Awesome Robotics Manipulation · full_paper

CycleVLA: Proactive Self-Correcting Vision-Language-Action Models via Subtask Backtracking and Minimum Bayes Risk Decoding

arXiv · 2026-01-05 · Low-Level Learning-Based Action Modelling / Input Modelling / 2D Vision Language Action Models with Robustness / Failure Analysis & Recovery

VLA安全机器人学习
新标签打开 AlphaXiv新标签打开 Paper新标签打开 Code
CycleVLA: Proactive Self-Correcting Vision-Language-Action Models via Subtask Backtracking and Minimum Bayes Risk Decoding figure
AlphaXiv 中文论文页面（可滚动查看）