mlsanityData to 7 Oct 2026

Erasing Undesirable Influence in Diffusion Models

arXiv 2401.05779 ↗First posted 11 Jan 2024cs.CVcited by 52

On the map: Diffusion models → Robustness, privacy and adversarial aspects

Abstract (arXiv)

Diffusion models are highly effective at generating high-quality images but pose risks, such as the unintentional generation of NSFW (not safe for work) content. Although various techniques have been proposed to mitigate unwanted influences in diffusion models while preserving overall performance, achieving a balance between these goals remains challenging. In this work, we introduce EraseDiff, an algorithm designed to preserve the utility of the diffusion model on retained data while removing the unwanted information associated with the data to be forgotten. Our approach formulates this task as a constrained optimization problem using the value function, resulting in a natural first-order algorithm for solving the optimization problem. By altering the generative process to deviate away from the ground-truth denoising trajectory, we update parameters for preservation while controlling constraint reduction to ensure effective erasure, striking an optimal trade-off. Extensive experiments and thorough comparisons with state-of-the-art algorithms demonstrate that EraseDiff effectively preserves the model's utility, efficacy, and efficiency.

Key facts. Cited by 52 papers in the map (data through 7 Oct 2026). Placed in Robustness, privacy and adversarial aspects, one of 14 directions in Diffusion models. Builds on 52 of its 74 references that are in the map. First posted on arXiv 11 Jan 2024.

Builds on

52 of 74 references.

Closest ancestors · 12

Same direction, newest first

  1. 2023One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applications26 Dec
  2. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation19 Oct
  3. 2023Unified Concept Editing in Diffusion Models25 Aug
  4. 2023Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models30 Mar
  5. 2023Ablating Concepts in Text-to-Image Diffusion Models23 Mar
  6. 2023Erasing Concepts from Diffusion Models13 Mar
  7. 2023TrojDiff: Trojan Attacks on Diffusion Models with Diverse Targets10 Mar
  8. 2023Raising the Cost of Malicious AI-Powered Image Editing13 Feb
  9. 2023Extracting Training Data from Diffusion Models30 Jan
  10. 2022Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models9 Nov
  11. 2022Differentially Private Diffusion Models18 Oct
  12. 2022Red-Teaming the Stable Diffusion Safety Filter3 Oct

Other references · 40

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2020Denoising Diffusion Implicit Modelscited by 9,655
  5. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,861
  6. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  7. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  8. 2016Deep Learning with Differential Privacycited by 2,682
  9. 2020Extracting Training Data from Large Language Modelscited by 2,144
  10. 2019Machine Unlearningcited by 1,020
  11. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 925
  12. 2022Quantifying Memorization Across Neural Language Modelscited by 755
28 more
  1. 2019Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networkscited by 599
  2. 2019Meta-Learning with Implicit Gradientscited by 523
  3. 2019Certified Data Removal from Machine Learning Modelscited by 512
  4. 2019Making AI Forget You: Data Deletion in Machine Learningcited by 447
  5. 2021Conflict-Averse Gradient Descent for Multi-task Learningcited by 441
  6. 2022Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Modelscited by 418
  7. 2018Assessing Generative Models via Precision and Recallcited by 401
  8. 2021Remember What You Want to Forget: Algorithms for Machine Unlearningcited by 360
  9. 2015Gradient-based Hyperparameter Optimization through Reversible Learningcited by 357
  10. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  11. 2022Memorization Without Overfitting: Analyzing the Training Dynamics of Large Language Modelscited by 218
  12. 2021Fast Yet Effective Machine Unlearningcited by 209
  13. 2023Understanding and Mitigating Copying in Diffusion Modelscited by 205
  14. 2023Glaze: Protecting Artists from Style Mimicry by Text-to-Image Modelscited by 200
  15. 2020fastai: A Layered API for Deep Learningcited by 183
  16. 2023Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Modelscited by 175
  17. 2022Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?cited by 150
  18. 2023Boundary Unlearningcited by 132
  19. 2022Measuring Forgetting of Memorized Training Examplescited by 118
  20. 2022BOME! Bilevel Optimization Made Easy: A Simple First-Order Approachcited by 103
  21. 2023On Provable Copyright Protection for Generative Modelscited by 97
  22. 2022Deep Unlearning via Randomized Conditionally Independent Hessianscited by 83
  23. 2021A Value-Function-based Interior-point Method for Non-convex Bi-level Optimizationcited by 60
  24. 2020A Non-Parametric Test to Detect Data-Copying in Generative Modelscited by 52
  25. 2022When do GANs replicate? On the choice of dataset sizecited by 34
  26. 2022Deep Regression Unlearningcited by 32
  27. 2022Learning with Recoverable Forgettingcited by 29
  28. 2021This Person (Probably) Exists. Identity Membership Attacks Against GAN Generated Facescited by 25

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 12; 2025: 15; 2026: 25.

52 papers.

