Bespoke Non-Stationary Solvers for Fast Sampling of Diffusion and Flow Models

arXiv:2403.01329v1 Announce Type: new
Abstract: This paper introduces Bespoke Non-Stationary (BNS) Solvers, a solver distillation approach to improve sample efficiency of Diffusion and Flow models. BNS solvers are based on a family of non-stationary solvers that provably subsumes existing numerical ODE solvers and consequently demonstrate considerable improvement in sample approximation (PSNR) over these baselines. Compared to model distillation, BNS solvers benefit from a tiny parameter space ($<$200 parameters), fast optimization (two orders of magnitude faster), maintain diversity of samples, and …

abstract arxiv cs.ai cs.cv cs.lg diffusion distillation efficiency family flow improvement numerical paper sample sampling solver type