FlipAttack: Jailbreak LLMs via Flipping

Computer Science > Cryptography and Security [Submitted on 2 Oct 2024 (v1), last revised 15 May 2026 (this version, v2)] FlipAttack: Jailbreak LLMs via Flipping Yue Liu, Xiaoxin He, Miao Xiong, Jinlan Fu, Shumin Deng, Yingwei Ma, Jiaheng Zhang, Bryan Hooi This paper proposes a simple yet effective jailbreak attack named FlipAttack against black-box LLMs. First, from the autoregressive nature, we reveal that LLMs tend to understand the text from left to right and find that they struggle to comprehend the text when noise is added to the left side. Motivated by these insights, we propose to disguise the harmful prompt by constructing left-side noise merely based on the prompt itself, then generalize this idea to 4 flipping modes. Second, we verify the strong ability of LLMs to perform the text-flipping task, and then develop 4 variants to guide LLMs to denoise, understand, and execute harmful behaviors accurately. These designs keep FlipAttack universal, stealthy, and simple, allowing it to jailbreak black-box LLMs within only 1 query. Experiments on 8 LLMs demonstrate the superiority of FlipAttack. Remarkably, it achieves \sim98\% attack success rate on GPT-4o, and \sim98\% bypass rate against 5 guardrail models on average. The codes are available at GitHub\footnote{this https URL}. Comments: 43 pages, 31 figures Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI) Cite as: arXiv:2410.02832 [cs.CR]   (or arXiv:2410.02832v2 [cs.CR] for this version)   https://doi.org/10.48550/arXiv.2410.02832 Focus to learn more Submission history From: Yue Liu [view email] [v1] Wed, 2 Oct 2024 08:41:23 UTC (3,591 KB) [v2] Fri, 15 May 2026 07:55:39 UTC (3,297 KB) Access Paper: HTML (experimental) view license Current browse context: cs.CR < prev   |   next > new | recent | 2024-10 Change to browse by: cs cs.AI References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)