FuzzLLM: A Novel and Universal Fuzzing Framework for Proactively Discovering Jailbreak Vulnerabilities in Large Language Models

arXiv:2309.05274v2 Announce Type: replace
Abstract: Jailbreak vulnerabilities in Large Language Models (LLMs), which exploit meticulously crafted prompts to elicit content that violates service guidelines, have captured the attention of research communities. While model owners can defend against individual jailbreak prompts through safety training strategies, this relatively passive approach struggles to handle the broader category of similar jailbreaks. To tackle this issue, we introduce FuzzLLM, an automated fuzzing framework designed to proactively test and discover jailbreak vulnerabilities in LLMs. We utilize …

arxiv attention can communities cs.cr exploit framework fuzzing guidelines jailbreak language language models large llms novel prompts research safety service strategies training vulnerabilities