mlsanityData to 7 Oct 2026

Rolling Diffusion Models

arXiv 2402.09470 ↗First posted 12 Feb 2024cs.LG stat.MLcited by 102

On the map: Diffusion models → Flow matching, time series and science

Abstract (arXiv)

Diffusion models have recently been increasingly applied to temporal data such as video, fluid mechanics simulations, or climate data. These methods generally treat subsequent frames equally regarding the amount of noise in the diffusion process. This paper explores Rolling Diffusion: a new approach that uses a sliding window denoising process. It ensures that the diffusion process progressively corrupts through time by assigning more noise to frames that appear later in a sequence, reflecting greater uncertainty about the future as the generation process unfolds. Empirically, we show that when the temporal dynamics are complex, Rolling Diffusion is superior to standard diffusion. In particular, this result is demonstrated in a video prediction task using the Kinetics-600 video dataset and in a chaotic fluid dynamics forecasting experiment.

Key facts. Cited by 102 papers in the map (data through 7 Oct 2026). Placed in Flow matching, time series and science, one of 14 directions in Diffusion models. Builds on 50 of its 56 references that are in the map. First posted on arXiv 12 Feb 2024.

Builds on

50 of 56 references.

Closest ancestors · 3

Same direction, newest first

  1. 2023PDE-Refiner: Achieving Accurate Long Rollouts with Neural PDE Solvers10 Aug
  2. 2023DYffusion: A Dynamics-informed Diffusion Model for Spatiotemporal Forecasting3 Jun
  3. 2021CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation7 Jul

Other references · 47

  1. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  3. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  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. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,067
  7. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  8. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  9. 2022Video Diffusion Modelscited by 2,280
  10. 2017The Kinetics Human Action Video Datasetcited by 1,894
  11. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,765
  12. 2022Make-A-Video: Text-to-Video Generation without Text-Video Datacited by 1,725
35 more
  1. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,465
  2. 2020DiffWave: A Versatile Diffusion Model for Audio Synthesiscited by 1,339
  3. 2022Scaling Autoregressive Models for Content-Rich Text-to-Image Generationcited by 1,287
  4. 2022Imagic: Text-Based Real Image Editing with Diffusion Modelscited by 1,172
  5. 2021Variational Diffusion Modelscited by 1,145
  6. 2022Diffusion-LM Improves Controllable Text Generationcited by 1,006
  7. 2022On Distillation of Guided Diffusion Modelscited by 721
  8. 2021VideoGPT: Video Generation using VQ-VAE and Transformerscited by 651
  9. 2023Language Model Beats Diffusion -- Tokenizer is Key to Visual Generationcited by 633
  10. 2022Phenaki: Variable Length Video Generation From Open Domain Textual Descriptioncited by 475
  11. 2021Machine learning accelerated computational fluid dynamicscited by 461
  12. 2023Simple diffusion: End-to-end diffusion for high resolution imagescited by 413
  13. 2022MAGVIT: Masked Generative Video Transformercited by 341
  14. 2018A Short Note about Kinetics-600cited by 331
  15. 2023Photorealistic Video Generation with Diffusion Modelscited by 304
  16. 2022Flexible Diffusion Modeling of Long Videoscited by 294
  17. 2023Understanding Diffusion Objectives as the ELBO with Simple Data Augmentationcited by 293
  18. 2023Preserve Your Own Correlation: A Noise Prior for Video Diffusion Modelscited by 284
  19. 2021N"UWA: Visual Synthesis Pre-training for Neural visUal World creAtioncited by 253
  20. 2022Diffusion Probabilistic Modeling for Video Generationcited by 237
  21. 2022Scalable Adaptive Computation for Iterative Generationcited by 190
  22. 2019Scaling Autoregressive Video Modelscited by 189
  23. 2023Video Probabilistic Diffusion Models in Projected Latent Spacecited by 171
  24. 2023GenCast: Diffusion-based ensemble forecasting for medium-range weathercited by 169
  25. 2017Self-Supervised Visual Planning with Temporal Skip Connectionscited by 137
  26. 2023AR-Diffusion: Auto-Regressive Diffusion Model for Text Generationcited by 113
  27. 2021FitVid: Overfitting in Pixel-Level Video Predictioncited by 86
  28. 2021CCVS: Context-aware Controllable Video Synthesiscited by 68
  29. 2020Transformation-based Adversarial Video Prediction on Large-Scale Datacited by 66
  30. 2022Learning to correct spectral methods for simulating turbulent flowscited by 49
  31. 2023TEDi: Temporally-Entangled Diffusion for Long-Term Motion Synthesiscited by 48
  32. 2022Temporally Consistent Transformers for Video Generationcited by 43
  33. 2022Transframer: Arbitrary Frame Prediction with Generative Modelscited by 37
  34. 2023E3 TTS: Easy End-to-End Diffusion-based Text to Speechcited by 30
  35. 2023A Neural PDE Solver with Temporal Stencil Modelingcited by 22

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 16; 2025: 48; 2026: 38.

102 papers.

