Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual Storytelling
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915559217561600 |
|---|---|
| author | Zhang, Deyue Yang, Dongdong Mu, Junjie Zou, Quancheng Ying, Zonghao Xu, Wenzhuo Liu, Zhao Wang, Xuan Zhang, Xiangzheng |
| author_facet | Zhang, Deyue Yang, Dongdong Mu, Junjie Zou, Quancheng Ying, Zonghao Xu, Wenzhuo Liu, Zhao Wang, Xuan Zhang, Xiangzheng |
| contents | Multimodal large language models (MLLMs) exhibit remarkable capabilities but remain susceptible to jailbreak attacks exploiting cross-modal vulnerabilities. In this work, we introduce a novel method that leverages sequential comic-style visual narratives to circumvent safety alignments in state-of-the-art MLLMs. Our method decomposes malicious queries into visually innocuous storytelling elements using an auxiliary LLM, generates corresponding image sequences through diffusion models, and exploits the models' reliance on narrative coherence to elicit harmful outputs. Extensive experiments on harmful textual queries from established safety benchmarks show that our approach achieves an average attack success rate of 83.5\%, surpassing prior state-of-the-art by 46\%. Compared with existing visual jailbreak methods, our sequential narrative strategy demonstrates superior effectiveness across diverse categories of harmful content. We further analyze attack patterns, uncover key vulnerability factors in multimodal safety mechanisms, and evaluate the limitations of current defense strategies against narrative-driven attacks, revealing significant gaps in existing protections. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_15068 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual Storytelling Zhang, Deyue Yang, Dongdong Mu, Junjie Zou, Quancheng Ying, Zonghao Xu, Wenzhuo Liu, Zhao Wang, Xuan Zhang, Xiangzheng Cryptography and Security Artificial Intelligence Multimodal large language models (MLLMs) exhibit remarkable capabilities but remain susceptible to jailbreak attacks exploiting cross-modal vulnerabilities. In this work, we introduce a novel method that leverages sequential comic-style visual narratives to circumvent safety alignments in state-of-the-art MLLMs. Our method decomposes malicious queries into visually innocuous storytelling elements using an auxiliary LLM, generates corresponding image sequences through diffusion models, and exploits the models' reliance on narrative coherence to elicit harmful outputs. Extensive experiments on harmful textual queries from established safety benchmarks show that our approach achieves an average attack success rate of 83.5\%, surpassing prior state-of-the-art by 46\%. Compared with existing visual jailbreak methods, our sequential narrative strategy demonstrates superior effectiveness across diverse categories of harmful content. We further analyze attack patterns, uncover key vulnerability factors in multimodal safety mechanisms, and evaluate the limitations of current defense strategies against narrative-driven attacks, revealing significant gaps in existing protections. |
| title | Sequential Comics for Jailbreaking Multimodal Large Language Models via Structured Visual Storytelling |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2510.15068 |