arXiv preprint / cs.CV

VAMR: Multi-Question Agentic Reasoning for Efficient Long-Form Video Understanding

作者：Runquan Gui, Hanzhu Chen, Zehao Wang, Hanxin Zhu, Xin Li, Zhibo Chen · 单位：University of Science and Technology of China · 会议/期刊：arXiv preprint · 方向：cs.CV · 发布日期：2026-10-09

2026-10-09视觉感知
新标签打开 AlphaXiv
VAMR: Multi-Question Agentic Reasoning for Efficient Long-Form Video Understanding figure
AlphaXiv 中文概览（可滚动查看）