# Revisiting Precision and Recall Definition for Generative Model Evaluation

arXiv: https://arxiv.org/abs/1905.05441 · HTML: https://mlsanity.com/paper/1905.05441/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 May 2019.

## Built on by (most cited first, 9 of 9)

- [The Vendi Score: A Diversity Evaluation Metric for Machine Learning](https://mlsanity.com/paper/2210.02410/) (2022)
- [Pros and Cons of GAN Evaluation Measures: New Developments](https://mlsanity.com/paper/2103.09396/) (2021)
- [Do Quantum Circuit Born Machines Generalize?](https://mlsanity.com/paper/2207.13645/) (2022)
- [Beyond Scale: The Diversity Coefficient as a Data Quality Metric for Variability in Natural Language Data](https://mlsanity.com/paper/2306.13840/) (2023)
- [Generalization Metrics for Practical Quantum Advantage in Generative Models](https://mlsanity.com/paper/2201.08770/) (2022)
- [Are Large Language Models Fit For Guided Reading?](https://mlsanity.com/paper/2305.10645/) (2023)
- [Enhanced Generative Model Evaluation with Clipped Density and Coverage](https://mlsanity.com/paper/2507.01761/) (2025)
- [Evaluating quantum generative models via imbalanced data classification benchmarks](https://mlsanity.com/paper/2308.10847/) (2023)
- [Approximating f-Divergences with Rank Statistics](https://mlsanity.com/paper/2601.22784/) (2026)

## Abstract (arXiv)

In this article we revisit the definition of Precision-Recall (PR) curves for generative models proposed by Sajjadi et al. (arXiv:1806.00035). Rather than providing a scalar for generative quality, PR curves distinguish mode-collapse (poor recall) and bad quality (poor precision). We first generalize their formulation to arbitrary measures, hence removing any restriction to finite support. We also expose a bridge between PR curves and type I and type II error rates of likelihood ratio classifiers on the task of discriminating between samples of the two distributions. Building upon this new perspective, we propose a novel algorithm to approximate precision-recall curves, that shares some interesting methodological properties with the hypothesis testing technique from Lopez-Paz et al (arXiv:1610.06545). We demonstrate the interest of the proposed formulation over the original approach on controlled multi-modal datasets.