Predicting Prompt Refusal in Language Models
-
arxiv.org
Clear