الملخص
Concept erasure removes copyright-protected, privacy-sensitive, or otherwise undesirable concepts from pretrained text-to-image diffusion models to support content governance and compliance. As erasure requests arrive over time, models must remove new targets without undoing prior erasures. Existing methods do not constrain interference across edits: residual perturbations outside the retain set interact and accumulate, degrading unrelated generations and sometimes collapsing previously erased targets into noise. We propose CEASE (Continual Erasure via Adaptive Subspace Editing), a training-free method that imposes two subspace constraints on a closed-form solver. CEASE adds the token representation of the shared replacement to the solver's invariance matrix and, when interference is detected, projects the current update onto the orthogonal complement of dominant output directions extracted from cumulative past updates. A closed-form decomposition attributes the accumulated interference to repeated activation of the shared replacement and overlap between successive update directions, showing that the two constraints suppress these respective sources. Across continual erasure of celebrities, artistic styles, and instances, CEASE achieves the most consistent erase-preserve trade-off, while existing methods either degrade general generation or insufficiently erase targets.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Wu, Y., Lu, H., Luo, J., cao, W., Zhu, X., & Liu, Y. (2026). Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference. https://omanscience.com/ar/articles/continual-concept-erasure-in-diffusion-models-by-suppressing-cross-edit-interference
MLA 9
Wu, Yongliang, et al. "Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference." https://omanscience.com/ar/articles/continual-concept-erasure-in-diffusion-models-by-suppressing-cross-edit-interference.
شيكاغو (المؤلف–التاريخ)
Wu, Yongliang, Haori Lu, Jinqi Luo, Wei cao, Xingyu Zhu, and Yaoyao Liu. 2026. "Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference." https://omanscience.com/ar/articles/continual-concept-erasure-in-diffusion-models-by-suppressing-cross-edit-interference.
هارفارد
Wu, Y., Lu, H., Luo, J., cao, W., Zhu, X. and Liu, Y. (2026) 'Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference', Available at: https://omanscience.com/ar/articles/continual-concept-erasure-in-diffusion-models-by-suppressing-cross-edit-interference.
فانكوفر
Wu Y, Lu H, Luo J, cao W, Zhu X, Liu Y. Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference. https://omanscience.com/ar/articles/continual-concept-erasure-in-diffusion-models-by-suppressing-cross-edit-interference
IEEE
Y. Wu, H. Lu, J. Luo, W. cao, X. Zhu, and Y. Liu, "Continual Concept Erasure in Diffusion Models by Suppressing Cross-Edit Interference," https://omanscience.com/ar/articles/continual-concept-erasure-in-diffusion-models-by-suppressing-cross-edit-interference.