Diffusion and flow models achieve high generative quality but remain computationally expensive due to slow multi-step sampling. Distillation methods accelerate them by training fast student generators, yet most existing objectives lack a unified theoretical foundation. In this work, we propose Di-Bregman, a compact framework that formulates diffusion distillation as Bregman divergence-based density-ratio matching. This convex-analytic view connects several existing objectives through a common lens. Experiments on CIFAR-10 and text-to-image generation demonstrate that Di-Bregman achieves improved one-step FID over reverse-KL distillation and maintains high visual fidelity compared to the teacher model. Our results highlight Bregman density-ratio matching as a practical and theoretically-grounded route toward efficient one-step diffusion generation.
@article{arxiv.2510.16983,
title = {One-step Diffusion Models with Bregman Density Ratio Matching},
author = {Yuanzhi Zhu and Eleftherios Tsonis and Lucas Degeorge and Vicky Kalogeiton},
journal= {arXiv preprint arXiv:2510.16983},
year = {2025}
}