Multi-View Relational Distillation for Spatial Reasoning with Vision-Language Models figure
AlphaXiv 中文概览(可滚动查看)