FM-VLA: Force-based Memory for Vision-Language-Action Models in Contact-Rich Manipulation figure
AlphaXiv 中文概览(可滚动查看)