Search Results for author:
Found 2 papers, 2 papers with code
Emulated Disalignment: Safety Alignment for Large Language Models May Backfire!
Large language models (LLMs) need to undergo safety alignment to ensure safe conversations with humans.
Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey
Large Language Models (LLMs) are now commonplace in conversation applications.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
