FrameSkip: Learning from Fewer but More Informative Frames in VLA Training
返回详情
VLA / Vision-Language-Action 每日论文卡