mlsanityData to 14 Oct 2026

On Exact Inversion of DPM-Solvers

arXiv 2311.18387 ↗First posted 30 Nov 2023cs.CV cs.LGcited by 47

On the map: Diffusion models → Diffusion foundations and sampling

Abstract (arXiv)

Diffusion probabilistic models (DPMs) are a key component in modern generative models. DPM-solvers have achieved reduced latency and enhanced quality significantly, but have posed challenges to find the exact inverse (i.e., finding the initial noise from the given image). Here we investigate the exact inversions for DPM-solvers and propose algorithms to perform them when samples are generated by the first-order as well as higher-order DPM-solvers. For each explicit denoising step in DPM-solvers, we formulated the inversions using implicit methods such as gradient descent or forward step method to ensure the robustness to large classifier-free guidance unlike the prior approach using fixed-point iteration. Experimental results demonstrated that our proposed exact inversion methods significantly reduced the error of both image and noise reconstructions, greatly enhanced the ability to distinguish invisible watermarks and well prevented unintended background changes consistently during image editing. Project page: urlhttps://smhongok.github.io/inv-dpm.html.

Key facts. Cited by 47 papers in the map (data through 14 Oct 2026). Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models. Builds on 41 of its 45 references that are in the map. First posted on arXiv 30 Nov 2023.

Builds on

41 of 45 references.

Closest ancestors · 14

Same direction, newest first

  1. 2022Fast Sampling of Diffusion Models via Operator Learning24 Nov
  2. 2022DPM-Solver++: Fast Solver for Guided Sampling of Diffusion Probabilistic Models2 Nov
  3. 2022On Distillation of Guided Diffusion Models6 Oct
  4. 2022Classifier-Free Diffusion Guidance26 Jul
  5. 2022DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps2 Jun
  6. 2022Accelerating Diffusion Models via Early Stop of the Diffusion Process25 May
  7. 2022Fast Sampling of Diffusion Models with Exponential Integrator29 Apr
  8. 2021High-Resolution Image Synthesis with Latent Diffusion Models20 Dec
  9. 2021Tackling the Generative Learning Trilemma with Denoising Diffusion GANs15 Dec
  10. 2021Variational Diffusion Models1 Jul
  11. 2021Diffusion Models Beat GANs on Image Synthesis11 May
  12. 2020Denoising Diffusion Implicit Models6 Oct
2 more
  1. 2020Denoising Diffusion Probabilistic Models19 Jun
  2. 2020Improved Techniques for Training Score-Based Generative Models16 Jun

Other references · 27

  1. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  2. 2021GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Modelscited by 3,614
  3. 2022DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generationcited by 3,300
  4. 2022Prompt-to-Prompt Image Editing with Cross Attention Controlcited by 2,431
  5. 2015Variational Inference with Normalizing Flowscited by 2,170
  6. 2021SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equationscited by 1,937
  7. 2022RePaint: Inpainting using Denoising Diffusion Probabilistic Modelscited by 1,430
  8. 2019Normalizing Flows for Probabilistic Modeling and Inferencecited by 1,323
  9. 2022Denoising Diffusion Restoration Modelscited by 963
  10. 2019Image2StyleGAN: How to Embed Images Into the StyleGAN Latent Space?cited by 675
  11. 2021DiffusionCLIP: Text-Guided Diffusion Models for Robust Image Manipulationcited by 642
  12. 2020In-Domain GAN Inversion for Real Image Editingcited by 420
15 more
  1. 2021SRDiff: Single Image Super-Resolution with Diffusion Probabilistic Modelscited by 415
  2. 2021GAN Inversion: A Surveycited by 367
  3. 2019Image2StyleGAN++: How to Edit the Embedded Images?cited by 356
  4. 2017Compressed Sensing using Generative Modelscited by 339
  5. 2022Inversion-Based Style Transfer with Diffusion Modelscited by 291
  6. 2022EDICT: Exact Diffusion Inversion via Coupled Transformationscited by 258
  7. 2022Dual Diffusion Implicit Bridges for Image-to-Image Translationcited by 213
  8. 2021High-Fidelity GAN Inversion for Image Attribute Editingcited by 197
  9. 2023Tree-Ring Watermarks: Fingerprints for Diffusion Images that are Invisible and Robustcited by 187
  10. 2023Solving Inverse Problems with Latent Diffusion Models via Hard Data Consistencycited by 182
  11. 2023Localizing Object-level Shape Variations with Text-to-Image Diffusion Modelscited by 142
  12. 2023TrojDiff: Trojan Attacks on Diffusion Models with Diverse Targetscited by 88
  13. 2023Effective Real Image Editing with Accelerated Iterative Diffusion Inversioncited by 79
  14. 2020Composing Normalizing Flows for Inverse Problemscited by 48
  15. 2023Exact Diffusion Inversion via Bi-directional Integration Approximationcited by 44

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 16; 2025: 16; 2026: 14.

47 papers.

