Fetching the paper…

MME: A Comprehensive Evaluation Benchmark for Multimodal Large Language Models · Around