TAG: Target-Agnostic Guidance for Stable Object-Centric Inference in Vision-Language-Action Models

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