Fetching the paper…

Automatic Pseudo-Harmful Prompt Generation for Evaluating False Refusals in Large Language Models · Around