الملخص
Concept erasure is essential for the safe deployment of text-to-image diffusion models, as they may reproduce harmful, copyrighted, or privacy-sensitive content learned from unconstrained large-scale data. Existing methods typically erase unwanted concepts while preserving general generation capability by redirecting target-related text-to-image mappings. However, recent studies show that erased models may still retain visual generative trajectories of target concepts, leaving them vulnerable to adversarial recovery attacks and revealing a fundamental gap between redirecting text-to-image mappings and truly removing visual knowledge. To bridge this gap, we propose VisualErase, a new paradigm that redirects concept-bearing visual generative trajectories toward explicitly defined concept-removed outcomes. To enable this redirection, we use structure-preserving image editing to construct content-aligned, concept-removed counterparts for source images, providing explicit visual endpoints that retain non-target content. We then derive a denoising target from each source-to-counterpart pair and use a dual-branch redirection loss to align both text-conditioned and unconditional predictions with this target, since conditional supervision alone does not explicitly constrain generation without textual guidance. To mitigate the adverse effects of concept erasure on non-target generation, we jointly optimize the redirection loss with a counterpart retention loss that matches denoising predictions from the frozen pretrained model. Across style, celebrity, and nudity erasure, VisualErase limits the maximum attack success rate over seven attacks to 0%, 8%, and 0.1%, respectively, while retaining general generation quality. These results highlight the importance of visual trajectory redirection for robust concept erasure beyond text-to-image mappings alone.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Xiang, Q., Zhang, M., Wang, K., Hu, Y., Hou, J., & Nie, L. (2026). VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models. https://omanscience.com/ar/articles/visualerase-dual-branch-visual-trajectory-redirection-for-robust-concept-erasure-in-text-to-image-diffusion-models
MLA 9
Xiang, Qianlong, et al. "VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models." https://omanscience.com/ar/articles/visualerase-dual-branch-visual-trajectory-redirection-for-robust-concept-erasure-in-text-to-image-diffusion-models.
شيكاغو (المؤلف–التاريخ)
Xiang, Qianlong, Miao Zhang, Kun Wang, Yupeng Hu, Junhui Hou, and Liqiang Nie. 2026. "VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models." https://omanscience.com/ar/articles/visualerase-dual-branch-visual-trajectory-redirection-for-robust-concept-erasure-in-text-to-image-diffusion-models.
هارفارد
Xiang, Q., Zhang, M., Wang, K., Hu, Y., Hou, J. and Nie, L. (2026) 'VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models', Available at: https://omanscience.com/ar/articles/visualerase-dual-branch-visual-trajectory-redirection-for-robust-concept-erasure-in-text-to-image-diffusion-models.
فانكوفر
Xiang Q, Zhang M, Wang K, Hu Y, Hou J, Nie L. VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models. https://omanscience.com/ar/articles/visualerase-dual-branch-visual-trajectory-redirection-for-robust-concept-erasure-in-text-to-image-diffusion-models
IEEE
Q. Xiang, M. Zhang, K. Wang, Y. Hu, J. Hou, and L. Nie, "VisualErase: Dual-Branch Visual Trajectory Redirection for Robust Concept Erasure in Text-to-Image Diffusion Models," https://omanscience.com/ar/articles/visualerase-dual-branch-visual-trajectory-redirection-for-robust-concept-erasure-in-text-to-image-diffusion-models.