mlsanityData to 7 Oct 2026

Cold Diffusion: Inverting Arbitrary Image Transforms Without Noise

arXiv 2208.09392 ↗First posted 19 Aug 2022cs.CV cs.LGcited by 275

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Standard diffusion models involve an image transform -- adding Gaussian noise -- and an image restoration operator that inverts this degradation. We observe that the generative behavior of diffusion models is not strongly dependent on the choice of image degradation, and in fact an entire family of generative models can be constructed by varying this choice. Even when using completely deterministic degradations (e.g., blur, masking, and more), the training and test-time update rules that underlie diffusion models can be easily generalized to create generative models. The success of these fully deterministic models calls into question the community's understanding of diffusion models, which relies on noise in either gradient Langevin dynamics or variational inference, and paves the way for generalized diffusion models that invert arbitrary processes. Our code is available at https://github.com/arpitbansal297/Cold-Diffusion-Models

Key facts. Cited by 275 papers in the map (data through 7 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 31 of its 38 references that are in the map. First posted on arXiv 19 Aug 2022.

Builds on

31 of 38 references.

Closest ancestors · 7

Same direction, newest first

  1. 2022Elucidating the Design Space of Diffusion-Based Generative Models1 Jun
  2. 2021Cascaded Diffusion Models for High Fidelity Image Generation30 May
  3. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  4. 2021Improved Denoising Diffusion Probabilistic Models18 Feb
  5. 2020Score-Based Generative Modeling through Stochastic Differential Equations26 Nov
  6. 2020Denoising Diffusion Implicit Models6 Oct
  7. 2020Denoising Diffusion Probabilistic Models19 Jun

Other references · 24

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  4. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  5. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  6. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  7. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  8. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,595
  9. 2016Improved Techniques for Training GANscited by 2,938
  10. 2019Benchmarking Neural Network Robustness to Common Corruptions and Perturbationscited by 2,857
  11. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  12. 2021Image Super-Resolution via Iterative Refinementcited by 1,506
12 more
  1. 2018Demystifying MMD GANscited by 1,268
  2. 2022Denoising Diffusion Restoration Modelscited by 951
  3. 2016The Little Engine that Could: Regularization by Denoising (RED)cited by 303
  4. 2021Deblurring via Stochastic Refinementcited by 248
  5. 2018A Note on the Inception Scorecited by 225
  6. 2021SNIPS: Solving Noisy Inverse Problems Stochasticallycited by 191
  7. 2022Generative Modelling With Inverse Heat Dissipationcited by 164
  8. 2016Mode Regularized Generative Adversarial Networkscited by 123
  9. 2020Adversarial score matching and improved sampling for image generationcited by 95
  10. 2017Learned D-AMP: Principled Neural Network based Compressive Image Recoverycited by 65
  11. 2021Stochastic Image Denoising by Sampling from the Posterior Distributioncited by 45
  12. 2019LOGAN: Latent Optimisation for Generative Adversarial Networkscited by 42

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 20; 2023: 89; 2024: 60; 2025: 66; 2026: 40.

275 papers.

Sort:

By area:

