Reasoning-aware Speculative Decoding for Efficient Vision-Language-Action Models in Autonomous Driving figure
AlphaXiv 中文概览(可滚动查看)