Awesome Robotics Manipulation · full_paper

METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model

作者：Yankai Fu, Ning Chen, Junkai Zhao, Shaozhe Shan, Guocai Yao, Pengwei Wang, Zhongyuan Wang, Shanghang Zhang · 单位：State Key Laboratory of Multimedia Information Processing, School of Computer Science Peking University, Beijing Academy of Artificial Intelligence · 会议/期刊：arXiv · 日期：2025-11-21 · 来源：Manipulation Tasks / Dexterous Manipulation / VLA

灵巧操作视频规划视觉语言动作操作
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Project
METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model figure
AlphaXiv 中文论文页面（可滚动查看）