
HX
Hong Xi Tae, Jiaming Zhang, Wenwen He, Xuan Wang, Wei Yang Bryan Lim
· 1 min read
ResearcharXiv cs.CV
Concept Score Relearning: A Unified Cross-Architecture Attack on Concept Erasure
arXiv:2609.33445v1 Announce Type: new
Abstract: Concept erasure aims to suppress undesirable knowledge in text-to-image generative models. However, existing robustness evaluations typically rely on relearning attacks tailored to specific model architectures. We study concept reactivation across two substantially different generative paradigms: noise-prediction U-Nets and flow-matching Transformers. We introduce \textbf{Concept Score Relearning (CSR)}, a unified parameter-level framework that reactivates erased concepts by optimizing each model within its native prediction space. CSR requires no external target-concept image dataset and applies the same concept-directed objective to both U-Net-based Stable Diffusion and Transformer-based FLUX. Experiments across diverse concepts and multiple erasure methods demonstrate consistent concept reactivation across both architectures, highlighting the cross-architecture applicability of CSR and the persistent recoverability of apparently erased concepts. For strict nudity, CSR reaches average ASRs of 50.47\% on FLUX and 40.29\% on Stable Diffusion, consistently ranking first across all evaluated safety settings.
Original source
This story was published by arXiv cs.CV and written by Hong Xi Tae, Jiaming Zhang, Wenwen He, Xuan Wang, Wei Yang Bryan Lim. SyncAI.news shows a preview; the complete article is on the publisher's site.
Read the full story on arxiv.org


