REVIEW 7 cited by
Erasing Undesirable Concepts in Diffusion Models with Adversarial Preservation
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
Diffusion models excel at generating visually striking content from text but can inadvertently produce undesirable or harmful content when trained on unfiltered internet data. A practical solution is to selectively removing target concepts from the model, but this may impact the remaining concepts. Prior approaches have tried to balance this by introducing a loss term to preserve neutral content or a regularization term to minimize changes in the model parameters, yet resolving this trade-off remains challenging. In this work, we propose to identify and preserving concepts most affected by parameter changes, termed as \textit{adversarial concepts}. This approach ensures stable erasure with minimal impact on the other concepts. We demonstrate the effectiveness of our method using the Stable Diffusion model, showing that it outperforms state-of-the-art erasure methods in eliminating unwanted content while maintaining the integrity of other unrelated elements. Our code is available at https://github.com/tuananhbui89/Erasing-Adversarial-Preservation.
Forward citations
Cited by 7 Pith papers
-
AutoAnchor: Stable Diffusion Unlearning Using Cross-Attention as a Manifold Surrogate
Cross-attention maps serve as a tractable surrogate for manifold proximity, enabling automatic synthesis of anchors that suppress normal-space drift in diffusion unlearning.
-
SuMa: A Subspace Mapping Approach for Robust and Effective Concept Erasure in Text-to-Image Diffusion Models
SuMa erases narrow concepts from text-to-image models by mapping the concept's token subspace onto a nearby reference subspace, achieving robustness against adversarial attacks with image quality close to standard era...
-
Semantic Steering for Controllable Generation: Tuning-Free Concept Erasure in Multimodal Diffusion Transformers
A single steering vector extracted from the middle text layer of a multimodal diffusion transformer can erase concepts like nudity, celebrities, and art styles at inference time, with no training.
-
Erase but Preserve: Controllable Removal of Copyrighted Animation Characters via Optimized Semantic Anchors
Optimized continuous embedding anchors with structure-aware adaptive replacement erase copyrighted animation characters from text-to-image diffusion outputs while preserving image context.
-
A Unified Framework for Diffusion Model Unlearning with f-Divergence
Diffusion model unlearning is generalized from KL/MSE to any f-divergence, with closed-form Hellinger and chi-square losses and a variational min-max form.
-
Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Concepts
ANT erases unwanted concepts from text-to-image diffusion models by reversing the guidance direction only during mid-to-late denoising, preserving early structural quality.
-
TRACE: Trajectory-Constrained Concept Erasure in Diffusion Models
TRACE combines a closed-form cross-attention nullification with a late-timestep fine-tuning loss to erase concepts from diffusion models, claiming better erasure and fidelity than published baselines.
Discussion (0). Continue with ORCID to comment.