Sort:

By area:

  1. 2024Taming Rectified Flow for Inversion and Editingcited by 204
  2. 2024Improving the Training of Rectified Flowscited by 108
  3. 2024DiT4Edit: Diffusion Transformer for Image Editingcited by 101
  4. 2024An Undetectable Watermark for Generative Image Modelscited by 81
  5. 2024Watermark Anything with Localized Messagescited by 53
  6. 2024Video Seal: Open and Efficient Video Watermarkingcited by 48
  7. 2024ROBIN: Robust and Invisible Watermarks for Diffusion Models with Adversarial Optimizationcited by 46
  8. 2023Lightning-Fast Image Inversion and Editing for Text-to-Image Diffusion Modelscited by 35
  9. 2024Unveil Inversion and Invariance in Flow Transformer for Versatile Image Editingcited by 34
  10. 2024BELM: Bidirectional Explicit Linear Multi-step Sampler for Exact Inversion in Diffusion Modelscited by 29
  11. 2024DiffuseTrace: A Transparent and Flexible Watermarking Scheme for Latent Diffusion Modelcited by 21
  12. 2024SmoothCache: A Universal Inference Acceleration Technique for Diffusion Transformerscited by 21
35 more
  1. 2025IntrinsicEdit: Precise generative image manipulation in intrinsic spacecited by 17
  2. 2025Pixel Seal: Adversarial-only training for invisible image and video watermarkingcited by 15
  3. 2024Dual-Schedule Inversion: Training- and Tuning-Free Inversion for Real Image Editingcited by 8
  4. 2024Gradient-free Decoder Inversion in Latent Diffusion Modelscited by 5
  5. 2025MarkDiffusion: An Open-Source Toolkit for Generative Watermarking of Latent Diffusion Modelscited by 5
  6. 2025Secure and Robust Watermarking for AI-generated Images: A Comprehensive Surveycited by 5
  7. 2024Oscillation Inversion: Understand the structure of Large Flow Model through the Lens of Inversion Methodcited by 4
  8. 2024SuperMark: Robust and Training-free Image Watermarking via Diffusion-based Super-Resolutioncited by 4
  9. 2025HOFAR: High-Order Augmentation of Flow Autoregressive Transformerscited by 4
  10. 2025Fine-Tuning Diffusion Models via Intermediate Distribution Shapingcited by 4
  11. 2026SemBind: Binding Diffusion Watermarks to Semantics Against Black-Box Forgery Attackscited by 4
  12. 2025MeSS: City Mesh-Guided Outdoor Scene Generation with Cross-View Consistent Diffusioncited by 3
  13. 2025FlowOpt: Fast Optimization Through Whole Flow Processes for Training-Free Editingcited by 3
  14. 2026Learning to Watermark in the Latent Space of Generative Modelscited by 3
  15. 2025Learning Single Index Models with Diffusion Priorscited by 2
  16. 2025Instability in Diffusion ODEs: An Explanation for Inaccurate Image Reconstructioncited by 2
  17. 2025Zero-Shot Video Deraining with Video Diffusion Modelscited by 2
  18. 2026OSI: One-step Inversion Excels in Extracting Diffusion Watermarkscited by 2
  19. 2024CLUE-MARK: Watermarking Diffusion Models using CLWEcited by 1
  20. 2025Detection Limits and Statistical Separability of Tree Ring Watermarks in Rectified Flow-based Text-to-Image Generation Modelscited by 1
  21. 2025MUSE: Model-Agnostic Tabular Watermarking via Multi-Sample Selectioncited by 1
  22. 2026Attention to Neural Plagiarism: Diffusion Models Can Plagiarize Your Copyrighted Images!cited by 1
  23. 2026BiFM: Bidirectional Flow Matching for Few-Step Image Editing and Generationcited by 1
  24. 2026On the Robustness of Watermarking for Autoregressive Image Generationcited by 1
  25. 2026When Does High-CFG Diffusion Inversion Fail? A Controlled Study of Prompt--Latent Interactionscited by 1
  26. 2026MarkNull: Model-Agnostic Watermark Removal in AI-Generated Images via On-Manifold Latent Manipulationcited by 1
  27. 2025Sample-Efficient Optimisation over the Outputs of Generative Modelscited by 0
  28. 2025CSMapping: Scalable Crowdsourced Semantic Mapping and Topology Inference for Autonomous Drivingcited by 0
  29. 2025An Iteration-Free Fixed-Point Estimator for Diffusion Inversioncited by 0
  30. 2026PGID: Progressive Guided Inversion and Denoising for Robust Watermark Detectioncited by 0
  31. 2026Timestep Rescheduling in Diffusion Inversioncited by 0
  32. 2026ResEdit: Residual embeddings for precise generative image editingcited by 0
  33. 2026Wavelet-Guided Semantic Signal Compensation for Inversion-Free Image Editingcited by 0
  34. 2026One Inverse Step is a Convex Program: Bayes-Limit Calibration of Diffusion Inversioncited by 0
  35. 2026Multi-History-Step SDE Inversion for Image Editing with Superior Regional Awarenesscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023DPM-Solver-v3: Improved Diffusion ODE Solver with Empirical Model Statistics
  2. 2024Accelerating Diffusion Sampling with Optimized Time Steps
  3. 2025Exploiting the Exact Denoising Posterior Score in Training-Free Guidance of Diffusion Models
  4. 2023On Inference Stability for Diffusion Models
  5. 2024Denoising Diffusion Probabilistic Models in Six Simple Steps
  6. 2023DPM-OT: A New Diffusion Probabilistic Model Based on Optimal Transport
  7. 2022Diffusion Model Based Posterior Sampling for Noisy Linear Inverse Problems
  8. 2026Inverting the Generation Process of Denoising Diffusion Implicit Models: Empirical Evaluation and a Novel Method