mlsanityData to 14 Oct 2026

Probabilistic Precision and Recall Towards Reliable Evaluation of Generative Models

arXiv 2309.01590 ↗First posted 4 Sep 2023cs.LG cs.CVcited by 11

Not on the map. Why?

Abstract (arXiv)

Assessing the fidelity and diversity of the generative model is a difficult but important issue for technological advancement. So, recent papers have introduced k-Nearest Neighbor (kNN) based precision-recall metrics to break down the statistical distance into fidelity and diversity. While they provide an intuitive method, we thoroughly analyze these metrics and identify oversimplified assumptions and undesirable properties of kNN that result in unreliable evaluation, such as susceptibility to outliers and insensitivity to distributional changes. Thus, we propose novel metrics, P-precision and P-recall (PP&PR), based on a probabilistic approach that address the problems. Through extensive investigations on toy experiments and state-of-the-art generative models, we show that our PP&PR provide more reliable estimates for comparing fidelity and diversity than the existing metrics. The codes are available at urlhttps://github.com/kdst-team/Probablisticprecisionrecall.

Key facts. Cited by 11 papers in the map (data through 14 Oct 2026). Builds on 32 of its 42 references that are in the map. First posted on arXiv 4 Sep 2023.

Builds on

32 of 42 references.

References · 32

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2020Score-Based Generative Modeling through Stochastic Differential Equationscited by 9,122
  4. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  5. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 6,013
  6. 2022Generative Adversarial Networkscited by 5,627
  7. 2015Rethinking the Inception Architecture for Computer Visioncited by 5,475
  8. 2019Analyzing and Improving the Image Quality of StyleGANcited by 3,588
  9. 2016Improved Techniques for Training GANscited by 2,970
  10. 2018Large Scale GAN Training for High Fidelity Natural Image Synthesiscited by 2,626
  11. 2021Image Super-Resolution via Iterative Refinementcited by 1,511
  12. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
20 more
  1. 2015LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loopcited by 1,355
  2. 2018Demystifying MMD GANscited by 1,293
  3. 2020Training Generative Adversarial Networks with Limited Datacited by 1,186
  4. 2021Alias-Free Generative Adversarial Networkscited by 1,184
  5. 2021Variational Diffusion Modelscited by 1,153
  6. 2019StarGAN v2: Diverse Image Synthesis for Multiple Domainscited by 1,098
  7. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 944
  8. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  9. 2021Maximum Likelihood Training of Score-Based Diffusion Modelscited by 672
  10. 2021Score-based Generative Modeling in Latent Spacecited by 636
  11. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 434
  12. 2019Dreaming to Distill: Data-free Knowledge Transfer via DeepInversioncited by 423
  13. 2018Assessing Generative Models via Precision and Recallcited by 406
  14. 2022The Role of ImageNet Classes in Fr'echet Inception Distancecited by 213
  15. 2021How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Modelscited by 196
  16. 2020Image Generators with Conditionally-Independent Pixel Synthesiscited by 106
  17. 2018An empirical study on evaluation metrics of generative adversarial networkscited by 81
  18. 2021The Spatially-Correlative Loss for Various Image Translation Taskscited by 66
  19. 2022Multi-View Consistent Generative Adversarial Networks for 3D-aware Image Synthesiscited by 43
  20. 2022Arbitrary-Scale Image Synthesiscited by 20

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 2; 2025: 5; 2026: 4.

11 papers.

  1. 2025Position: All Current Generative Fidelity and Diversity Metrics are Flawedcited by 12
  2. 2025Sprint: Sparse-Dense Residual Fusion for Efficient Diffusion Transformerscited by 4
  3. 2025Blockwise Flow Matching: Improving Flow Matching Models For Efficient High-Quality Generationcited by 3
  4. 2024Unifying and extending Precision Recall metrics for assessing generative modelscited by 1
  5. 2025Enhanced Generative Model Evaluation with Clipped Density and Coveragecited by 1
  6. 2026AutoPKG: An Automated Framework for Dynamic E-commerce Product-Attribute Knowledge Graph Constructioncited by 1
  7. 2024A Unifying Information-theoretic Perspective on Evaluating Generative Modelscited by 0
  8. 2025A New Perspective on Precision and Recall for Generative Modelscited by 0
  9. 2026GICDM: Mitigating Hubness for Reliable Distance-Based Generative Model Evaluationcited by 0
  10. 2026Assessing Sample Quality in Conditional Generation under Compositional Shiftcited by 0
  11. 2026A Semiotics-Aware Framework for Evaluating Fidelity and Coverage in Natural Language Generationcited by 0