TADA: Improved Diffusion Sampling with Training-free Augmented Dynamics

TL;DR

This paper introduces TADA, a training-free diffusion sampling method that accelerates image generation by up to 186%, leveraging higher-dimensional noise and an ODE solver to produce detailed images efficiently.

Contribution

TADA presents a novel, training-free diffusion sampling approach that uses higher-dimensional noise and an ODE solver to significantly improve sampling speed and control over image detail.

Findings

Up to 186% faster sampling on ImageNet512.

Effective across various pretrained diffusion models.

Allows control over image detail without extra computation.

Abstract

Diffusion models have demonstrated exceptional capabilities in generating high-fidelity images but typically suffer from inefficient sampling. Many solver designs and noise scheduling strategies have been proposed to dramatically improve sampling speeds. In this paper, we introduce a new sampling method that is up to $186\%$ faster than the current state of the art solver for comparative FID on ImageNet512. This new sampling method is training-free and uses an ordinary differential equation (ODE) solver. The key to our method resides in using higher-dimensional initial noise, allowing to produce more detailed samples with less function evaluations from existing pretrained diffusion models. In addition, by design our solver allows to control the level of detail through a simple hyper-parameter at no extra computational cost. We present how our approach leverages momentum dynamics by establishing a fundamental equivalence between momentum diffusion models and conventional diffusion models with respect to their training paradigms. Moreover, we observe the use of higher-dimensional noise naturally exhibits characteristics similar to stochastic differential equations (SDEs). Finally, we demonstrate strong performances on a set of representative pretrained diffusion models, including EDM, EDM2, and Stable-Diffusion 3, which cover models in both pixel and latent spaces, as well as class and text conditional settings. The code is available at https://github.com/apple/ml-tada.