Attacks, Defenses and Evaluations for LLM Conversation Safety: A Survey
It addresses safety concerns for users and society in conversational AI, but is incremental as it synthesizes existing work.
This survey tackles the problem of harmful response generation in Large Language Models (LLMs) used in conversation applications by providing a comprehensive overview of recent studies on attacks, defenses, and evaluations, aiming to enhance understanding and encourage further research.
Large Language Models (LLMs) are now commonplace in conversation applications. However, their risks of misuse for generating harmful responses have raised serious societal concerns and spurred recent research on LLM conversation safety. Therefore, in this survey, we provide a comprehensive overview of recent studies, covering three critical aspects of LLM conversation safety: attacks, defenses, and evaluations. Our goal is to provide a structured summary that enhances understanding of LLM conversation safety and encourages further investigation into this important subject. For easy reference, we have categorized all the studies mentioned in this survey according to our taxonomy, available at: https://github.com/niconi19/LLM-conversation-safety.