Multicalibration for Unbiased Model-Based Prevalence Estimation

Excerpt

arXiv:2604.21549v2 Announce Type: replace Abstract: Estimating the prevalence of a category in a population using imperfect measurement devices (diagnostic tests, classifiers, or large language models) is fundamental to science, public health, and online trust and safety. Standard approaches correct for known device error rates but assume these rates remain stable across populations. We show this assumption fails under covariate shift and that multicalibration, which enforces calibration conditi