{"ID":2859663,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.04441","arxiv_id":"2510.04441","title":"Domain Generalization Under Posterior Drift","abstract":"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.","short_abstract":"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 domai...","url_abs":"https://arxiv.org/abs/2510.04441","url_pdf":"https://arxiv.org/pdf/2510.04441v2","authors":"[\"Yilun Zhu\",\"Naihao Deng\",\"Naichen Shi\",\"Aditya Gangrade\",\"Clayton Scott\"]","published":"2025-10-06T02:17:12Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"stat.ML\"]","methods":"[]","has_code":false}