mlsanityData to 14 Oct 2026

Fractal Generative Models

arXiv 2502.17437 ↗First posted 24 Feb 2025cs.LG cs.CVcited by 46

Not on the map. Why?

Cited by 22 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Modularization is a cornerstone of computer science, abstracting complex functions into atomic building blocks. In this paper, we introduce a new level of modularization by abstracting generative models into atomic generative modules. Analogous to fractals in mathematics, our method constructs a new type of generative model by recursively invoking atomic generative modules, resulting in self-similar fractal architectures that we call fractal generative models. As a running example, we instantiate our fractal framework using autoregressive models as the atomic generative modules and examine it on the challenging task of pixel-by-pixel image generation, demonstrating strong performance in both likelihood estimation and generation quality. We hope this work could open a new paradigm in generative modeling and provide a fertile ground for future research. Code is available at https://github.com/LTH14/fractalgen.

Key facts. Cited by 46 papers in the map (data through 14 Oct 2026). Builds on 47 of its 69 references that are in the map. First posted on arXiv 24 Feb 2025.

Builds on

47 of 69 references.

References · 47

  1. 2017Attention Is All You Needcited by 60,361
  2. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  3. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  4. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  5. 2020Denoising Diffusion Implicit Modelscited by 9,781
  6. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  7. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,585
  8. 2022Scalable Diffusion Models with Transformerscited by 6,054
  9. 2022Classifier-Free Diffusion Guidancecited by 5,847
  10. 2022Flow Matching for Generative Modelingcited by 5,702
  11. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  12. 2017Neural Discrete Representation Learningcited by 4,712
35 more
  1. 2021Zero-Shot Text-to-Image Generationcited by 4,288
  2. 2014Going Deeper with Convolutionscited by 4,241
  3. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  4. 2020Taming Transformers for High-Resolution Image Synthesiscited by 3,253
  5. 2016Improved Techniques for Training GANscited by 2,970
  6. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  7. 2023Consistency Modelscited by 1,822
  8. 2022Imagen Video: High Definition Video Generation with Diffusion Modelscited by 1,778
  9. 2019Generating Long Sequences with Sparse Transformerscited by 1,546
  10. 2017Accurate, Large Minibatch SGD: Training ImageNet in 1 Hourcited by 1,492
  11. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  12. 2022MaskGIT: Masked Generative Image Transformercited by 1,099
  13. 2024Visual Autoregressive Modeling: Scalable Image Generation via Next-Scale Predictioncited by 1,023
  14. 2016Conditional Image Generation with PixelCNN Decoderscited by 862
  15. 2014Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognitioncited by 720
  16. 2024Autoregressive Image Generation without Vector Quantizationcited by 651
  17. 2023Muse: Text-To-Image Generation via Masked Generative Transformerscited by 648
  18. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 492
  19. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  20. 2020Efficient Content-Based Sparse Attention with Routing Transformerscited by 471
  21. 2023Simple diffusion: End-to-end diffusion for high resolution imagescited by 422
  22. 2023Understanding Diffusion Objectives as the ELBO with Simple Data Augmentationcited by 304
  23. 2024Infinity: Scaling Bitwise AutoRegressive Modeling for High-Resolution Image Synthesiscited by 304
  24. 2022MAGE: MAsked Generative Encoder to Unify Representation Learning and Image Synthesiscited by 227
  25. 2016FractalNet: Ultra-Deep Neural Networks without Residualscited by 216
  26. 2023MEGABYTE: Predicting Million-byte Sequences with Multiscale Transformerscited by 170
  27. 2024HART: Efficient Visual Generation with Hybrid Autoregressive Transformercited by 132
  28. 2024Normalizing Flows are Capable Generative Modelscited by 91
  29. 2024Simpler Diffusion (SiD2): 1.5 FID on ImageNet512 with pixel-space diffusioncited by 75
  30. 2021Combiner: Full Attention Transformer with Sparse Computation Costcited by 64
  31. 2022General-purpose, long-context autoregressive modeling with Perceiver ARcited by 60
  32. 2024JetFormer: An Autoregressive Generative Model of Raw Images and Textcited by 46
  33. 2024Neural Flow Diffusion Models: Learnable Forward Process for Improved Diffusion Modellingcited by 45
  34. 2020Wavelet Flow: Fast Training of High Resolution Normalizing Flowscited by 33
  35. 2023Unveiling universal aspects of the cellular anatomy of the braincited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 25; 2026: 21.

46 papers.

