流匹配中的无分类器引导：非自治势、过冲与后验均值控制

中文摘要 AI 辅助

无分类器引导（CFG）提升了流匹配中的条件生成质量，但强引导会扭曲生成分布并降低多样性。我们通过将流匹配视为时变梯度流，并刻画CFG如何重塑其底层势，对此行为给出了几何解释。该视角解释了为何更强的对齐可能伴随均值位移和轨迹集中，并启发通过模型隐含的终端后验均值来控制引导。因此，我们提出了后验均值截断CFG（PMC-CFG），这是一种免训练、逐样本的方法，无需额外网络评估即可自适应地保留最强可行的引导。在合成基准和大规模图像生成基准上的实验表明，PMC-CFG在提升对齐-多样性权衡的同时，限制了引导引起的扭曲和集中，尤其在标称引导较大时收益显著。

英文摘要

Classifier-free guidance (CFG) improves conditional generation in Flow Matching, but strong guidance can distort the generated distribution and reduce diversity. We provide a geometric account of this behavior by viewing Flow Matching as a time-varying gradient flow and characterizing how CFG reshapes its underlying potential. This view explains how stronger alignment can be accompanied by mean displacement and trajectory concentration, and motivates controlling guidance through the model-implied terminal posterior mean. We therefore propose Posterior-Mean-Capped CFG (PMC-CFG), a training-free, per-sample method that adaptively retains the strongest feasible guidance without additional network evaluations. Experiments on synthetic and large-scale image-generation benchmarks show that PMC-CFG limits guidance-induced distortion and concentration while improving the alignment--diversity trade-off, with particularly strong benefits when nominal guidance is large.

发表机构

• School of Mathematical Sciences, Shanghai Jiao Tong University（上海交通大学数学科学学院）
• Institute of Natural Sciences, MOE-LSC, Shanghai Jiao Tong University（上海交通大学自然科学研究院）
• Qing Yuan Research Institute, Shanghai Jiao Tong University（上海交通大学清源研究院）
• CMA-Shanghai, Shanghai Jiao Tong University（上海交通大学上海数学中心）

机构由 AI 辅助整理，请以论文原文为准。