Awesome Robotics Manipulation · full_paper

Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons

arXiv · 2026-03-02 · High-Level Structured Planning / Multimodal Reasoning / Reward Reasoning

任务规划多模态推理强化学习感知操作
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Code
Robometer: Scaling General-Purpose Robotic Reward Models via Trajectory Comparisons figure
AlphaXiv 中文论文页面（可滚动查看）