Add SciSafeEval: The State-of-the-Art Benchmark for Safety Alignment of Large Language Models in Scientific Tasks

Hi Jun, We introduced SciSafeEval, the state-of-the-art benchmark for safety alignment of large language models in scientific tasks. More info could be found at https://arxiv.org/abs/2410.03769 and https://huggingface.co/datasets/Tianhao0x01/SciSafeEval . Best, Tianhao