Diffusion models (DMs) produce very detailed and high-quality images.

Of course, spotting near duplicates of training observations is only possible because these models yield realistic samples. This section describes additional details of the data sets, model architectures, and experimental setup. CIFAR-10 contains color images from 10 different categories and does not require further preprocessing. For CIFAR-10 and CelebA we used random horizontal flips during training as data augmentation. Full details of the model architecture are given in Table 1.

The limitations of rote memorization remind us of the human learning process of "learn by analogy"

We present the first investigation into these questions.

Diffusion models (DMs) produce very detailed and high-quality images.

In particular, it is vital to minimize a model's memorization of sensitive