ARC：用稀疏人工纠正做机器人策略的残差修正
When, Where, and What to Correct: Human-Guided Adaptive Residual Correction for Robot Policies
机器人模仿学习到末端精度总翻车？这篇 ARC 只用几次人工纠正加 RL 微调，成功率从 33% 拉到 75%，思路挺聪明。
arXiv 论文提出 Adaptive Residual Correction（ARC），在冻结的模仿学习策略之上用强化学习训练残差修正模块。人工稀疏纠正不被当作模仿目标，而是用来判断何时激活残差控制、在残差动作空间哪个区域自主探索。在真实机器人上的三项精密操作任务中，平均任务成功率从 33.3% 提升到 75.0%。人工纠正定义的 2-3 维子空间保留了七关节残差空间中 96.0%-97.7% 的平方纠正量。
When, Where, and What to Correct: Human-Guided Adaptive Residual Correction for Robot Policies
Imitation policies that reliably execute most of a manipulation trajectory can still fail in localized regions requiring last-mile precision. Updating the full policy or exploring residual actions everywhere treats this local failure as a global learning problem. We introduce Adaptive Residual Correction (ARC), a lightweight post-training module that uses reinforcement learning (RL) to learn a residual correction on top of a frozen policy while using sparse human corrections to structure autonomous exploration. Rather than treating human corrections as imitation targets, ARC uses pre-intervention states to determine 'when' residual control should become active and correction geometry to determine 'where' in residual action space autonomous refinement should explore; task reward determines 'what' residual is ultimately executed. Across three precision-manipulation tasks on a real robot, ARC improves average task success from 33.3% to 75.0%. The human corrections define task-dependent 2-3 dimensional subspaces of the seven-joint residual space that retain 96.0-97.7% of squared correction magnitude. Together, these results show that sparse human corrections can localize and constrain autonomous post-training without requiring the observed correction itself to serve as the final action target.