arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2312.07586cs.CVcs.AIcs.LGphysics.data-an
收藏

特征引导:大引导尺度下扩散模型的非线性修正

Characteristic Guidance: Non-linear Correction for Diffusion Model at Large Guidance Scale

Hong Kong University of Science and Technology(香港科技大学);Southern University of Science and Technology(南方科技大学)

Candi Zheng, Yuan Lan

AI总结:

针对无分类器引导在大引导尺度下忽略非线性效应的问题,提出特征引导方法,通过第一性原理非线性修正使扩散模型遵循Fokker-Planck方程,无需训练即可增强语义特征并减轻图像生成不规则性。

摘要:

去噪扩散概率模型(DDPM)的流行引导方法将不同的条件模型线性组合在一起,以提供对样本的增强控制。然而,这种方法忽略了在引导尺度较大时变得显著的非线性效应。为了解决这个问题,我们提出了特征引导(characteristic guidance),这是一种为无分类器引导提供第一性原理非线性修正的引导方法。这种修正以无需训练且与现有采样方法兼容的方式,迫使受引导的 DDPM 遵循扩散过程的 Fokker-Planck (FP) 方程。实验表明,特征引导增强了提示的语义特征并减轻了图像生成中的不规则性,证明了其在从模拟磁相变到潜空间采样等多种应用中的有效性。

英文摘要:

Popular guidance for denoising diffusion probabilistic model (DDPM) linearly combines distinct conditional models together to provide enhanced control over samples. However, this approach overlooks nonlinear effects that become significant when guidance scale is large. To address this issue, we propose characteristic guidance, a guidance method that provides first-principle non-linear correction for classifier-free guidance. Such correction forces the guided DDPMs to respect the Fokker-Planck (FP) equation of diffusion process, in a way that is training-free and compatible with existing sampling methods. Experiments show that characteristic guidance enhances semantic characteristics of prompts and mitigate irregularities in image generation, proving effective in diverse applications ranging from simulating magnet phase transitions to latent space sampling.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