Awesome Robotics Manipulation · full_paper

SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning

作者：Philip Schroeder, Thomas Weng, Karl Schmeckpeper, Eric Rosen, Stephen Hart, Ondrej Biza · 单位：MIT, RAI Institute · 会议/期刊：arXiv · 日期：2026-03-30 · 来源：High-Level Structured Planning / Multimodal Reasoning / Reward Reasoning

任务规划多模态推理视频规划强化学习触觉
新标签打开 AlphaXiv新标签打开 Paper
SOLE-R1: Video-Language Reasoning as the Sole Reward for On-Robot Reinforcement Learning figure
AlphaXiv 中文论文页面（可滚动查看）