mlsanityData to 7 Oct 2026

Generative Modeling via Drifting

arXiv 2602.04770 ↗First posted 4 Feb 2026cs.LG cs.CVcited by 98

Not on the map. Why?

Cited by 26 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Generative modeling can be formulated as learning a mapping f such that its pushforward distribution matches the data distribution. The pushforward behavior can be carried out iteratively at inference time, for example in diffusion and flow-based models. In this paper, we propose a new paradigm called Drifting Models, which evolve the pushforward distribution during training and naturally admit one-step inference. We introduce a drifting field that governs the sample movement and achieves equilibrium when the distributions match. This leads to a training objective that allows the neural network optimizer to evolve the distribution. In experiments, our one-step generator achieves state-of-the-art results on ImageNet at 256 x 256 resolution, with an FID of 1.54 in latent space and 1.61 in pixel space. We hope that our work opens up new opportunities for high-quality one-step generation.

Key facts. Cited by 98 papers in the map (data through 7 Oct 2026). Builds on 55 of its 73 references that are in the map. First posted on arXiv 4 Feb 2026.

Builds on

55 of 73 references.

References · 55

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 29,915
  4. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  5. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  6. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  7. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  8. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 8,964
  9. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,185
  10. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  11. 2021Masked Autoencoders Are Scalable Vision Learnerscited by 7,339
  12. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
38 more
  1. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 6,979
  2. 2022Scalable Diffusion Models with Transformerscited by 5,931
  3. 2022Classifier-Free Diffusion Guidancecited by 5,717
  4. 2022Flow Matching for Generative Modelingcited by 5,426
  5. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,581
  6. 2022Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flowcited by 3,550
  7. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,229
  8. 2020Exploring Simple Siamese Representation Learningcited by 2,853
  9. 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
  10. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,609
  11. 2020Improved Baselines with Momentum Contrastive Learningcited by 2,282
  12. 2015Variational Inference with Normalizing Flowscited by 2,161
  13. 2016Density estimation using Real NVPcited by 2,029
  14. 2020GLU Variants Improve Transformercited by 1,724
  15. 2018Group Normalizationcited by 1,514
  16. 2019Root Mean Square Layer Normalizationcited by 1,425
  17. 2023One-step Diffusion with Distribution Matching Distillationcited by 978
  18. 2023Stochastic Interpolants: A Unifying Framework for Flows and Diffusionscited by 897
  19. 2023ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoderscited by 703
  20. 2024Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Thinkcited by 666
  21. 2025Mean Flows for One-step Generative Modelingcited by 524
  22. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  23. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 489
  24. 2023Improved Techniques for Training Consistency Modelscited by 430
  25. 2023Simple diffusion: End-to-end diffusion for high resolution imagescited by 413
  26. 2025Reconstruction vs. Generation: Taming Optimization Dilemma in Latent Diffusion Modelscited by 376
  27. 2024One Step Diffusion via Shortcut Modelscited by 373
  28. 2023Understanding Diffusion Objectives as the ELBO with Simple Data Augmentationcited by 293
  29. 2020Query-Key Normalization for Transformerscited by 266
  30. 2015Generative Moment Matching Networkscited by 252
  31. 2023Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Modelscited by 232
  32. 2024Score identity Distillation: Exponentially Fast Distillation of Pretrained Diffusion Models for One-Step Generationcited by 211
  33. 2015Training generative neural networks via Maximum Mean Discrepancy optimizationcited by 149
  34. 2025Inductive Moment Matchingcited by 107
  35. 2025Improved Mean Flows: On the Challenges of Fastforward Generative Modelscited by 102
  36. 2024Flow map matching with stochastic interpolants: A mathematical framework for consistency modelscited by 99
  37. 2024Normalizing Flows are Capable Generative Modelscited by 89
  38. 2025PixelDiT: Pixel Diffusion Transformers for Image Generationcited by 75

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 98.

98 papers.

