mlsanityData to 7 Oct 2026

Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models

arXiv 2303.17591 ↗First posted 30 Mar 2023cs.CV cs.AI cs.LGcited by 297

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

also in 1 more
Abstract (arXiv)

The unlearning problem of deep learning models, once primarily an academic concern, has become a prevalent issue in the industry. The significant advances in text-to-image generation techniques have prompted global discussions on privacy, copyright, and safety, as numerous unauthorized personal IDs, content, artistic creations, and potentially harmful materials have been learned by these models and later utilized to generate and distribute uncontrolled content. To address this challenge, we propose Forget-Me-Not, an efficient and low-cost solution designed to safely remove specified IDs, objects, or styles from a well-configured text-to-image model in as little as 30 seconds, without impairing its ability to generate other content. Alongside our method, we introduce the Memorization Score (M-Score) and ConceptBench to measure the models' capacity to generate general concepts, grouped into three primary categories: ID, object, and style. Using M-Score and ConceptBench, we demonstrate that Forget-Me-Not can effectively eliminate targeted concepts while maintaining the model's performance on other concepts. Furthermore, Forget-Me-Not offers two practical extensions: a) removal of potentially harmful or NSFW content, and b) enhancement of model accuracy, inclusion and diversity through concept correction and disentanglement. It can also be adapted as a lightweight model patch for Stable Diffusion, allowing for concept manipulation and convenient distribution. To encourage future research in this critical area and promote the development of safe and inclusive generative models, we will open-source our code and ConceptBench at hrefhttps://github.com/SHI-Labs/Forget-Me-Nothttps://github.com/SHI-Labs/Forget-Me-Not.

Key facts. Cited by 297 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 61 of its 72 references that are in the map. First posted on arXiv 30 Mar 2023.

Builds on

61 of 72 references.

Closest ancestors · 4

Same direction, newest first

  1. 2023Ablating Concepts in Text-to-Image Diffusion Models23 Mar
  2. 2023Erasing Concepts from Diffusion Models13 Mar
  3. 2023Extracting Training Data from Diffusion Models30 Jan
  4. 2022Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models9 Nov

Other references · 57

  1. 2017Attention Is All You Needcited by 59,932
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  5. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  6. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  7. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  8. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  9. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  10. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,084
  11. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  12. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
38 more
  1. 2020End-to-End Object Detection with Transformerscited by 7,384
  2. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  3. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  4. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  5. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  6. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  7. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  8. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  9. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,286
  10. 2022Scaling Instruction-Finetuned Language Modelscited by 2,997
  11. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  12. 2022An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversioncited by 2,231
  13. 2014Conditional Generative Adversarial Netscited by 2,152
  14. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Stepscited by 1,961
  15. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,287
  16. 2021Conceptual 12M: Pushing Web-Scale Image-Text Pre-Training To Recognize Long-Tail Visual Conceptscited by 1,209
  17. 2021Alias-Free Generative Adversarial Networkscited by 1,181
  18. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,095
  19. 2022eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoiserscited by 921
  20. 2023Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Modelscited by 696
  21. 2023Text2Video-Zero: Text-to-Image Diffusion Models are Zero-Shot Video Generatorscited by 666
  22. 2023Muse: Text-To-Image Generation via Masked Generative Transformerscited by 640
  23. 2022Make-A-Scene: Scene-Based Text-to-Image Generation with Human Priorscited by 487
  24. 2022Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Modelscited by 418
  25. 2019SinGAN: Learning a Generative Model from a Single Natural Imagecited by 348
  26. 2022OneFormer: One Transformer to Rule Universal Image Segmentationcited by 327
  27. 2022VQGAN-CLIP: Open Domain Image Generation and Editing with Natural Language Guidancecited by 285
  28. 2021More Control for Free! Image Synthesis with Semantic Diffusion Guidancecited by 211
  29. 2016Unrolled Generative Adversarial Networkscited by 208
  30. 2022Versatile Diffusion: Text, Images and Variations All in One Diffusion Modelcited by 199
  31. 2023Fair Diffusion: Instructing Text-to-Image Generation Models on Fairnesscited by 149
  32. 2022GENIE: Higher-Order Denoising Diffusion Solverscited by 127
  33. 2022Zero-Shot Machine Unlearningcited by 125
  34. 2021Instance-Conditioned GANcited by 119
  35. 2022A Prompt Array Keeps the Bias Away: Debiasing Vision-Language Models with Adversarial Learningcited by 104
  36. 2023Learning to Unlearn: Instance-wise Unlearning for Pre-trained Classifierscited by 66
  37. 2020Learn to Forget: Machine Unlearning via Neuron Maskingcited by 56
  38. 2022Few Shot Generative Model Adaption via Relaxed Spatial Structural Alignmentcited by 54

50 most cited of 57 All 57 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 31; 2024: 90; 2025: 103; 2026: 73.

297 papers.

