时序对齐引导：扩散模型中的流形采样

摘要

扩散模型作为生成模型已取得显著成功。然而，即便是训练有素的模型，在生成过程中也可能累积误差。当施加任意引导以驱使样本朝向特定属性时，这些误差尤为突出，常常损害样本的保真度。本文提出了一种通用解决方案，以应对扩散模型中观察到的离流形现象。我们的方法利用时间预测器来估计每一步与期望数据流形的偏差，发现时间间隔越大，生成质量越低。随后，我们设计了一种新颖的引导机制——“时序对齐引导”（TAG），在生成过程中的每一步将样本吸引回期望的流形。通过大量实验，我们证明TAG能够在每一步持续生成与期望流形紧密对齐的样本，从而在各种下游任务中显著提升生成质量。

English

Diffusion models have achieved remarkable success as generative models. However, even a well-trained model can accumulate errors throughout the generation process. These errors become particularly problematic when arbitrary guidance is applied to steer samples toward desired properties, which often breaks sample fidelity. In this paper, we propose a general solution to address the off-manifold phenomenon observed in diffusion models. Our approach leverages a time predictor to estimate deviations from the desired data manifold at each timestep, identifying that a larger time gap is associated with reduced generation quality. We then design a novel guidance mechanism, `Temporal Alignment Guidance' (TAG), attracting the samples back to the desired manifold at every timestep during generation. Through extensive experiments, we demonstrate that TAG consistently produces samples closely aligned with the desired manifold at each timestep, leading to significant improvements in generation quality across various downstream tasks.

时序对齐引导：扩散模型中的流形采样

Temporal Alignment Guidance: On-Manifold Sampling in Diffusion Models

摘要

Support