mlsanityData to 7 Oct 2026

Safe Latent Diffusion: Mitigating Inappropriate Degeneration in Diffusion Models

arXiv 2211.05105 ↗First posted 9 Nov 2022cs.CV cs.AI cs.LGcited by 501

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

Abstract (arXiv)

Text-conditioned image generation models have recently achieved astonishing results in image quality and text alignment and are consequently employed in a fast-growing number of applications. Since they are highly data-driven, relying on billion-sized datasets randomly scraped from the internet, they also suffer, as we demonstrate, from degenerated and biased human behavior. In turn, they may even reinforce such biases. To help combat these undesired side effects, we present safe latent diffusion (SLD). Specifically, to measure the inappropriate degeneration due to unfiltered and imbalanced training sets, we establish a novel image generation test bed-inappropriate image prompts (I2P)-containing dedicated, real-world image-to-text prompts covering concepts such as nudity and violence. As our exhaustive empirical evaluation demonstrates, the introduced SLD removes and suppresses inappropriate image parts during the diffusion process, with no additional training required and no adverse effect on overall image quality or text alignment.

Key facts. Cited by 501 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 29 of its 60 references that are in the map. First posted on arXiv 9 Nov 2022.

Builds on

29 of 60 references.

References · 29

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  5. 2022Photorealistic Text-to-Image Diffusion Models with Deep Language Understandingcited by 6,322
  6. 2022Classifier-Free Diffusion Guidancecited by 5,717
  7. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,131
  8. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  9. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,409
  10. 2019Language Models as Knowledge Bases?cited by 1,922
  11. 2021LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairscited by 1,544
  12. 2018Datasheets for Datasetscited by 1,371
17 more
  1. 2016Man is to Computer Programmer as Woman is to Homemaker? Debiasing Word Embeddingscited by 1,304
  2. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
  3. 2022Imagic: Text-Based Real Image Editing with Diffusion Modelscited by 1,172
  4. 2016Semantics derived automatically from language corpora contain human-like biasescited by 894
  5. 2020How Much Knowledge Can You Pack Into the Parameters of a Language Model?cited by 713
  6. 2021Persistent Anti-Muslim Bias in Large Language Modelscited by 376
  7. 2021Self-Diagnosis and Self-Debiasing: A Proposal for Reducing Corpus-Based Bias in NLPcited by 336
  8. 2021Multimodal datasets: misogyny, pornography, and malignant stereotypescited by 326
  9. 2022Text2LIVE: Text-Driven Layered Image and Video Editingcited by 312
  10. 2019Measurement and Fairnesscited by 299
  11. 2019Mitigating Gender Bias in Natural Language Processing: Literature Reviewcited by 287
  12. 2020Large image datasets: A pyrrhic win for computer vision?cited by 237
  13. 2021Large Pre-trained Language Models Contain Human-like Biases of What is Right and Wrong to Docited by 178
  14. 2022Can Machines Help Us Answering Question 16 in Datasheets, and In Turn Reflecting on Inappropriate Content?cited by 150
  15. 2020REVISE: A Tool for Measuring and Mitigating Bias in Visual Datasetscited by 136
  16. 2020Image Representations Learned With Unsupervised Pre-Training Contain Human-like Biasescited by 107
  17. 2018Using Image Fairness Representations in Diversity-Based Re-ranking for Recommendationscited by 11

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 50; 2024: 136; 2025: 167; 2026: 147.

501 papers.

Sort:

By area:

