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