mlsanityData to 7 Oct 2026

One-step Latent-free Image Generation with Pixel Mean Flows

arXiv 2601.22158 ↗First posted 29 Jan 2026cs.CVcited by 49

Not on the map. Why?

Abstract (arXiv)

Modern diffusion/flow-based models for image generation typically exhibit two core characteristics: (i) using multi-step sampling, and (ii) operating in a latent space. Recent advances have made encouraging progress on each aspect individually, paving the way toward one-step diffusion/flow without latents. In this work, we take a further step towards this goal and propose "pixel MeanFlow" (pMF). Our core guideline is to formulate the network output space and the loss space separately. The network target is designed to be on a presumed low-dimensional image manifold (i.e., x-prediction), while the loss is defined via MeanFlow in the velocity space. We introduce a simple transformation between the image manifold and the average velocity field. In experiments, pMF achieves strong results for one-step latent-free generation on ImageNet at 256x256 resolution (2.22 FID) and 512x512 resolution (2.48 FID), filling a key missing piece in this regime. We hope that our study will further advance the boundaries of diffusion/flow-based generative models.

Key facts. Cited by 49 papers in the map (data through 7 Oct 2026). Builds on 45 of its 54 references that are in the map. First posted on arXiv 29 Jan 2026.

Builds on

45 of 54 references.

References · 45

  1. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  6. 2020Denoising Diffusion Implicit Modelscited by 9,655
  7. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  8. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  9. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  10. 2022Scalable Diffusion Models with Transformerscited by 5,931
  11. 2022Classifier-Free Diffusion Guidancecited by 5,717
  12. 2022Flow Matching for Generative Modelingcited by 5,426
33 more
  1. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,736
  2. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  3. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 2,994
  4. 2016Improved Techniques for Training GANscited by 2,938
  5. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  6. 2023Consistency Modelscited by 1,778
  7. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,479
  8. 2022Building Normalizing Flows with Stochastic Interpolantscited by 964
  9. 2022On Distillation of Guided Diffusion Modelscited by 721
  10. 2023ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoderscited by 703
  11. 2024Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Thinkcited by 666
  12. 2025Mean Flows for One-step Generative Modelingcited by 524
  13. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  14. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 489
  15. 2023Consistency Trajectory Models: Learning Probability Flow ODE Trajectory of Diffusioncited by 448
  16. 2023Improved Techniques for Training Consistency Modelscited by 430
  17. 2023Simple diffusion: End-to-end diffusion for high resolution imagescited by 413
  18. 2024One Step Diffusion via Shortcut Modelscited by 373
  19. 2023Understanding Diffusion Objectives as the ELBO with Simple Data Augmentationcited by 293
  20. 2024Simplifying, Stabilizing and Scaling Continuous-Time Consistency Modelscited by 276
  21. 2024Consistency Models Made Easycited by 140
  22. 2025Inductive Moment Matchingcited by 107
  23. 2025Improved Mean Flows: On the Challenges of Fastforward Generative Modelscited by 102
  24. 2024Flow map matching with stochastic interpolants: A mathematical framework for consistency modelscited by 99
  25. 2025PixelFlow: Pixel-Space Generative Models with Flowcited by 91
  26. 2025PixNerd: Pixel Neural Field Diffusioncited by 75
  27. 2025PixelDiT: Pixel Diffusion Transformers for Image Generationcited by 75
  28. 2024Simpler Diffusion (SiD2): 1.5 FID on ImageNet512 with pixel-space diffusioncited by 74
  29. 2025DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generationcited by 55
  30. 2025DiP: Taming Diffusion Models in Pixel Spacecited by 42
  31. 2025There is No VAE: End-to-End Pixel-Space Generative Modeling via Self-Supervised Pre-trainingcited by 23
  32. 2025Bidirectional Normalizing Flow: From Data to Noise and Backcited by 7
  33. 2025MeanFlow Transformers with Representation Autoencoderscited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 49.

49 papers.

