Fetching the paper…

MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models · Around