JEPA-VLA: Video Predictive Embedding is Needed for VLA Models
返回详情
VLA / Vision-Language-Action 每日论文卡