arXivDaily arXiv每日学术速递 周一至周五更新

AI 大模型

视频大模型

视频理解、视频生成、视频语言模型和时序视觉推理。

2025-10-27 至 2025-10-27 共收录 2 信号源:cs.CV, eess.IV, cs.MM

1. 长视频与时序推理 2 篇

2506.15745 2025-10-27 eess.IV cs.LG 83%

InfiniPot-V: Memory-Constrained KV Cache Compression for Streaming Video Understanding

Minsoo Kim, Kyuhong Shim, Jungwook Choi, Simyung Chang

专题命中 长视频与时序推理 :video understanding(title,abstract);long video(abstract);分类 eess.IV

Comments NeurIPS 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.06485 2025-10-27 cs.CV cs.AI cs.CL 79%

Video-RTS: Rethinking Reinforcement Learning and Test-Time Scaling for Efficient and Enhanced Video Reasoning

Ziyang Wang, Jaehong Yoon, Shoubin Yu, Md Mohaiminul Islam, Gedas Bertasius, Mohit Bansal

机构 * UNC Chapel Hill(北卡罗来纳大学教堂山分校) Nanyang Technological University(南洋理工大学)

专题命中 长视频与时序推理 :video reasoning(title,abstract);分类 cs.CV

Comments EMNLP 2025. The first two authors contributed equally. Project page: https://sites.google.com/cs.unc.edu/videorts2025/

详情

展开后加载摘要…

URL PDF HTML 收藏