Orthogonal Concept Erasure for Diffusion Models
This paper introduces Orthogonal Concept Erasure (OCE), which uses multiplicative parameter updates for precise concept removal while preserving generative capacity, supporting multi-concept erasure with high speed.
Article intelligence
Key points
- Existing editing-based methods rely on additive updates that interfere with generative capacity.
- OCE uses orthogonal transformations as multiplicative updates, preserving neuron direction and angular geometry.
- OCE introduces a subspace-level objective for multi-concept erasure, achieving superior performance.
Why it matters
This matters because existing editing-based methods rely on additive updates that interfere with generative capacity.
Technical impact
May affect model selection, inference cost, product capability, and evaluation benchmarks.
[2605.28902] Orthogonal Concept Erasure for Diffusion Models
[Submitted on 27 May 2026]
Title:Orthogonal Concept Erasure for Diffusion Models
View a PDF of the paper titled Orthogonal Concept Erasure for Diffusion Models, by Yuhao Sun and 5 other authors
View PDF HTML (experimental)
Abstract:Concept erasure has emerged as a promising approach to mitigate undesired or unsafe content in diffusion models, yet existing methods still face significant limitations. While training-based methods are effective, their high computational cost limits scalability. Editing-based methods are more efficient and deployment-friendly, yet they struggle to simultaneously achieve precise concept erasure and preserve overall generative capacity. We identify this core limitation of the editing-based methods as reliance on additive parameter updates. Our empirical analysis reveals that concept semantics primarily depend on neuron direction rather than neuron magnitude, while overall generative capacity relies on the angular geometry of neurons. As additive updates inherently entangle direction, magnitude, and angular geometry, they inevitably introduce unintended interference between concept erasure and overall generation performance. To address this, we propose Orthogonal Concept Erasure (OCE), which reformulates editing-based erasure as multiplicative parameter updates from a geometric perspective. Specifically, OCE applies layer-wise orthogonal transformations derived from a closed-form solution to the parameters, enabling precise concept erasure while preserving the neuron magnitude and angular geometry. Furthermore, to address conflicting constraints in multi-concept erasure, OCE introduces a subspace-level objective with structured subspace manipulation, yielding a more effective and scalable erasure. Extensive experiments on single- and multi-concept erasure demonstrate that OCE outperforms existing methods in concept erasure and non-target preservation, erasing up to 100 concepts in 4.3 s. Code: this https URL.
Comments: Accepted by ICML 2026 Oral
Subjects:
Artificial Intelligence (cs.AI)
Cite as: arXiv:2605.28902 [cs.AI]
(or arXiv:2605.28902v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2605.28902
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Yuhao Sun [view email] [v1] Wed, 27 May 2026 15:58:20 UTC (18,125 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Orthogonal Concept Erasure for Diffusion Models, by Yuhao Sun and 5 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.AI
new | recent | 2026-05
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)