AnyBreak: Towards Jailbreaking Any T2I Model for Any Concept
cs.CV
Submitted: 2026-11-09
Updated: 2026-11-09
Comments: Accepted for poster presentation at ICME 2026
Code: https://github.com/alex000kim/nsfw
License: http://creativecommons.org/licenses/by/4.0/
The gist: Text-to-image (T2I) models have demonstrated remarkable capabilities in image generation and editing, yet this progress has also introduced significant security risks.
Terminology
Abstract
Text-to-image (T2I) models have demonstrated remarkable capabilities in image generation and editing, yet this progress has also introduced significant security risks. Prior work has shown that adversarial prompt can induce the generation of Not-Safe-For-Work (NSFW) content, including pornographic, violent, and graphic imagery. However, such attacks are typically mitigated by prompt filters and often rely on white-box assumptions. In this paper, we propose AnyBreak, a novel jailbreak framework that constructs adversarial images via multi-scale progressive semantic alignment. During the generation process, these adversarial images steer the generation trajectory in latent space towards unsafe data manifolds, enabling cross-modal transfer of malicious concepts from the text modality to the image modality. AnyBreak operates using only clean textual prompts and adversarial images at inference time, and can jailbreak any T2I models towards any malicious concepts in a fully black-box setting. We further introduce AnyBreak-V, an enhanced variant that incorporates a compensation term to bypass post-hoc safety checkers. Extensive experiments demonstrate that AnyBreak achieves high attack success rates while preserving visual fidelity and semantic consistency with the input prompts under black-box and clean-text constraints. Our findings reveal that existing T2I defense leave critical cross-modal attack surfaces insufficiently addressed. The code is available at https://github.com/luqi-glory/AnyBreak.
Sources
- Imperceptible and Reversible Adversarial Examples against Vision-Language Models for Privacy Protection
- A Dynamic Prognostic Prediction Method for Colorectal Cancer Liver Metastasis
- 4D-ACFNet: A 4D Attention Mechanism-Based Prognostic Framework for Colorectal Cancer Liver Metastasis Integrating Multimodal Spatiotemporal Features
- ActDistill: General Action-Guided Self-Derived Distillation for Efficient Vision-Language-Action Models
- Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?
- AdvI2I: Adversarial Image Attack on Image-to-Image Diffusion models
- Universal Prompt Optimizer for Safe Text-to-Image Generation
- Red-Teaming the Stable Diffusion Safety Filter
- Detecting Language Model Attacks with Perplexity
- Score-Based Generative Modeling through Stochastic Differential Equations
- Z-Image: An Efficient Image Generation Foundation Model with Single-Stream Diffusion Transformer
- Explaining and Harnessing Adversarial Examples
- Adversarial Perturbations Cannot Reliably Protect Artists From Generative AI
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs