Domain generalization (DG) is the problem of generalizing from several distributions (or domains), for which labeled training data are available, to a new test domain for which no labeled data is available. For the prevailing benchmark datasets in DG, there exists a single classifier that performs well across all domains. In this work, we study a fundamentally different regime where the domains satisfy a \emph{posterior drift} assumption, in which the optimal classifier might vary substantially with domain. We establish a decision-theoretic framework for DG under posterior drift, and investigate the practical implications of this framework through experiments on language and vision tasks.
@article{arxiv.2510.04441,
title = {Domain Generalization Under Posterior Drift},
author = {Yilun Zhu and Naihao Deng and Naichen Shi and Aditya Gangrade and Clayton Scott},
journal= {arXiv preprint arXiv:2510.04441},
year = {2026}
}