ConLA: Contrastive Latent Action Learning from Human Videos for Robotic Manipulation

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