QueryAttack: Jailbreaking Aligned Large Language Models Using Structured Non-natural Query Language
This addresses security risks for users of aligned LLMs by exposing vulnerabilities in safety mechanisms, though it is incremental as it builds on prior jailbreak research.
The paper tackles the problem of bypassing safety alignment in large language models by translating malicious queries into structured non-natural language, achieving high attack success rates and jailbreaking various defenses, with a tailored defense reducing ASR by up to 64% on GPT-4-1106.
Recent advances in large language models (LLMs) have demonstrated remarkable potential in the field of natural language processing. Unfortunately, LLMs face significant security and ethical risks. Although techniques such as safety alignment are developed for defense, prior researches reveal the possibility of bypassing such defenses through well-designed jailbreak attacks. In this paper, we propose QueryAttack, a novel framework to examine the generalizability of safety alignment. By treating LLMs as knowledge databases, we translate malicious queries in natural language into structured non-natural query language to bypass the safety alignment mechanisms of LLMs. We conduct extensive experiments on mainstream LLMs, and the results show that QueryAttack not only can achieve high attack success rates (ASRs), but also can jailbreak various defense methods. Furthermore, we tailor a defense method against QueryAttack, which can reduce ASR by up to $64\%$ on GPT-4-1106. Our code is available at https://github.com/horizonsinzqs/QueryAttack.