EmbodiedBench: Comprehensive Benchmarking Multi-modal Large Language Models for Vision-Driven Embodied Agents figure
AlphaXiv 中文论文页面(可滚动查看)