C^2FG：通过分数差异分析控制无分类器引导
C^2FG: Control Classifier-Free Guidance via Score Discrepancy Analysis
把 CFG 的引导权重从固定/启发式改为随扩散时间步指数衰减的控制函数，且免训练、即插即用，可直接替换现有条件扩散采样器中的 guidance 调度，用于提升条件生成质量并与其他引导策略正交叠加。
原文依据 a novel, training-free, and plug-in method that aligns the guidance strength with the diffusion dynamics via an exponential decay control function
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论无分类器引导（CFG）是现代条件扩散模型的基石，然而其对固定或启发式动态引导权重的依赖主要是经验性的，并且忽视了扩散过程的内在动态。本文对无分类器引导进行了严格的理论分析。具体而言，我们基于扩散过程，建立了不同时间步上条件分布与无条件分布之间分数差异的严格上界。这一发现解释了固定权重策略的局限性，并为时间相关的引导奠定了有原则的基础。受此启发，我们提出了控制无分类器引导（C^2FG），这是一种新颖的、免训练的、即插即用的方法，通过指数衰减控制函数使引导强度与扩散动态相匹配。大量实验表明，C^2FG 是有效的，可广泛适用于多种生成任务，并且与现有策略具有正交性。
英文原文摘要 1054 字符
Classifier-Free Guidance (CFG) is a cornerstone of modern conditional diffusion models, yet its reliance on the fixed or heuristic dynamic guidance weight is predominantly empirical and overlooks the inherent dynamics of the diffusion process. In this paper, we provide a rigorous theoretical analysis of the Classifier-Free Guidance. Specifically, we establish strict upper bounds on the score discrepancy between conditional and unconditional distributions at different timesteps based on the diffusion process.This finding explains the limitations of fixed-weight strategies and establishes a principled foundation for time-dependent guidance. Motivated by this insight, we introduce **Control Classifier-Free Guidance (C^2FG)**, a novel, training-free, and plug-in method that aligns the guidance strength with the diffusion dynamics via an exponential decay control function. Extensive experiments demonstrate that C^2FG is effective and broadly applicable across diverse generative tasks, while also exhibiting orthogonality to existing strategies.