DrAttack: Prompt Decomposition and Reconstruction Makes Powerful LLM Jailbreakers

arXiv:2402.16914v1 Announce Type: new
Abstract: The safety alignment of Large Language Models (LLMs) is vulnerable to both manual and automated jailbreak attacks, which adversarially trigger LLMs to output harmful content. However, current methods for jailbreaking LLMs, which nest entire harmful prompts, are not effective at concealing malicious intent and can be easily identified and rejected by well-aligned LLMs. This paper discovers that decomposing a malicious prompt into separated sub-prompts can effectively obscure its underlying malicious intent by presenting it in …

alignment arxiv attacks automated can cs.ai cs.cl cs.cr current intent jailbreak jailbreaking language language models large llm llms malicious nest prompt prompts safety trigger vulnerable