Sort:

By area:

  1. 2024Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusioncited by 634
  2. 2025Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusioncited by 569
  3. 2024From Slow Bidirectional to Fast Autoregressive Video Diffusion Modelscited by 401
  4. 2024Diffusion Models Are Real-Time Game Enginescited by 282
  5. 2025Rolling Forcing: Autoregressive Long Video Diffusion in Real Timecited by 209
  6. 2025Self-Forcing++: Towards Minute-Scale High-Quality Video Generationcited by 176
  7. 2025History-Guided Video Diffusioncited by 165
  8. 2024FIFO-Diffusion: Generating Infinite Videos from Text without Trainingcited by 134
  9. 2025Test-Time Training Done Rightcited by 121
  10. 2025AdaWorld: Learning Adaptable World Models with Latent Actionscited by 108
  11. 2025Mixture of Contexts for Long Video Generationcited by 90
  12. 2025RELIC: Interactive Video World Model with Long-Horizon Memorycited by 89
38 more
  1. 2024Simpler Diffusion (SiD2): 1.5 FID on ImageNet512 with pixel-space diffusioncited by 74
  2. 2025MotionStream: Real-Time Video Generation with Interactive Motion Controlscited by 71
  3. 2025Stable Video Infinity: Infinite-Length Video Generation with Error Recyclingcited by 68
  4. 2025Large Video Planner Enables Generalizable Robot Controlcited by 67
  5. 2024Progressive Autoregressive Video Diffusion Modelscited by 65
  6. 2024Speculative Diffusion Decoding: Accelerating Language Generation through Diffusioncited by 57
  7. 2024On conditional diffusion models for PDE simulationscited by 56
  8. 2026Helios: Real Real-Time Long Video Generation Modelcited by 56
  9. 2025AR-Diffusion: Asynchronous Video Generation with Auto-Regressive Diffusioncited by 53
  10. 2024Streaming Diffusion Policy: Fast Policy Synthesis with Variable Noise Diffusion Modelscited by 51
  11. 2025A Fourier Space Perspective on Diffusion Modelscited by 48
  12. 2025End-to-End Training for Autoregressive Video Diffusion via Self-Resamplingcited by 45
  13. 2024One Diffusion to Generate Them Allcited by 41
  14. 2024Autoregressive Diffusion Transformer for Text-to-Speech Synthesiscited by 34
  15. 2025Streaming Flow Policy: Simplifying diffusion/flow-matching policies by treating action trajectories as flow trajectoriescited by 34
  16. 2025Learning 3D Persistent Embodied World Modelscited by 29
  17. 2025A Survey of Interactive Generative Videocited by 23
  18. 2025Error Analyses of Auto-Regressive Video Diffusion Models: A Unified Frameworkcited by 20
  19. 2025LoViC: Efficient Long Video Generation with Context Compressioncited by 19
  20. 2026INSPATIO-WORLD: A Real-Time 4D World Simulator via Spatiotemporal Autoregressive Modelingcited by 18
  21. 2025MALT Diffusion: Memory-Augmented Latent Transformers for Any-Length Video Generationcited by 14
  22. 2025AR-1-to-3: Single Image to Consistent 3D Object Generation via Next-View Predictioncited by 14
  23. 2025WorldWeaver: Generating Long-Horizon Video Worlds via Rich Perceptioncited by 14
  24. 2025Position: Interactive Generative Video as Next-Generation Game Enginecited by 13
  25. 2025MaskFlow: Discrete Flows For Flexible and Efficient Long Video Generationcited by 12
  26. 2025Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithmscited by 12
  27. 2025ScalingNoise: Scaling Inference-Time Search for Generating Infinite Videoscited by 12
  28. 2025Data Synthesis with Diverse Styles for Face Recognition via 3DMM-Guided Diffusioncited by 12
  29. 2025Elucidated Rolling Diffusion Models for Probabilistic Forecasting of Complex Dynamicscited by 12
  30. 2026Mode Seeking meets Mean Seeking for Fast Long Video Generationcited by 12
  31. 2025From Sparse Signal to Smooth Motion: Real-Time Motion Generation with Rolling Prediction Modelscited by 11
  32. 2025Diffusion-Based Hierarchical Graph Neural Networks for Simulating Nonlinear Solid Mechanicscited by 10
  33. 2024CL-DiffPhyCon: Closed-loop Diffusion Control of Complex Physical Systemscited by 8
  34. 2025Dynamical Diffusion: Learning Temporal Dynamics with Diffusion Modelscited by 8
  35. 2025Speculative Jacobi-Denoising Decoding for Accelerating Autoregressive Text-to-image Generationcited by 7
  36. 2025WorldPack: Dynamic Frame Compression for Long-context Video World Modelingcited by 7
  37. 2025RFLAV: Rolling Flow matching for infinite Audio Video generationcited by 6
  38. 2025Diffusion Buffer: Online Diffusion-based Speech Enhancement with Sub-Second Latencycited by 6

50 most cited of 102 All 102 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025Generating time-consistent dynamics with discriminator-guided image diffusion models
  2. 2026When Denoising Hurts: Rethinking the Terminal Step of Diffusion Time Series Forecasters -- Extended Version
  3. 2026PDE-regularized Dynamics-informed Diffusion with Uncertainty-aware Filtering for Long-Horizon Dynamics
  4. 2023Diffusion Models for Time Series Applications: A Survey
  5. 2026Drift Estimation for Stochastic Differential Equations with Denoising Diffusion Models
  6. 2025ADiff4TPP: Asynchronous Diffusion Models for Temporal Point Processes
  7. 2026A Tutorial on Diffusion Theory: From Differential Equations to Diffusion Models
  8. 2025The Principles of Diffusion Models