mlsanityData to 7 Oct 2026

Reliable Fidelity and Diversity Metrics for Generative Models

arXiv 2002.09797 ↗First posted 23 Feb 2020cs.CV cs.LG stat.MLcited by 425

Not on the map. Why?

Cited by 119 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Devising indicative evaluation metrics for the image generation task remains an open problem. The most widely used metric for measuring the similarity between real and generated images has been the Fr'echet Inception Distance (FID) score. Because it does not differentiate the fidelity and diversity aspects of the generated images, recent papers have introduced variants of precision and recall metrics to diagnose those properties separately. In this paper, we show that even the latest version of the precision and recall metrics are not reliable yet. For example, they fail to detect the match between two identical distributions, they are not robust against outliers, and the evaluation hyperparameters are selected arbitrarily. We propose density and coverage metrics that solve the above issues. We analytically and experimentally show that density and coverage provide more interpretable and reliable signals for practitioners than the existing metrics. Code: https://github.com/clovaai/generative-evaluation-prdc.

Key facts. Cited by 425 papers in the map (data through 7 Oct 2026). First posted on arXiv 23 Feb 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 32; 2022: 47; 2023: 62; 2024: 91; 2025: 97; 2026: 96.

425 papers.

Sort:

By area:

  1. 2022StyleGAN-XL: Scaling StyleGAN to Large Diverse Datasetscited by 489
  2. 2022TabDDPM: Modelling Tabular Data with Diffusion Modelscited by 373
  3. 2022The Vendi Score: A Diversity Evaluation Metric for Machine Learningcited by 299
  4. 2023Imitating Human Behaviour with Diffusion Modelscited by 260
  5. 2022The Role of ImageNet Classes in Fr'echet Inception Distancecited by 211
  6. 2021Few-shot Image Generation via Cross-domain Correspondencecited by 206
  7. 2023Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion modelscited by 199
  8. 2021How Faithful is your Synthetic Data? Sample-level Metrics for Evaluating and Auditing Generative Modelscited by 193
  9. 2022RenderDiffusion: Image Diffusion for 3D Reconstruction, Inpainting and Generationcited by 162
  10. 2023LayoutDM: Discrete Diffusion Model for Controllable Layout Generationcited by 141
  11. 2023DNA-GPT: Divergent N-Gram Analysis for Training-Free Detection of GPT-Generated Textcited by 127
  12. 2021Pros and Cons of GAN Evaluation Measures: New Developmentscited by 121
38 more
  1. 2021Instance-Conditioned GANcited by 119
  2. 2021Image-to-Image Translation: Methods and Applicationscited by 116
  3. 2024Why Tabular Foundation Models Should Be a Research Prioritycited by 103
  4. 2024Improving Text-to-Image Consistency via Automatic Prompt Optimizationcited by 88
  5. 2022STaSy: Score-based Tabular data Synthesiscited by 85
  6. 2023KoLA: Carefully Benchmarking World Knowledge of Large Language Modelscited by 85
  7. 2022PoseGPT: Quantization-based 3D Human Motion Generation and Forecastingcited by 78
  8. 2023CoDi: Co-evolving Contrastive Diffusion Models for Mixed-type Tabular Synthesiscited by 78
  9. 2023Generating images of rare concepts using pre-trained diffusion modelscited by 78
  10. 2022Variational Model Inversion Attackscited by 75
  11. 2021Unleashing Transformers: Parallel Token Prediction with Discrete Absorbing Diffusion for Fast High-Resolution Image Generation from Vector-Quantized Codescited by 71
  12. 2021The Spatially-Correlative Loss for Various Image Translation Taskscited by 66
  13. 2022Synthetic Data -- what, why and how?cited by 66
  14. 2023Generating and Imputing Tabular Data via Diffusion and Flow-based Gradient-Boosted Treescited by 60
  15. 2022Plug & Play Attacks: Towards Robust and Flexible Model Inversion Attackscited by 58
  16. 2021Rebooting ACGAN: Auxiliary Classifier GANs with Stable Trainingcited by 57
  17. 2021Using latent space regression to analyze and leverage compositionality in GANscited by 55
  18. 2023Generative Table Pre-training Empowers Models for Tabular Predictioncited by 55
  19. 2023Towards Understanding the Interplay of Generative Artificial Intelligence and the Internetcited by 54
  20. 2021Adversarial Text-to-Image Synthesis: A Reviewcited by 52
  21. 2024On the Content Bias in Fr'echet Video Distancecited by 51
  22. 2023PanoDiffusion: 360-degree Panorama Outpainting via Diffusioncited by 50
  23. 2022Denoising Likelihood Score Matching for Conditional Score-based Data Generationcited by 45
  24. 2022On Evaluation Metrics for Graph Generative Modelscited by 44
  25. 2024VidMuse: A Simple Video-to-Music Generation Framework with Long-Short-Term Modelingcited by 39
  26. 2022StudioGAN: A Taxonomy and Benchmark of GANs for Image Synthesiscited by 38
  27. 2022Self-Distilled StyleGAN: Towards Generation from Internet Photoscited by 37
  28. 2023Don't Play Favorites: Minority Guidance for Diffusion Modelscited by 37
  29. 2022Mutual Information Divergence: A Unified Metric for Multimodal Generative Modelscited by 35
  30. 2022Rarity Score : A New Metric to Evaluate the Uncommonness of Synthesized Imagescited by 35
  31. 2024Consistency-diversity-realism Pareto fronts of conditional image generative modelscited by 35
  32. 2023Retrieval-Augmented Layout Transformer for Content-Aware Layout Generationcited by 34
  33. 2023Feedback-guided Data Synthesis for Imbalanced Classificationcited by 33
  34. 2024Diff-A-Riff: Musical Accompaniment Co-creation via Latent Diffusion Modelscited by 33
  35. 2022Full-band General Audio Synthesis with Score-based Diffusioncited by 32
  36. 2022Evaluating generative models in high energy physicscited by 31
  37. 2024On the Challenges and Opportunities in Generative AIcited by 31
  38. 2024Simple and Fast Distillation of Diffusion Modelscited by 31

50 most cited of 425 All 425 (JSON)