SeeTraceAct: Visibility-Aware Latent Planning from Cross-Embodiment Demonstration Videos

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