通过热力学奇点检测扩散模型的记忆化现象
Detection of Memorization in Diffusion Models Through Thermodynamic Singularity
扩散模型老是背训练图？这篇论文给出一个热力学判据，能直接检测模型是不是在记忆化，做扩散模型的同学可以看看。
扩散模型的训练目标本应是学习底层密度函数的分布，但记忆化会让反向过程转而学习单个训练样本。这篇 arXiv 论文对记忆化的反向过程给出了分析性刻画，并提出一个检测热力学奇点的判据。实验将该判据应用于多种被刻意设计为促进或抑制记忆化的神经网络，能可靠区分正常学习与记忆化的网络。
Detection of Memorization in Diffusion Models Through Thermodynamic Singularity
In diffusion models, proper distributional learning entails reversing a diffusion process over underlying density functions, whereas a key failure is memorization, in which the reverse process instead learns diffusion from individual training samples. We provide an analytical characterization of such memorized reverse processes, along with a criterion that detects the associated thermodynamic singularity. The proposed criterion reliably distinguishes between properly learned and memorizing networks across a range of neural network models designed to promote or mitigate memorization.
