{"id":"arxiv/2603.23783","url":"https://arxiv.org/abs/2603.23783","tags":["cs.LG","cs.AI","math.OC","math.PR","stat.ML"],"score":0,"title":"Probabilistic Geometric Alignment via Bayesian Latent Transport for Domain-Adaptive Foundation Models","domain":"cs.LG","source":"arxiv","authors":["Aueaphum Aueawatthanaphisut","Kuepon Auewattanapisut"],"reviews":0,"subject":"cs.LG","summary":"Adapting large-scale foundation models to new domains with limited supervision remains a fundamental challenge due to latent distribution mismatch, unstable optimization dynamics, and miscalibrated uncertainty propagation. This paper introduces an uncertainty-aware probabilistic latent transport framework that formulates domain adaptation as a stochastic geometric alignment problem in representation space. A Bayesian transport operator is proposed to redistribute latent probability mass along Wasserstein-type geodesic trajectories, while a PAC-Bayesian regularization mechanism constrains…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-03-24","reactions":{}}