ST-VLA: Enabling 4D-Aware Spatiotemporal Understanding for General Robot Manipulation

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