Sort:

By area:

  1. 2023Unified Concept Editing in Diffusion Modelscited by 377
  2. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 376
  3. 2024MACE: Mass Concept Erasure in Diffusion Modelscited by 298
  4. 2024MUSE: Machine Unlearning Six-Way Evaluation for Language Modelscited by 251
  5. 2024Rethinking Machine Unlearning for Large Language Modelscited by 247
  6. 2023Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?cited by 239
  7. 2023Model Sparsity Can Simplify Machine Unlearningcited by 206
  8. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Nowcited by 204
  9. 2023Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Modelscited by 175
  10. 2024Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Modelscited by 164
  11. 2024Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearningcited by 160
  12. 2023Can Sensitive Information Be Deleted From LLMs? Objectives for Defending Against Extraction Attackscited by 156
38 more
  1. 2023Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Modelscited by 148
  2. 2024SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generationcited by 125
  3. 2023Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Eraserscited by 118
  4. 2023Circumventing Concept Erasure Methods For Text-to-Image Generative Modelscited by 81
  5. 2025SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoderscited by 75
  6. 2023Nightshade: Prompt-Specific Poisoning Attacks on Text-to-Image Generative Modelscited by 72
  7. 2023Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generationcited by 69
  8. 2024Jailbreaking Prompt Attack: A Controllable Adversarial Attack against Diffusion Modelscited by 68
  9. 2023A Comprehensive Survey of Forgetting in Deep Learning Beyond Continual Learningcited by 66
  10. 2023Toward effective protection against diffusion based mimicry through score distillationcited by 64
  11. 2023Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Modelscited by 62
  12. 2024Get What You Want, Not What You Don't: Image Content Suppression for Text-to-Image Diffusion Modelscited by 62
  13. 2025Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Conceptscited by 61
  14. 2024R.A.C.E.: Robust Adversarial Concept Erasure for Secure Text-to-Image Diffusion Modelcited by 56
  15. 2024Erasing Undesirable Influence in Diffusion Modelscited by 52
  16. 2024ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruningcited by 51
  17. 2024Scissorhands: Scrub Data Influence via Connection Sensitivity in Networkscited by 50
  18. 2024Direct Unlearning Optimization for Robust and Safe Text-to-Image Modelscited by 50
  19. 2023Smooth Diffusion: Crafting Smooth Latent Spaces in Diffusion Modelscited by 48
  20. 2024Erasing Undesirable Concepts in Diffusion Models with Adversarial Preservationcited by 48
  21. 2024UnlearnCanvas: Stylized Image Dataset for Enhanced Machine Unlearning Evaluation in Diffusion Modelscited by 47
  22. 2024UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AIcited by 46
  23. 2023Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomycited by 44
  24. 2024Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearningcited by 44
  25. 2024Unified Gradient-Based Machine Unlearning with Remain Geometry Enhancementcited by 43
  26. 2023Towards Safe Self-Distillation of Internet-Scale Text-to-Image Diffusion Modelscited by 41
  27. 2024STEREO: A Two-Stage Framework for Adversarially Robust Concept Erasing from Text-to-Image Diffusion Modelscited by 39
  28. 2024Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Conceptscited by 39
  29. 2024Copyright Protection in Generative AI: A Technical Perspectivecited by 35
  30. 2024Adversarial Attacks and Defenses on Text-to-Image Diffusion Models: A Surveycited by 34
  31. 2024Erasing Conceptual Knowledge from Language Modelscited by 34
  32. 2025Fantastic Targets for Concept Erasure in Diffusion Models and Where To Find Themcited by 34
  33. 2023All but One: Surgical Concept Erasing with Model Preservation in Text-to-Image Diffusion Modelscited by 33
  34. 2024The Stronger the Diffusion Model, the Easier the Backdoor: Data Poisoning to Induce Copyright Breaches Without Adjusting Finetuning Pipelinecited by 33
  35. 2025SPEED: Scalable, Precise, and Efficient Concept Erasure for Diffusion Modelscited by 33
  36. 2023Latent Space Editing in Transformer-Based Flow Matchingcited by 32
  37. 2024Attacks and Defenses for Generative Diffusion Models: A Comprehensive Surveycited by 32
  38. 2024Jailbreak Attacks and Defenses against Multimodal Generative Models: A Surveycited by 31

50 most cited of 297 All 297 (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. 2023Understanding (Un)Intended Memorization in Text-to-Image Generative Models
  3. 2024Unlearning or Concealment? A Critical Analysis and Evaluation Metrics for Unlearning in Diffusion Models
  4. 2026Certifying Concept Unlearning in Text-to-Image Diffusion Models
  5. 2025ForgetMe: Evaluating Selective Forgetting in Generative Models
  6. 2026TILDE: TILt-based Distributional Erasure for Concept Unlearning
  7. 2025Finding DoRI: Discovery of Retained Images in Diffusion Models
  8. 2024Unlearning Concepts in Diffusion Model via Concept Domain Correction and Concept Preserving Gradient