Sort:

By area:

  1. 2026Representation Fr'echet Loss for Visual Generationcited by 32
  2. 2026Action-to-Action Flow Matchingcited by 15
  3. 2026PixelGen: Improving Pixel Diffusion with Perceptual Supervisioncited by 10
  4. 2026Representation Distribution Matching for One-Step Visual Generationcited by 10
  5. 2026Self-conditioned Flow Map Language Models via Fixed-point Flowscited by 9
  6. 2026Posterior Refinement: Fast Language Generation via Any-Order Flow Mapscited by 8
  7. 2026Parallel Decoding Distillation for Fast Image and Video Generationcited by 7
  8. 2026Calibrated Test-Time Guidance for Bayesian Inferencecited by 4
  9. 2026CMuon: Accelerating and Stabilizing Diffusion Transformer Training via Chunked Momentum Orthogonalizationcited by 3
  10. 2026AdvFD: Boosting Visual Generation via Adversarial Fr'echet Distance Losscited by 3
  11. 2026An Empirical Study of Training Pixel-Space Text-to-Image Diffusion Modelscited by 3
  12. 2026V-Co: A Closer Look at Visual Representation Alignment via Co-Denoisingcited by 2
37 more
  1. 2026FREPix: Frequency-Heterogeneous Flow Matching for Pixel-Space Image Generationcited by 2
  2. 2026Registers Matter for Pixel-Space Diffusion Transformerscited by 2
  3. 2026Distilling Drifting Transformers with Representation Autoencoderscited by 2
  4. 2026CrossFlow: One-Step Generation Across Latent and Pixel Spacescited by 2
  5. 2026Trajectory Forcing: Structure-First Generation with Controllable Semantic Trajectoriescited by 2
  6. 2026NullFlow: One-Step Generative Reconstructioncited by 2
  7. 2026Three-Body Scattering for Generative Modelingcited by 2
  8. 2026Amortized Moment Matching for Visual Generationcited by 2
  9. 2026FA-RDP: A Frequency-Adaptive Reactive Diffusion Policy for Contact-Rich Manipulationcited by 2
  10. 2026Transition Flow Matchingcited by 1
  11. 2026PixelFlowCast: Latent-Free Precipitation Nowcasting via Pixel Mean Flowscited by 1
  12. 2026Drift Flow Matchingcited by 1
  13. 2026Few-step Cofolding with All-Atom Flow Mapscited by 1
  14. 2026Image2Sim: Scaling Embodied Navigation via Generative Neural Simulatorcited by 1
  15. 2026Unifying Distributional Training for One-Step Visual Generationcited by 1
  16. 2026CoGenCast: A Coupled Autoregressive-Flow Generative Framework for Time Series Forecastingcited by 0
  17. 2026HybridFlow: A 2-NFE Generative Policy for Real-Time Robotic Manipulationcited by 0
  18. 2026Point-MF: Stabilizing One-Step Mean Flows for Single-Image Point Cloud Reconstructioncited by 0
  19. 2026Autoregressive One-Step Generative Modeling for Dynamical System Forecastingcited by 0
  20. 2026Accelerating Redshift-Conditioned Galaxy Image Synthesis with One-step Generative Modelingcited by 0
  21. 2026Cross-scale Aligned Supervision for Training GANscited by 0
  22. 2026Neuron Populations Exhibit Divergent Selectivity with Scalecited by 0
  23. 2026Multimarginal flow matching with optimal transport potentialscited by 0
  24. 2026Robust Watermarks Meet Backdoored Models: Evading Diffusion Semantic Watermarks via Stealthy Backdoorcited by 0
  25. 2026Beyond Token-Level Cross-Entropy: Fr'echet Distributional Post-Training for Autoregressive Image Generationcited by 0
  26. 2026Where Does Generative Difficulty Reside? An Empirical Study of Target Representationscited by 0
  27. 2026Rethinking Pixel Mean Flows via Interval Denoisercited by 0
  28. 2026FiRe: Fixed-Noise Refinement for Visual Counterfactual Explanationscited by 0
  29. 2026Continuous Adversarial MeanFlow Transfercited by 0
  30. 2026Efficient JPEG Restoration in the Wavelet Domain via Mean Flowscited by 0
  31. 2026Bio-MF: Low-Latency and High-Fidelity EEG-to-fNIRS Cross-Modal Generation for Hybrid Motor-Imagery Brain--Computer Interfacescited by 0
  32. 2026PixelART: Image-to-Layer Decomposition without Latents or Text-to-Image Pretrainingcited by 0
  33. 2026One-Step Generative Modeling via Unbalanced Optimal Transportcited by 0
  34. 2026From Feed-Forward to Flow: Unifying Reconstruction and Generation Is Easier Than You Thinkcited by 0
  35. 2026RepFlow: Reciprocal Supervision Improves Generation and Representation in Flow Modelscited by 0
  36. 2026Acceleration of Diffusion Language Model through Discrete Average Generatorcited by 0
  37. 2026Sphere Encoder 2cited by 0