A Plug-in Interpretation of Conditioning in Score-Based Diffusion Models

2026-08-19

AI Summary

提出了一种基于目标和条件多速联合扩散机制的扩散模型调节方法，通过插入修正项实现透明可控的条件生成，并引入对数福克-普朗克残差正则化以提高采样质量。

Abstract

We propose a conditioning mechanism for diffusion models based on multi-speed joint diffusion of the target and the condition. The mechanism learns an unconditional joint score network and enforces conditioning at inference via a plug-in correction term. The plug-in term separates the conditioning contribution from the learned unconditional dynamics, offering a transparent view of how the condition steers generation of the target distribution. Building on this, we derive explicit conditional reverse-time SDEs and approximate probability-flow ODEs, enabling principled and directly comparable conditional samplers. To reduce the induced ODE--SDE discrepancy, we introduce a log-Fokker--Planck residual regularization that improves ODE sampling quality. Experiments on conditional image generation tasks demonstrate competitive performance and support the effectiveness of the plug-in conditioning view. Additional ODE--SDE comparison experiments show that the log-Fokker--Planck residual regularization improves deterministic ODE sampling.

Problem

• conditioning
• diffusion models
• score-based
• plug-in correction
• conditional generation

Innovation

• multi-speed joint diffusion
• plug-in correction term
• log-Fokker--Planck residual regularization