ControlVLA: Few-shot Object-centric Adaptation for Pre-trained Vision-Language-Action Models

返回详情 VLA / Vision-Language-Action 每日论文卡