Precise, Fast, and Low-cost Concept Erasure in Value Space: Orthogonal Complement Matters
Yuan Wang, Ouxiang Li, Tingting Mu, Yanbin Hao, Kuien Liu, Xiang Wang, Xiangnan He
Abstract
Single-Concept Erasure Multi-Concept Erasure Original Image (a) Erasure Efficacy and Prior Preservation (b) Transferability Figure 1. The proposed Adaptive Value Decomposer (AdaVD) demonstrates a satisfactory balance between erasure efficacy and prior preservation and an effective transferability across T2I diffusion models. (a) Compared to SLD [39], AdaVD enables precise concept erasure without compromising prior knowledge for non-target concepts at both single-and multi-concept erasure. This is facilitated by a precise disentanglement of target semantics (e.g., "Snoopy") and a robust preservation of non-target ones (e.g., "Teddy"), with visualization interpretation marked by . (b) AdaVD can be transferred to various T2I models, e.g., SDXL [31], DreamShaper [6], Chilloutmix [5].
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
3 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- High-Resolution Image Synthesis with Latent Diffusion Models cites Denoising Diffusion Probabilistic Models
- Improved Denoising Diffusion Probabilistic Models cites Denoising Diffusion Probabilistic Models
- Adding Conditional Control to Text-to-Image Diffusion Models cites High-Resolution Image Synthesis with Latent Diffusion Models
- Adding Conditional Control to Text-to-Image Diffusion Models cites Improved Denoising Diffusion Probabilistic Models
- Adding Conditional Control to Text-to-Image Diffusion Models cites Learning Transferable Visual Models From Natural Language Supervision
- Concept Removal for Frontier Image Generative Models cites High-Resolution Image Synthesis with Latent Diffusion Models
- SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoders cites High-Resolution Image Synthesis with Latent Diffusion Models
- Orthogonal Concept Erasure for Diffusion Models cites High-Resolution Image Synthesis with Latent Diffusion Models
- Orthogonal Concept Erasure for Diffusion Models cites Denoising Diffusion Probabilistic Models
Cited by top-tier papers3
- SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse AutoencodersEnrico Cassano, Riccardo Renzulli, Marco Nurisso, Mirko Zaffaroni et al.ICML 2026 · 7 citations
- Concept Removal for Frontier Image Generative ModelsAditya Kumar, Pierre Joly, Adam Dziedzic, Franziska BoenischICML 2026
- Orthogonal Concept Erasure for Diffusion ModelsYuhao Sun, Lingyun Yu, Hao-Xiang Xu, Fengyuan Miao et al.ICML 2026
Builds on21
- Learning Transferable Visual Models From Natural Language SupervisionAlec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh et al.ICML 2021 · 47,906 citations
- Denoising Diffusion Probabilistic ModelsJonathan Ho, Ajay Jain, Pieter AbbeelNeurIPS 2020 · 35,902 citations
- High-Resolution Image Synthesis with Latent Diffusion ModelsRobin Rombach, Andreas Blattmann, Dominik Lorenz, Patrick Esser et al.CVPR 2022 · 13,123 citations
- Adding Conditional Control to Text-to-Image Diffusion ModelsLvmin Zhang, Anyi Rao, Maneesh AgrawalaICCV 2023 · 6,759 citations
- Improved Denoising Diffusion Probabilistic ModelsAlexander Quinn Nichol, Prafulla DhariwalICML 2021 · 5,234 citations
Related papers
- One-dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing ApplicationsMengyao Lyu, Yuhong Yang, Haiwen Hong, Hui Chen et al.CVPR 2024 · 16 citations
- Where Concept Erasure Should Occur: Concept–Layer Alignment in Text-to-Video Diffusion ModelsYiwei Xie, Ping Liu, Zheng ZhangICML 2026
- ICE: Intercede Concept Erasure in Text-to-Image Diffusion ModelsYizhou Lin, Nisha Huang, Kaer Huang, Henglin Liu et al.ACM MM 2025 · 1 citation
- Now You See It, Now You Don't: Instant Concept Erasure for Safe Text-to-Image and Video GenerationShristi Das Biswas, Arani Roy, Kaushik RoyCVPR Findings 2026
- Z-Erase: Enabling Concept Erasure in Single Stream Diffusion TransformersNanxiang Jiang, Zhaoxin Fan, Baisen Wang, Daiheng Gao et al.ICML 2026 · 2 citations
