Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos
返回详情
VLA / Vision-Language-Action 每日论文卡