Sort:

By area:

  1. 2025PixelFlow: Pixel-Space Generative Models with Flowcited by 97
  2. 2025PixelDiT: Pixel Diffusion Transformers for Image Generationcited by 83
  3. 2025PixNerd: Pixel Neural Field Diffusioncited by 82
  4. 2025DeCo: Frequency-Decoupled Pixel Diffusion for End-to-End Image Generationcited by 60
  5. 2025ResearchCodeBench: Benchmarking LLMs on Implementing Novel Machine Learning Research Codecited by 36
  6. 2025STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesiscited by 32
  7. 2025Bridging Continuous and Discrete Tokens for Autoregressive Visual Generationcited by 29
  8. 2025FARMER: Flow AutoRegressive Transformer over Pixelscited by 18
  9. 2025Hyperspherical Latents Improve Continuous-Token Autoregressive Generationcited by 16
  10. 2026PixelGen: Improving Pixel Diffusion with Perceptual Supervisioncited by 14
  11. 2025Locality-aware Parallel Decoding for Efficient Autoregressive Image Generationcited by 10
  12. 2025SuperFlow: Training Flow Matching Models with RL on the Flycited by 10
34 more
  1. 2026Autoregressive Image Generation with Masked Bit Modelingcited by 9
  2. 2026HyperDiT: Hyper-Connected Transformers for High-Fidelity Pixel-Space Diffusioncited by 9
  3. 2025Demystifying Diffusion Objectives: Reweighted Losses are Better Variational Boundscited by 8
  4. 2026Image Generation with a Sphere Encodercited by 8
  5. 2026PixelU: A U-Shaped Transformer for Efficient End-to-End Pixel Diffusioncited by 8
  6. 2025Stochastic Interpolants via Conditional Dependent Couplingcited by 7
  7. 20253D Scene Understanding Through Local Random Access Sequence Modelingcited by 5
  8. 2025Next Visual Granularity Generationcited by 5
  9. 2026FrequencyBooster: Full-Frequency Modeling for High-Fidelity Pixel Diffusioncited by 4
  10. 2026JANUS: A Lightweight Framework for Jailbreaking Text-to-Image Models via Distribution Optimizationcited by 3
  11. 2026Normalizing Flows with Iterative Denoisingcited by 3
  12. 2026FREPix: Frequency-Heterogeneous Flow Matching for Pixel-Space Image Generationcited by 3
  13. 2025ARINAR: Bi-Level Autoregressive Feature-by-Feature Generative Modelscited by 2
  14. 2025Fractal-IR: A Unified Framework for Efficient and Scalable Image Restorationcited by 2
  15. 2025Generation then Reconstruction: Accelerating Masked Autoregressive Models via Two-Stage Samplingcited by 2
  16. 2025Luminance-Aware Statistical Quantization: Unsupervised Hierarchical Learning for Illumination Enhancementcited by 2
  17. 2025Rethinking Generative Image Pretraining: How Far Are We From Scaling Up Next-Pixel Prediction?cited by 2
  18. 2026V-Co: A Closer Look at Visual Representation Alignment via Co-Denoisingcited by 2
  19. 2026Signal or Noise? Understanding Generative Models for Real-World Sensor Time Seriescited by 2
  20. 2025FRN: Fractal-Based Recursive Spectral Reconstruction Networkcited by 1
  21. 2025Ctrl-Z Sampling: Scaling Diffusion Sampling with Controlled Random Zigzag Explorationscited by 1
  22. 2025Heptapod: Language Modeling on Visual Signalscited by 1
  23. 2025RePack then Refine: Efficient Diffusion Transformer with Vision Foundation Modelcited by 1
  24. 2026Agentic Flow Steering and Parallel Rollout Search for Spatially Grounded Text-to-Image Generationcited by 1
  25. 2026DuSPiT: Dual-Branch Sub-Patch Pixel Diffusion Transformercited by 1
  26. 2025Fractal Flow: Hierarchical and Interpretable Normalizing Flow via Topic Modeling and Recursive Strategycited by 0
  27. 2026ViT Registers and Fractal ViTcited by 0
  28. 2026Revisiting Autoregressive Models for Generative Image Classificationcited by 0
  29. 2026Coevolving Representations in Joint Image-Feature Diffusioncited by 0
  30. 2026RaPD: Resolution-Agnostic Pixel Diffusion via Semantics-Enriched Implicit Representationscited by 0
  31. 2026Parallel Rollout Approximation for Pixel-Space Autoregressive Image Generationcited by 0
  32. 2026PixelDiT2: Representation-Grounded Pixel Diffusion Transformerscited by 0
  33. 2026PixelMoE: Scaling Pixel-Space Diffusion with LLM-Style Mixture-of-Expertscited by 0
  34. 2026Generative Representation Learning for Scalable Latent Tokenizationcited by 0