Your Diffusion Model is Secretly a Noise Classifier and Benefits from Contrastive Training

TL;DR

This paper reveals that diffusion models function as noise classifiers and introduces contrastive training to improve denoising, especially in out-of-distribution regions, enhancing sample quality and sampling speed.

Contribution

The authors propose a novel contrastive training objective that leverages the implicit log-likelihood ratio in diffusion models to improve out-of-distribution denoising performance.

Findings

Contrastive training improves OOD denoising.

Enhanced parallel sampling performance.

Significant speedup in sample generation.

Abstract

Diffusion models learn to denoise data and the trained denoiser is then used to generate new samples from the data distribution. In this paper, we revisit the diffusion sampling process and identify a fundamental cause of sample quality degradation: the denoiser is poorly estimated in regions that are far Outside Of the training Distribution (OOD), and the sampling process inevitably evaluates in these OOD regions. This can become problematic for all sampling methods, especially when we move to parallel sampling which requires us to initialize and update the entire sample trajectory of dynamics in parallel, leading to many OOD evaluations. To address this problem, we introduce a new self-supervised training objective that differentiates the levels of noise added to a sample, leading to improved OOD denoising performance. The approach is based on our observation that diffusion models implicitly define a log-likelihood ratio that distinguishes distributions with different amounts of noise, and this expression depends on denoiser performance outside the standard training distribution. We show by diverse experiments that the proposed contrastive diffusion training is effective for both sequential and parallel settings, and it improves the performance and speed of parallel samplers significantly.