ProGAL-VLA: Grounded Alignment through Prospective Reasoning in Vision-Language-Action Models
返回详情
VLA / Vision-Language-Action 每日论文卡