Sort:

By area:

  1. 2026Representation Fr'echet Loss for Visual Generationcited by 32
  2. 2026Sinkhorn-Drifting Generative Modelscited by 20
  3. 2026One-Step Generative Modeling via Wasserstein Gradient Flowscited by 18
  4. 2026Generative Drifting is Secretly Score Matching: a Spectral and Variational Perspectivecited by 17
  5. 2026Gradient Flow Drifting: Generative Modeling via Wasserstein Gradient Flows of KDE-Approximated Divergencescited by 15
  6. 2026Continuous Latent Diffusion Language Modelcited by 12
  7. 2026Drifting Fields are not Conservativecited by 11
  8. 2026ELT: Elastic Looped Transformers for Visual Generationcited by 11
  9. 2026A Unified View of Score-Based and Drifting Modelscited by 8
  10. 2026On the Wasserstein Gradient Flow Interpretation of Drifting Modelscited by 6
  11. 2026Kernel-Gradient Drifting Modelscited by 6
  12. 2026Drift-Based Policy Optimization: Native One-Step Policy Learning for Online Robot Controlcited by 4
38 more
  1. 2026OpenHLM: An Empirical Recipe for Whole-Body Humanoid Loco-Manipulationcited by 4
  2. 2026Friction-Augmented Drifting Models for Resource-Efficient Domain Translationcited by 3
  3. 2026Geometry-Aware Discretization Error of Diffusion Modelscited by 3
  4. 2026Contrastive Distribution Matching for Amortized Sequential Monte Carlo in Discrete Diffusioncited by 3
  5. 2026Aligning Few-Step Generative Models by Amortizing Sample-based Variational Inferencecited by 3
  6. 2026Fast Organic Crystal Structure Prediction with Unit Cell Flow Matchingcited by 3
  7. 2026Let It Be Simple: One-Step Action Generation for Vision-Language-Action Modelscited by 3
  8. 2026PhysDrift: Bridging the Embodiment Gap in Humanoid Co-Speech Motion Generationcited by 3
  9. 2026Ada3Drift: Adaptive Training-Time Drifting for One-Step 3D Visuomotor Robotic Manipulationcited by 2
  10. 2026Setting-Matched and Semantics-Scaled Benchmarking of One-Step Generative Models Against Multistep Diffusion and Flow Modelscited by 2
  11. 2026V-Co: A Closer Look at Visual Representation Alignment via Co-Denoisingcited by 2
  12. 2026MRI-to-CT synthesis using drifting modelscited by 2
  13. 2026Generative Drifting for Conditional Medical Image Generationcited by 2
  14. 2026Speech Enhancement Based on Drifting Modelscited by 2
  15. 2026Teacher-Feature Drifting: One-Step Diffusion Distillation with Pretrained Diffusion Representationscited by 2
  16. 2026DriftXpress: Faster Drifting Models via Projected RKHS Fieldscited by 2
  17. 2026Finite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Modelscited by 2
  18. 2026Drift Q-Learningcited by 2
  19. 2026Drifting Preference Optimization for One-Step Generative Modelscited by 2
  20. 2026CLEAR: Cognition and Latent Evaluation for Adaptive Routing in End-to-End Autonomous Drivingcited by 2
  21. 2026Distilling Drifting Transformers with Representation Autoencoderscited by 2
  22. 2026Trajectory Forcing: Structure-First Generation with Controllable Semantic Trajectoriescited by 2
  23. 2026DriftWorld: Fast World Modeling through Driftingcited by 2
  24. 2026Three-Body Scattering for Generative Modelingcited by 2
  25. 2026Amortized Moment Matching for Visual Generationcited by 2
  26. 2026Beckmann Transport Models: From Autonomous Flows to One-Step Mapscited by 2
  27. 2026DriftingVLA: Native One-Step Vision-Language-Action Generation via Per-Dimension Temporal Driftingcited by 2
  28. 2026Receding-Horizon Control via Drifting Modelscited by 1
  29. 2026Positive-Only Drifting Policy Optimizationcited by 1
  30. 2026MISTY: High-Throughput Motion Planning via Mixer-based Single-step Driftingcited by 1
  31. 2026Analytical Correction for Subsampling Bias in Drifting Modelscited by 1
  32. 2026Lookahead Drifting Modelcited by 1
  33. 2026Drifting Field Policy: A One-Step Generative Policy via Wasserstein Gradient Flowcited by 1
  34. 2026Consistency Regularised Gradient Flows for Inverse Problemscited by 1
  35. 2026To discretize continually: Mean shift interacting particle systems for Bayesian inferencecited by 1
  36. 2026Efficient Image Synthesis with Sphere Latent Encodercited by 1
  37. 2026Drift Flow Matchingcited by 1
  38. 2026One-Step Distillation of Discrete Diffusion Image Generators via Fixed-Point Iterationcited by 1

50 most cited of 98 All 98 (JSON)