Sort:

By area:

  1. 2024ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruningcited by 51
  2. 2024Machine Unlearning in Generative AI: A Surveycited by 49
  3. 2024UnlearnCanvas: Stylized Image Dataset for Enhanced Machine Unlearning Evaluation in Diffusion Modelscited by 47
  4. 2024Machine Unlearning: A Comprehensive Surveycited by 33
  5. 2024Jailbreak Attacks and Defenses against Multimodal Generative Models: A Surveycited by 31
  6. 2024MUNBa: Machine Unlearning via Nash Bargainingcited by 17
  7. 2024On Catastrophic Inheritance of Large Foundation Modelscited by 15
  8. 2024On the Limitations and Prospects of Machine Unlearning for Generative AIcited by 12
  9. 2024Replication in Visual Diffusion Models: A Survey and Outlookcited by 11
  10. 2025Erasing Concepts, Steering Generations: A Comprehensive Survey of Concept Suppressioncited by 10
  11. 2025Continual Unlearning for Text-to-Image Diffusion Models: A Regularization Perspectivecited by 9
  12. 2025SAEmnesia: Erasing Concepts in Diffusion Models with Supervised Sparse Autoencoderscited by 8
38 more
  1. 2024U Can't Gen This? A Survey of Intellectual Property Protection Methods for Data in Generative AIcited by 7
  2. 2024Model Integrity when Unlearning with T2I Diffusion Modelscited by 7
  3. 2025ReTrack: Data Unlearning in Diffusion Models through Redirecting the Denoising Trajectorycited by 5
  4. 2025Locality-Aware Continual Unlearning for Diffusion Modelscited by 5
  5. 2025EMMA: Concept Erasure Benchmark with Comprehensive Semantic Metrics and Diverse Categoriescited by 4
  6. 2026RAZOR: Ratio-Aware Layer Editing for Targeted Unlearning in Vision Transformers and Diffusion Modelscited by 3
  7. 2025The Linear Geometry of Interpretable Tokens: Jailbreaking Attacks and Defenses for Unlearned Diffusion Modelscited by 2
  8. 2025Not Every Time and Frequency Need to Be Forgotten in Diffusion Unlearningcited by 2
  9. 2025SPQR: A Multi-Dimensional Benchmark for Safety Alignment under Benign Model Adaptationcited by 2
  10. 2025M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Modelscited by 2
  11. 2026Manifold Generalization Provably Proceeds Memorization in Diffusion Modelscited by 2
  12. 2026Unlearning in Diffusion Models: A Unified Framework with KL Divergence and Likelihood Constraintscited by 2
  13. 2025RUB: Evaluating Residual Knowledge in Unlearned Modelscited by 1
  14. 2025Comprehensive Evaluation and Analysis for NSFW Concept Erasure in Text-to-Image Diffusion Modelscited by 1
  15. 2025IMU: Influence-guided Machine Unlearningcited by 1
  16. 2025Beyond Fixed Anchors: Precisely Erasing Concepts with Sibling Exclusive Counterpartscited by 1
  17. 2026Critic-Guided Reinforcement Unlearning in Text-to-Image Diffusioncited by 1
  18. 2026Designing to Forget: Deep Semi-parametric Models for Unlearningcited by 1
  19. 2026Disentangled Sparse Representations for Concept-Separated Diffusion Unlearningcited by 1
  20. 2026Training data attribution in diffusion models via mirrored unlearning and noise-consistent skewcited by 1
  21. 2026PIU: Proximity-guided Identity Unlearning in ID-Conditioned Diffusion Modelscited by 1
  22. 2024Attention Shift: Steering AI Away from Unsafe Contentcited by 0
  23. 2025Comprehensive Assessment and Analysis for NSFW Content Erasure in Text-to-Image Diffusion Modelscited by 0
  24. 2026GUDA: Counterfactual Group-wise Training Data Attribution for Diffusion Models via Unlearningcited by 0
  25. 2026DICE: Disentangling Artist Style from Content via Contrastive Subspace Decomposition in Diffusion Modelscited by 0
  26. 2026Efficient Unlearning through Maximizing Relearning Convergence Delaycited by 0
  27. 2026SafeDiffusion-R1: Online Reward Steering for Safe Diffusion Post-Trainingcited by 0
  28. 2026Robust and Generalizable Safety Steering for Text-to-Image Diffusion Transformerscited by 0
  29. 2026Concept Removal for Frontier Image Generative Modelscited by 0
  30. 2026IREU: Identity-Related Encoder-Only Unlearning for Customized Portrait Generationcited by 0
  31. 2026DriftScope: Measuring The Hidden Effects of Diffusion Model Adaptationcited by 0
  32. 2026TILDE: TILt-based Distributional Erasure for Concept Unlearningcited by 0
  33. 2026AutoAnchor: Stable Diffusion Unlearning Using Cross-Attention as a Manifold Surrogatecited by 0
  34. 2026Inference-Time Concept Suppression and Video-Centric Evaluation for Text-to-Video Modelscited by 0
  35. 2026Spectral Saliency for Machine Unlearningcited by 0
  36. 2026TINA+: Probing Residual Visual Knowledge in Unlearned Diffusion Models via Diffusion-Consistent Text-Free Inversioncited by 0
  37. 2026Mitigating Sequential Reappearance in Diffusion Data-Point Unlearningcited by 0
  38. 2026PruneForget: Joint Unlearning and Pruning of Vision Modelscited by 0

50 most cited of 52 All 52 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Robust Concept Erasure in Diffusion Models: A Theoretical Perspective on Security and Robustness
  2. 2023Unlearnable Examples for Diffusion Models: Protect Data from Unauthorized Exploitation
  3. 2025TRACE: Trajectory-Constrained Concept Erasure in Diffusion Models
  4. 2025Hiding Images in Diffusion Models by Editing Learned Score Functions
  5. 2025Training-Free Safe Denoisers for Safe Use of Diffusion Models
  6. 2022Image Embedding for Denoising Generative Models
  7. 2026Erasing Without Collateral Damage: Precise Concept Removal in Diffusion Models
  8. 2023Restoration by Generation with Constrained Priors