Awesome Robotics Manipulation · full_paper

ARM: Advantage Reward Modeling for Long-Horizon Manipulation

arXiv · 2026-04-03 · High-Level Structured Planning / Multimodal Reasoning / Reward Reasoning

任务规划多模态推理强化学习操作
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Code
ARM: Advantage Reward Modeling for Long-Horizon Manipulation figure
AlphaXiv 中文论文页面（可滚动查看）