AI RESEARCH
Rao-Blackwellized Score Matching on Manifolds
arXiv CS.LG
•
ArXi:2605.25567v1 Announce Type: cross We study denoising score matching (DSM) when the latent distribution is ed on a smooth embedded manifold $M \subset \mathbb{R}^D$. Under ambient Gaussian corruption, the tangent denoising target contains a singular normal-fiber noise channel whose variance diverges as $d/\sigma^2$ as $\sigma \to 0