  1. 2023Erasing Concepts from Diffusion Modelscited by 530
  2. 2023Unified Concept Editing in Diffusion Modelscited by 377
  3. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generationcited by 376
  4. 2023Ablating Concepts in Text-to-Image Diffusion Modelscited by 312
  5. 2024MACE: Mass Concept Erasure in Diffusion Modelscited by 298
  6. 2023Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Modelscited by 297
  7. 2023Text-to-image Diffusion Models in Generative AI: A Surveycited by 264
  8. 2023Ring-A-Bell! How Reliable are Concept Removal Methods for Diffusion Models?cited by 239
  9. 2024Diffusion Model-Based Image Editing: A Surveycited by 212
  10. 2023Understanding and Mitigating Copying in Diffusion Modelscited by 205
  11. 2023To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Nowcited by 204
  12. 2023MMA-Diffusion: MultiModal Attack on Diffusion Modelscited by 184
38 more
  1. 2023Selective Amnesia: A Continual Learning Approach to Forgetting in Deep Generative Modelscited by 175
  2. 2023Unsafe Diffusion: On the Generation of Unsafe Images and Hateful Memes From Text-To-Image Modelscited by 171
  3. 2023One-Dimensional Adapter to Rule Them All: Concepts, Diffusion Models and Erasing Applicationscited by 164
  4. 2024Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Modelscited by 164
  5. 2023Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Promptscited by 161
  6. 2023Holistic Evaluation of Text-To-Image Modelscited by 160
  7. 2023Fair Diffusion: Instructing Text-to-Image Generation Models on Fairnesscited by 149
  8. 2023Concept Sliders: LoRA Adaptors for Precise Control in Diffusion Modelscited by 148
  9. 2023LEDITS++: Limitless Image Editing using Text-to-Image Modelscited by 147
  10. 2024SAFREE: Training-Free and Adaptive Guard for Safe Text-to-Image And Video Generationcited by 125
  11. 2024Reliable and Efficient Concept Erasure of Text-to-Image Diffusion Modelscited by 123
  12. 2025Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrailscited by 122
  13. 2023Receler: Reliable Concept Erasing of Text-to-Image Diffusion Models via Lightweight Eraserscited by 118
  14. 2024EraseAnything: Enabling Concept Erasure in Rectified Flow Transformerscited by 93
  15. 2023Finetuning Text-to-Image Diffusion Models for Fairnesscited by 90
  16. 2023Circumventing Concept Erasure Methods For Text-to-Image Generative Modelscited by 81
  17. 2023FLIRT: Feedback Loop In-context Red Teamingcited by 81
  18. 2023Diffusion Models Without Attentioncited by 75
  19. 2025SAeUron: Interpretable Concept Unlearning in Diffusion Models with Sparse Autoencoderscited by 75
  20. 2023Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generationcited by 69
  21. 2024Unlearning Concepts in Diffusion Model via Concept Domain Correction and Concept Preserving Gradientcited by 69
  22. 2024Jailbreaking Prompt Attack: A Controllable Adversarial Attack against Diffusion Modelscited by 68
  23. 2024Latent Guard: a Safety Framework for Text-to-image Generationcited by 68
  24. 2024T2VSafetyBench: Evaluating the Safety of Text-to-Video Generative Modelscited by 66
  25. 2024SynthCLIP: Are We Ready for a Fully Synthetic CLIP Training?cited by 64
  26. 2024SafeGen: Mitigating Sexually Explicit Content Generation in Text-to-Image Modelscited by 64
  27. 2024MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?cited by 64
  28. 2023Diffusion Models for Open-Vocabulary Segmentationcited by 63
  29. 2024GuardT2I: Defending Text-to-Image Models from Adversarial Promptscited by 61
  30. 2025Set You Straight: Auto-Steering Denoising Trajectories to Sidestep Unwanted Conceptscited by 61
  31. 2024Style Vectors for Steering Generative Large Language Modelcited by 56
  32. 2024R.A.C.E.: Robust Adversarial Concept Erasure for Secure Text-to-Image Diffusion Modelcited by 56
  33. 2024Erasing Undesirable Influence in Diffusion Modelscited by 52
  34. 2024UnsafeBench: Benchmarking Image Safety Classifiers on Real-World and AI-Generated Imagescited by 52
  35. 2024ConceptPrune: Concept Editing in Diffusion Models via Skilled Neuron Pruningcited by 51
  36. 2024Scissorhands: Scrub Data Influence via Connection Sensitivity in Networkscited by 50
  37. 2024Balancing Act: Distribution-Guided Debiasing in Diffusion Modelscited by 50
  38. 2024Direct Unlearning Optimization for Robust and Safe Text-to-Image Modelscited by 50

50 most cited of 501 All 501 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Test-Time Scaling for Safe Text-Guided Image Generation via Intermediate Clean Estimates
  2. 2024Safety Alignment Backfires: Preventing the Re-emergence of Suppressed Concepts in Fine-tuned Text-to-Image Diffusion Models
  3. 2025Evaluating Robustness in Latent Diffusion Models via Embedding Level Augmentation
  4. 2024Controllable Generation with Text-to-Image Diffusion Models: A Survey
  5. 2024Kaleido Diffusion: Improving Conditional Diffusion Models with Autoregressive Latent Modeling
  6. 2026i1: A Simple and Fully Open Recipe for Strong Text-to-Image Models
  7. 2024On Improved Conditioning Mechanisms and Pre-training Strategies for Diffusion Models
  8. 2025Latent Diffusion Unlearning: Protecting Against Unauthorized Personalization Through Trajectory Shifted Perturbations