  1. 2022Diffusion Models: A Comprehensive Survey of Methods and Applicationscited by 1,107
  2. 2022Building Normalizing Flows with Stochastic Interpolantscited by 964
  3. 2023Structure and Content-Guided Video Synthesis with Diffusion Modelscited by 589
  4. 2023Universal Guidance for Diffusion Modelscited by 426
  5. 2022A Survey on Generative Diffusion Modelcited by 238
  6. 2023A Comprehensive Survey of AI-Generated Content (AIGC): A History of Generative AI from GAN to ChatGPTcited by 221
  7. 2024Diffusion Model-Based Image Editing: A Surveycited by 212
  8. 2023Inversion by Direct Iteration: An Alternative to Denoising Diffusion for Image Restorationcited by 164
  9. 2023Solving Linear Inverse Problems Provably via Posterior Sampling with Latent Diffusion Modelscited by 156
  10. 2023Ambient Diffusion: Learning Clean Distributions from Corrupted Datacited by 108
  11. 2023DYffusion: A Dynamics-informed Diffusion Model for Spatiotemporal Forecastingcited by 105
  12. 2022How to Backdoor Diffusion Models?cited by 102
38 more
  1. 2022Soft Diffusion: Score Matching for General Corruptionscited by 101
  2. 2023Stochastic interpolants with data-dependent couplingscited by 94
  3. 2022Blurring Diffusion Modelscited by 89
  4. 2023Minimizing Trajectory Curvature of ODE-based Generative Modelscited by 87
  5. 2023Refusion: Enabling Large-Size Realistic Image Restoration with Latent-Space Diffusion Modelscited by 86
  6. 2022BeLFusion: Latent Diffusion for Behavior-Driven Human Motion Predictioncited by 82
  7. 2023Beyond First-Order Tweedie: Solving Inverse Problems using Latent Diffusioncited by 78
  8. 2022Multiscale Structure Guided Diffusion for Image Deblurringcited by 76
  9. 2023Diffusion Models for Black-Box Optimizationcited by 75
  10. 2024What Matters When Repurposing Diffusion Models for General Dense Perception Tasks?cited by 74
  11. 2023Restoration-Degradation Beyond Linear Diffusions: A Non-Asymptotic Analysis For DDIM-Type Samplerscited by 73
  12. 2023A Survey on Audio Diffusion Models: Text To Speech Synthesis and Enhancement in Generative AIcited by 68
  13. 2023Residual Denoising Diffusion Modelscited by 68
  14. 20223D-LDM: Neural Implicit 3D Shape Generation with Latent Diffusion Modelscited by 67
  15. 2024A Survey on Diffusion Models for Time Series and Spatio-Temporal Datacited by 66
  16. 2024Diffusion Models, Image Super-Resolution And Everything: A Surveycited by 63
  17. 2023Towards Coherent Image Inpainting Using Denoising Diffusion Implicit Modelscited by 61
  18. 2023On the Design Fundamentals of Diffusion Models: A Surveycited by 59
  19. 2023Consistent Diffusion Models: Mitigating Sampling Drift by Learning to be Consistentcited by 58
  20. 2023CoreDiff: Contextual Error-Modulated Generalized Diffusion Model for Low-Dose CT Denoising and Generalizationcited by 51
  21. 2022Efficient Diffusion Models for Vision: A Surveycited by 50
  22. 2023TransFusion -- A Transparency-Based Diffusion Model for Anomaly Detectioncited by 43
  23. 2022DiffusER: Discrete Diffusion via Edit-based Reconstructioncited by 41
  24. 2023Information-Theoretic Diffusioncited by 41
  25. 2023LayoutDiffusion: Improving Graphic Layout Generation by Discrete Diffusion Probabilistic Modelscited by 40
  26. 2023Continuous-Time Functional Diffusion Processescited by 39
  27. 2024Schedule Your Edit: A Simple yet Effective Diffusion Noise Schedule for Image Editingcited by 38
  28. 2023PromptPaint: Steering Text-to-Image Generation Through Paint Medium-like Interactionscited by 34
  29. 2022Multiresolution Textual Inversioncited by 33
  30. 2024The Stronger the Diffusion Model, the Easier the Backdoor: Data Poisoning to Induce Copyright Breaches Without Adjusting Finetuning Pipelinecited by 33
  31. 2022f-DM: A Multi-stage Diffusion Model via Progressive Signal Transformationcited by 32
  32. 2022Cold Diffusion for Speech Enhancementcited by 31
  33. 2023Memory-Efficient 3D Denoising Diffusion Models for Medical Image Processingcited by 30
  34. 2024Stretching Each Dollar: Diffusion Training from Scratch on a Micro-Budgetcited by 30
  35. 2023Blackout Diffusion: Generative Diffusion Models in Discrete-State Spacescited by 29
  36. 2023Shelving, Stacking, Hanging: Relational Pose Diffusion for Multi-modal Rearrangementcited by 29
  37. 2024Blue noise for diffusion modelscited by 29
  38. 2023Diffusion Probabilistic Fieldscited by 28

50 most cited of 275 All 275 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2022Diffusion Models in Vision: A Survey
  2. 2024Non-Normal Diffusion Models
  3. 2025Diffusion Models: A Mathematical Introduction
  4. 2022Progressive Deblurring of Diffusion Models for Coarse-to-Fine Image Synthesis
  5. 2024GUD: Generation with Unified Diffusion
  6. 2022Image Embedding for Denoising Generative Models
  7. 2021Variational Diffusion Models
  8. 2023Neural Diffusion Models