解决扩散模型引导过强导致的过饱和问题，提升图像质量。
EP-CFG: Energy-Preserving Classifier-Free Guidance
- 通过保持条件预测的能量分布，动态调整引导输出
- 在不同引导强度下均保持自然图像质量和细节
- 计算开销极小，适合实际部署
分类器自由引导（CFG）广泛用于扩散模型，但在高引导强度下常引入过度对比和过饱和伪影。本文提出EP-CFG（能量保持的分类器自由引导），通过在引导过程中保持条件预测的能量分布来解决该问题。方法在每一步去噪时简单地将引导输出的能量重缩放至与条件预测一致，可选的鲁棒变体进一步抑制伪影。实验表明，EP-CFG在不同引导强度下均保持自然图像质量与细节，同时保留CFG的语义对齐优势，且计算开销极低。
原文摘要 · Abstract (English)
Classifier-free guidance (CFG) is widely used in diffusion models but often introduces over-contrast and over-saturation artifacts at higher guidance strengths. We present EP-CFG (Energy-Preserving Classifier-Free Guidance), which addresses these issues by preserving the energy distribution of the conditional prediction during the guidance process. Our method simply rescales the energy of the guided output to match that of the conditional prediction at each denoising step, with an optional robust variant for improved artifact suppression. Through experiments, we show that EP-CFG maintains natural image quality and preserves details across guidance strengths while retaining CFG's semantic alignment benefits, all with minimal computational overhead.
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。