjin Artificial Intelligence Innovation Center, Tianjin(天津人工智能创新中心,天津) Institute of Computing and Intelligence, Harbin Institute of Technology (Shenzhen)(计算与智能学院,哈尔滨工业大学(深圳))

AI总结 Spatial-VLN通过增强空间感知和探索机制,提升零样本视觉-语言导航在复杂环境中的泛化与鲁棒性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12697 2026-01-21 cs.CV cs.CG

Fusing in 3D: Free-Viewpoint Fusion Rendering with a 3D Infrared-Visible Scene Representation

在三维中融合:基于三维红外-可见场景表示的自由视角融合渲染

Chao Yang, Deshui Miao, Chao Tian, Guoqing Zhu, Yameng Gu, Zhenyu He

机构 * School of Computer Science(计算机科学学院) Harbin Institute of Technology, Shenzhen(哈尔滨工业大学深圳研究院)

AI总结 本文提出一种基于三维红外-可见场景表示的融合框架,通过跨模态调整模块和融合损失确保融合图像保留关键特征,有效解决传统方法在复杂场景下的信息丢失问题。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.12076 2026-01-21 cs.CV cs.CL

CroBIM-V: Memory-Quality Controlled Remote Sensing Referring Video Object Segmentation

CroBIM-V: 基于内存质量控制的遥感遥感指代视频目标分割

H. Jiang, Y. Sun, Z. Dong, T. Liu, Y. Gu

机构 * School of Electronics and Information Engineering, Harbin Institute of Technology(电子信息工程学院,哈尔滨工业大学)

AI总结 本文提出MQC-SAM框架,通过内存质量控制和解耦注意力机制,解决遥感视频中目标分割的显著性弱和噪声干扰问题,实现高精度分割。

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.06909 2026-01-21 cs.CV

UDPNet: Unleashing Depth-based Priors for Robust Image Dehazing

UDPNet: 释放基于深度的先验以实现鲁棒的图像去雾

Zengyuan Zuo, Junjun Jiang, Gang Wu, Xianming Liu

机构 * School of Computer Science and Technology, Harbin Institute of Technology(计算机科学与技术学院,哈尔滨工业大学)

AI总结 UDPNet通过引入深度先验提升图像去雾性能,有效整合深度信息以提高鲁棒性和去雾效果。

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.18177 2026-01-21 cs.CV cs.LG cs.MA

Scene-Aware Vectorized Memory Multi-Agent Framework with Cross-Modal Differentiated Quantization VLMs for Visually Impaired Assistance

具有跨模态差异化量化功能的场景感知向量内存多智能体框架用于视障辅助

Xiangxiang Wang, Xuanyu Wang, YiJia Luo, Yongbin Yu, Manping Fan, Jingtao Zhang, Liyong Ren

机构 * School of Information Software Engineering, University of Electronic Science Sichuan Provincial Key Laboratory for Human Disease Gene Study, Sichuan Academy of Medical Sciences \& Sichuan Provincial People's Hospital, University of Electronic Science Faculty of Computing, Harbin Institute of Technology, Harbin, China

AI总结 本文提出了一种具有跨模态差异化量化的多智能体框架,通过减少内存消耗和提升处理效率,为视障人士提供更高效的环境感知与辅助导航支持。

Comments 28 pages,9 figures

详情

展开后加载摘要…

URL PDF HTML 收藏