cs.CVSep 30, 2026

When the Editing Intent Is Split: A Cross-Modal Jailbreak Attack for Large Image Editing Models

Authors: Xianmin Ye, Zhiyuan Fan, Tianju Liu, Yuhang Wang, Zhenxing Niu, Haichang Gao

Organizations: School of Computer Science and Technology, Xidian University

Abstract

Large image editing models increasingly infer editing intent from both textual and visual inputs, creating an underexplored safety risk when harmful intent is distributed across modalities. In this work, we formalize this split-intent vulnerability and propose the Split-Intent Jailbreak Attack (SIJA), which asymmetrically distributes an unsafe editing intent by specifying the editing referent in text while concealing the safety-critical action and specification through graphical visual semantics in an auxiliary cue. To systematically study this threat, we introduce the Split-Intent Image Editing Safety Benchmark (SI2ESBench) for evaluating image editing models under split-intent inputs. Extensive experiments on representative commercial and open-source models demonstrate that SIJA effectively compromises state-of-the-art image editing systems, achieving ASRs of 68.28% on GPT Image 2 and 91.38% on Qwen-Image-3.0-Pro. Our results expose a mismatch between multimodal editing and safety assessment: the distributed intent can be recovered to execute the requested transformation, yet its harmfulness often remains undetected by existing safeguards. To mitigate this vulnerability, we propose editing-intent reconstruction, a defense principle that explicitly evaluates the complete transformation implied by the joint multimodal context and substantially improves rejection of split-intent attacks. Our findings expose a cross-modal safety gap in modern image editing systems and highlight the need to assess the jointly inferred transformation rather than individual inputs in isolation. Project page: https://github.com/yexianmin/SIJA.

Figures & tables

Appendix figures & tables14 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models

    Feb 10, 2026Jiacheng Hou, Yining Sun, Ruochong Jin +4Jailbreak AttacksMultimodal Jailbreak Attacks

  2. When Think-with-Image Meets Safety: What Determines Multimodal Jailbreak Robustness?

    May 27, 2026Yuan Tian, Bing Hu, Fang Wu +3VLM RobustnessTool Use in VLMs

  3. VETO: Towards Protecting Images From Frontier AI Editing

    Jul 29, 2026Jonas Grebe, Hossein Shakibania, Tobias Braun +2Image Editing EvaluationAdversarial Examples