无需分类器自由引导的扩散模型

Diffusion Models without Classifier-free Guidance

February 17, 2025

作者: Zhicong Tang, Jianmin Bao, Dong Chen, Baining Guo

cs.AI

摘要

本文提出了一种新颖的训练目标——模型引导（Model-guidance, MG），旨在替代并消除广泛使用的无分类器引导（Classifier-free Guidance, CFG）。我们的创新方法超越了仅对数据分布进行标准建模的局限，将条件后验概率纳入考量。该技术源于CFG的思想，简单而有效，可作为即插即用模块应用于现有模型。我们的方法显著加速了训练过程，使推理速度翻倍，并实现了与甚至超越当前采用CFG的扩散模型相媲美的卓越质量。大量实验验证了该方法在不同模型和数据集上的有效性、效率及可扩展性。最终，我们在ImageNet 256基准测试中取得了1.34的FID值，确立了最新的性能记录。代码已公开于https://github.com/tzco/Diffusion-wo-CFG。

English

This paper presents Model-guidance (MG), a novel objective for training diffusion model that addresses and removes of the commonly used Classifier-free guidance (CFG). Our innovative approach transcends the standard modeling of solely data distribution to incorporating the posterior probability of conditions. The proposed technique originates from the idea of CFG and is easy yet effective, making it a plug-and-play module for existing models. Our method significantly accelerates the training process, doubles the inference speed, and achieve exceptional quality that parallel and even surpass concurrent diffusion models with CFG. Extensive experiments demonstrate the effectiveness, efficiency, scalability on different models and datasets. Finally, we establish state-of-the-art performance on ImageNet 256 benchmarks with an FID of 1.34. Our code is available at https://github.com/tzco/Diffusion-wo-CFG.