Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes

arXiv:2403.00867v1 Announce Type: cross
Abstract: Large Language Models (LLMs) are becoming a prominent generative AI tool, where the user enters a query and the LLM generates an answer. To reduce harm and misuse, efforts have been made to align these LLMs to human values using advanced training techniques such as Reinforcement Learning from Human Feedback (RLHF). However, recent studies have highlighted the vulnerability of LLMs to adversarial jailbreak attempts aiming at subverting the embedded safety guardrails. To address this challenge, …

abstract advanced ai tool arxiv attacks cs.ai cs.cl cs.cr cs.lg generative gradient harm human jailbreak language language models large language large language models llm llms loss misuse query reduce tool type values