mlsanityData to 14 Oct 2026

Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design

arXiv 1902.00275 ↗First posted 1 Feb 2019cs.LG cs.NE stat.MLcited by 262

Not on the map. Why?

Cited by 42 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Flow-based generative models are powerful exact likelihood models with efficient sampling and inference. Despite their computational efficiency, flow-based models generally have much worse density modeling performance compared to state-of-the-art autoregressive models. In this paper, we investigate and improve upon three limiting design choices employed by flow-based models in prior work: the use of uniform noise for dequantization, the use of inexpressive affine flows, and the use of purely convolutional conditioning networks in coupling layers. Based on our findings, we propose Flow++, a new flow-based model that is now the state-of-the-art non-autoregressive model for unconditional density estimation on standard image benchmarks. Our work has begun to close the significant performance gap that has so far existed between autoregressive models and flow-based models. Our implementation is available at https://github.com/aravindsrinivas/flowpp

Key facts. Cited by 262 papers in the map (data through 14 Oct 2026). First posted on arXiv 1 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 72; 2022: 57; 2023: 57; 2024: 36; 2025: 25; 2026: 15.

262 papers.

Sort:

By area:

  1. 2022Flow Matching for Generative Modelingcited by 5,702
  2. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  3. 2021Cascaded Diffusion Models for High Fidelity Image Generationcited by 1,163
  4. 2021Variational Diffusion Modelscited by 1,153
  5. 2021Maximum Likelihood Training of Score-Based Diffusion Modelscited by 672
  6. 2021VideoGPT: Video Generation using VQ-VAE and Transformerscited by 658
  7. 2021Argmax Flows and Multinomial Diffusion: Learning Categorical Distributionscited by 574
  8. 2025Diffusion Transformers with Representation Autoencoderscited by 295
  9. 2021Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Modelscited by 252
  10. 2021E(n) Equivariant Normalizing Flowscited by 165
  11. 2022Concrete Score Matching: Generalized Score Matching for Discrete Datacited by 148
  12. 2022CITRIS: Causal Identifiability from Temporal Intervened Sequencescited by 133
38 more
  1. 2023Score-Based Diffusion Models as Principled Priors for Inverse Imagingcited by 127
  2. 2021ArtFlow: Unbiased Image Style Transfer via Reversible Neural Flowscited by 113
  3. 2023Diffusion Models for Image Restoration and Enhancement: A Comprehensive Surveycited by 109
  4. 2022Let us Build Bridges: Understanding and Extending Diffusion Generative Modelscited by 103
  5. 2021Soft Truncation: A Universal Training Technique of Score-based Diffusion Model for High Precision Score Estimationcited by 97
  6. 2022Poisson Flow Generative Modelscited by 95
  7. 2021Invertible Image Signal Processingcited by 94
  8. 2024Normalizing Flows are Capable Generative Modelscited by 91
  9. 2021Colorization Transformercited by 89
  10. 2023Reflected Diffusion Modelscited by 87
  11. 2021Diffusion Normalizing Flowcited by 77
  12. 2021Hierarchical Conditional Flow: A Unified Framework for Image Super-Resolution and Image Rescalingcited by 73
  13. 2022BDDM: Bilateral Denoising Diffusion Models for Fast and High-Quality Speech Synthesiscited by 64
  14. 2022Towards Blind Watermarking: Combining Invertible and Non-invertible Mechanismscited by 61
  15. 2021Hierarchical VAEs Know What They Don't Knowcited by 58
  16. 2023Improved Techniques for Maximum Likelihood Estimation for Diffusion ODEscited by 54
  17. 2021Smooth Normalizing Flowscited by 53
  18. 2021Soft then Hard: Rethinking the Quantization in Neural Image Compressioncited by 51
  19. 2021Densely connected normalizing flowscited by 49
  20. 2024Deep Generative Models through the Lens of the Manifold Hypothesis: A Survey and New Connectionscited by 47
  21. 2024JetFormer: An Autoregressive Generative Model of Raw Images and Textcited by 46
  22. 2022A Tale of Two Flows: Cooperative Learning of Langevin Flow and Normalizing Flow Toward Energy-Based Modelcited by 44
  23. 2023Information-Theoretic Diffusioncited by 44
  24. 2022Model-Based Image Signal Processors via Learnable Dictionariescited by 43
  25. 2021MALI: A memory efficient and reverse accurate integrator for Neural ODEscited by 42
  26. 2021Resampling Base Distributions of Normalizing Flowscited by 42
  27. 2022Maximum Likelihood Training of Implicit Nonlinear Diffusion Modelscited by 42
  28. 2021Bilateral Denoising Diffusion Modelscited by 36
  29. 2022FP-Diffusion: Improving Score-based Diffusion Models by Enforcing the Underlying Score Fokker-Planck Equationcited by 36
  30. 2021Normalizing field flows: Solving forward and inverse stochastic differential equations using physics-informed flow modelscited by 35
  31. 2023A Survey on Graph Diffusion Models: Generative AI in Science for Molecule, Protein and Materialcited by 35
  32. 2021Featurized Density Ratio Estimationcited by 33
  33. 2023Reversible Vision Transformerscited by 33
  34. 2023Autoregressive Conditional Neural Processescited by 33
  35. 2025STARFlow: Scaling Latent Normalizing Flows for High-resolution Image Synthesiscited by 32
  36. 2022Universal approximation property of invertible neural networkscited by 30
  37. 2021Implicit Normalizing Flowscited by 28
  38. 2022Modeling sRGB Camera Noise with Normalizing Flowscited by 28

50 most cited of 262 All 262 (JSON)