Acting While Understanding: Asynchronous Semantic-Action Decoupling for Real-Time Vision-Language-Action Models figure
AlphaXiv 中文概览(可滚动查看)