A New Perspective on Precision and Recall for Generative Models

[Submitted on 4 Nov 2025 (v1), last revised 10 Jun 2026 (this version, v3)]

Authors:Benjamin Sykes (Unicaen, Ensicaen, Greyc), Loïc Simon (Unicaen, Ensicaen, Greyc), Julien Rabin (Unicaen, Ensicaen, Greyc), Jalal Fadili (Unicaen, Ensicaen, Greyc)

View PDF

Abstract:With the recent success of generative models in image and text, the question of their evaluation has recently gained a lot of attention. While most methods from the state of the art rely on scalar metrics, the introduction of Precision and Recall (PR) for generative model has opened up a new avenue of research. The associated PR curve allows for a richer analysis, but their estimation poses several challenges. In this paper, we present a new framework for estimating entire PR curves based on a binary classification standpoint. We conduct a thorough statistical analysis of the proposed estimates. As a byproduct, we obtain a minimax upper bound on the PR estimation risk. We also show that our framework extends several landmark PR metrics of the literature which by design are restrained to the extreme values of the curve. Finally, we study the different behaviors of the curves obtained experimentally in various settings.

Submission history

From: Benjamin Sykes [view email] [via CCSD proxy]
[v1] Tue, 4 Nov 2025 09:44:11 UTC (810 KB)
[v2] Mon, 18 May 2026 14:53:55 UTC (3,586 KB)
[v3] Wed, 10 Jun 2026 15:28:34 UTC (3,587 KB)