Awesome Robotics Manipulation · full_paper

DreamPlan: Efficient Reinforcement Fine-Tuning of Vision-Language Planners via Video World Models

作者：Emily Yue-Ting Jia, Weiduo Yuan, Tianheng Shi, Vitor Guizilini, Jiageng Mao, Yue Wang · 单位：USC Physical Superintelligence Lab, Toyota Research Institute · 会议/期刊：arXiv · 日期：2026-03-17 · 来源：High-Level Structured Planning / Video-based Planning

任务规划视频规划强化学习触觉世界模型
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Project
DreamPlan: Efficient Reinforcement Fine-Tuning of Vision-Language Planners via Video World Models figure
AlphaXiv 中文论文页面（可滚动查看）