改进扩散模型的条件生成，让引导更准确高效。
REG: Rectified Gradient Guidance for Conditional Diffusion Models
- 用联合分布替代边缘分布，理论更合理
- 在图像生成任务中显著提升FID和得分
- 适用于各类条件扩散模型，易集成
引导技术虽简单有效，但实际实现与理论初衷存在偏差。本文通过证明原有缩放边缘分布目标在理论上不成立，提出采用有效的缩放联合分布目标进行修正。进一步表明，现有引导方法仅为无未来预知约束下的不可行最优解的近似。基于此理论分析，我们提出一种通用增强方法——校正梯度引导（REG），可提升现有引导方法性能。1D与2D实验表明，REG比以往方法更逼近最优解；在类条件ImageNet及文本到图像生成任务中，引入REG后，各类设置下均一致改善了FID、Inception和CLIP分数。
原文摘要 · Abstract (English)
Guidance techniques are simple yet effective for improving conditional generation in diffusion models. Albeit their empirical success, the practical implementation of guidance diverges significantly from its theoretical motivation. In this paper, we reconcile this discrepancy by replacing the scaled marginal distribution target, which we prove theoretically invalid, with a valid scaled joint distribution objective. Additionally, we show that the established guidance implementations are approximations to the intractable optimal solution under no future foresight constraint. Building on these theoretical insights, we propose rectified gradient guidance (REG), a versatile enhancement designed to boost the performance of existing guidance methods. Experiments on 1D and 2D demonstrate that REG provides a better approximation to the optimal solution than prior guidance techniques, validating the proposed theoretical framework. Extensive experiments on class-conditional ImageNet and text-to-image generation tasks show that incorporating REG consistently improves FID and Inception/CLIP scores across various settings compared to its absence.
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。