从分类器视角重新理解扩散模型的条件生成机制
Studying Classifier(-Free) Guidance From a Classifier-Centric Perspective
- 通过分析分类器引导的原理，揭示条件生成的本质是避开决策边界
- 在1维和高维数据上验证：两类引导均使轨迹远离难学区域
- 提出用流匹配修复分布差距，提升模型在边界附近的生成效果
Classifier-free guidance已成为去噪扩散模型中条件生成的标准方法。然而，对其全面理解仍不充分。本文开展实证研究，从分类器引导的根源出发，追溯其推导的关键假设，并系统分析分类器的作用。在1维数据上发现，分类器引导与classifier-free引导均通过将去噪扩散路径推向决策边界之外来实现条件生成，即那些条件信息纠缠、难以学习的区域。为验证该分类器中心视角在高维数据上的有效性，我们测试了针对预训练扩散模型学习分布与真实数据分布之间差距（尤其是在决策边界附近）设计的流匹配后处理步骤是否能提升性能。多种数据集上的实验验证了该分类器中心视角的合理性。
原文摘要 · Abstract (English)
Classifier-free guidance has become a staple for conditional generation with denoising diffusion models. However, a comprehensive understanding of classifier-free guidance is still missing. In this work, we carry out an empirical study to provide a fresh perspective on classifier-free guidance. Concretely, instead of solely focusing on classifier-free guidance, we trace back to the root, i.e., classifier guidance, pinpoint the key assumption for the derivation, and conduct a systematic study to understand the role of the classifier. On 1D data, we find that both classifier guidance and classifier-free guidance achieve conditional generation by pushing the denoising diffusion trajectories away from decision boundaries, i.e., areas where conditional information is usually entangled and is hard to learn. To validate this classifier-centric perspective on high-dimensional data, we assess whether a flow-matching postprocessing step that is designed to narrow the gap between a pre-trained diffusion model's learned distribution and the real data distribution, especially near decision boundaries, can improve the performance. Experiments on various datasets verify our classifier-centric understanding.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。