Trustworthy Evaluation of Generative AI Models

Abstract

Generative AI (GenAI) models have recently achieved remarkable empirical performance in various applications, however, their evaluations yet lack uncertainty quantification. In this paper, we propose a method to compare two generative models based on an unbiased estimator of their relative performance gap. Statistically, our estimator achieves parametric convergence rate and asymptotic normality, which enables valid inference. Computationally, our method is efficient and can be accelerated by parallel computing and leveraging pre-storing intermediate results. On simulated datasets with known ground truth, we show our approach effectively controls type I error and achieves power comparable with commonly used metrics. Furthermore, we demonstrate the performance of our method in evaluating diffusion models on real image datasets with statistical confidence.

Tasks

Reproductions