The Moving Eye: Enhancing VLA Spatial Generalization via Hybrid Dynamic Data Collection figure
AlphaXiv 中文概览(可滚动查看)