Metaphor-based Jailbreaking Attacks on Text-to-Image Models

📝 Original Info

• Title: Metaphor-based Jailbreaking Attacks on Text-to-Image Models
• ArXiv ID: 2512.10766
• Date: 2025-12-06
• Authors: Chenyu Zhang, Yiwen Ma, Lanjun Wang, Wenhui Li, Yi Tu, An-An Liu

📝 Abstract

Text-to-image (T2I) models commonly incorporate defense mechanisms to prevent the generation of sensitive images. Unfortunately, recent jailbreaking attacks have shown that adversarial prompts can effectively bypass these mechanisms and induce T2I models to produce sensitive content, revealing critical safety vulnerabilities. However, existing attack methods implicitly assume that the attacker knows the type of deployed defenses, which limits their effectiveness against unknown or diverse defense mechanisms. In this work, we introduce MJA, a metaphor-based jailbreaking attack method inspired by the Taboo game, aiming to effectively and efficiently attack diverse defense mechanisms without prior knowledge of their type by generating metaphor-based adversarial prompts. Specifically, MJA consists of two modules: an LLM-based multi-agent generation module (MLAG) and an adversarial prompt optimization module (APO). MLAG decomposes the generation of metaphor-based adversarial prompts into three subtasks: metaphor retrieval, context matching, and adversarial prompt generation. Subsequently, MLAG coordinates three LLM-based agents to generate diverse adversarial prompts by exploring various metaphors and contexts. To enhance attack effi...

📄 Full Content

...(본문 내용이 길어 생략되었습니다. 사이트에서 전문을 확인해 주세요.)