多模态 LLM 越狱的概率建模：从量化到应用

Computer Science > Cryptography and Security arXiv:2503.06989 (cs) [Submitted on 10 Mar 2025 (v1), last revised 22 Sep 2026 (this version, v5)] Title:Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application Authors:Wenzhuo Xu, Zhipeng Wei, Xiongtao Sun, Zonghao Ying, Deyue Zhang, Dongdong Yang, Xiangzheng Zhang, Quanchen Zou View a PDF of the paper titled Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application, by Wenzhuo Xu and 7 other authors View PDF HTML (experimental) Abstract:Recently, Multimodal Large Language Models (MLLMs) have demonstrated their superior ability in understanding multimodal content. However, they remain vulnerable to jailbreak attacks, which exploit weaknesses in their safety alignment to generate harmful responses. Previous studies categorize jailbreaks as successful or failed based on whether responses contain malicious content. However, given the stochastic nature of MLLM responses, this binary classification of an input's ability to jailbreak MLLMs is inappropriate. Derived from this viewpoint, we introduce jailbreak probability to quantify the jailbreak potential of an input, which represents the likelihood that MLLMs generated a malicious response when prompted with this input. We approximate this probability through multiple queries to MLLMs. After modeling the relationship between input hidden states and their corresponding jailbreak probability using Jailbreak Probability Prediction Network (JPPN), we use continuous jailbreak probability for optimization. Specifically, we propose Jailbreak-Probability-based Attack (JPA) that optimizes adversarial perturbations on input image to maximize jailbreak probability, and further enhance it as Multimodal JPA (MJPA) by including monotonic text rephrasing. To counteract attacks, we also propose Jailbreak-Probability-based Finetuning (JPF), which minimizes jailbreak probability through MLLM parameter updates. Extensive experiments show that (1) (M)JPA yields significant improvements when attacking a wide range of models under both white and black box settings. (2) JPF vastly reduces jailbreaks by at most over 60\%. Both of the above results demonstrate the significance of introducing jailbreak probability to make nuanced distinctions among input jailbreak abilities. Comments: Subjects: Cryptography and Security (cs.CR); Computer Vision and Pattern Recognition (cs.CV) Cite as: arXiv:2503.06989 [cs.CR] (or arXiv:2503.06989v5 [cs.CR] for this version) https://doi.org/10.48550/arXiv.2503.06989 Focus to learn more arXiv-issued DOI via DataCite Submission history From: Wenzhuo Xu [view email] [v1] Mon, 10 Mar 2025 07:10:38 UTC (1,454 KB) [v2] Thu, 31 Jul 2025 10:26:35 UTC (1,341 KB) [v3] Wed, 6 Aug 2025 02:38:30 UTC (1,340 KB) [v4] Thu, 28 Aug 2025 08:43:31 UTC (1,340 KB) [v5] Tue, 22 Sep 2026 08:02:23 UTC (845 KB) Full-text links: Access Paper: View a PDF of the paper titled Probabilistic Modeling of Jailbreak on Multimodal LLMs: From Quantification to Application, by Wenzhuo Xu and 7 other authors View PDF HTML (experimental) TeX Source view license Current browse context: cs.CR new | recent | 2025-03 Change to browse by: cs cs.CV References & Citations NASA ADS Google Scholar Semantic Scholar export BibTeX citation Loading... BibTeX formatted citation × loading... Data provided by: 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 Code, Data and Media Associated with this Article alphaXiv Toggle alphaXiv (What is alphaXiv?) Links to Code Toggle CatalyzeX Code Finder for Papers (What is CatalyzeX?) DagsHub Toggle DagsHub (What is DagsHub?) GotitPub Toggle Gotit.pub (What is GotitPub?) Huggingface Toggle Hugging Face (What is Huggingface?) ScienceCast Toggle ScienceCast (What is ScienceCast?) Demos Demos Replicate Toggle Replicate (What is Replicate?) Spaces Toggle Hugging Face Spaces (What is Spaces?) Spaces Toggle TXYZ.AI (What is TXYZ.AI?) Related Papers Recommenders and Search Tools Link to Influence Flower Influence Flower (What are Influence Flowers?) Core recommender toggle CORE Recommender (What is CORE?) Author Venue Institution Topic About arXivLabs arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)