Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos

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