<ul><li>Introduction of a framework for efficient diffusion models for symmetric-space Riemannian manifolds.</li><li>Proposal of a new diffusion model for symmetric manifolds with spatially-varying covariance to bypass heat kernel computations.</li><li>Training algorithm minimizes an efficient objective derived via Ito's Lemma, leading to reduced computational complexity.</li><li>Empirical results show improved training speed and sample quality on synthetic datasets on various symmetric manifolds.</li></ul>

Efficient Diffusion Models for Symmetric Manifolds

Discover more