Reducing Temporal Redundancy for Efficient Vision-Language-Action Inference figure
AlphaXiv 中文概览(可滚动查看)