Entropic Riemannian Neural Optimal Transport

ArXi:2605.04255v1 Announce Type: cross Many machine learning problems involve data ed on curved spaces such as spheres, rotation groups, hyperbolic spaces, and general Riemannian manifolds, where Euclidean geometry can distort distances, averages, and the resulting optimal transport (OT) problem. Existing manifold OT methods have pursued amortized out-of-sample maps, while entropic regularization has made discrete OT scalable, but these advantages have remained largely disjoint.