Diffusion models can sometimes reproduce examples from their training data, a phenomenon known as memorization. Existing memorization detection methods for unconditional diffusion models typically ope...
The paper demonstrates that diffusion models encode memorization in the geometry of their learned energy landscape before it becomes visible in generated samples. By analyzing score divergence and basin volume, the authors show that localized basins form around training images early in training, and that cyclic denoising can recover these images even when one‑shot generation shows no copies. Experiments on Gaussian mixtures, CelebA, CIFAR‑10, and Stable Diffusion v1.4 confirm that basin geometry reveals memorization across various settings.
arXiv:2609.00955v1 Announce Type: new Abstract: Diffusion models achieve strong image generation quality but incur high iterative denoising costs. Analog compute-in-memory (CIM) can accelerate matrix...
The paper investigates why diffusion models, despite their practical success, can fully memorize training data when optimal. It shows that as training proceeds, the denoising objective causes overfitting at intermediate noise levels, creating a gap between validation and training performance. In a 2D toy model, this gap arises from the interplay between model error and the data distribution’s support density, leading to a smooth, generalizing flow field rather than exact recall of training points. The study also reveals that predictions from intermediate inference states become increasingly similar to training data in feature space, even as overall similarity to validation data remains unchanged.
arXiv:2610.19042v1 Announce Type: new Abstract: Understanding the mechanism behind why diffusion models memorize their training data remains an open question in generative modeling. In this paper, we...
arXiv:2609.37974v1 Announce Type: cross Abstract: Masked diffusion models (MDMs) generate text by unmasking several tokens per step, but they are trained and sampled under different conditions. The m...
Understanding the mechanism behind why diffusion models memorize their training data remains an open question in generative modeling. In this paper, we examine this question from two perspectives: (a)...
arXiv:2610.02780v1 Announce Type: new Abstract: Diffusion models can reach useful sample quality before copying training examples, but fast optimization can compress this generalization window by acc...
arXiv:2510. 17917v2 Announce Type: replace-cross Abstract: Data unlearning aims to remove the influence of specific training samples from a trained model.
arXiv:2608. 14632v1 Announce Type: cross Abstract: Diffusion large language models (D-LLMs) have emerged as a promising paradigm for text generation.
arXiv:2606. 20416v1 Announce Type: new Abstract: Diffusion models rely heavily on explicit timestep embeddings to modulate the denoising process across various noise scales.
arXiv:2609.01449v1 Announce Type: new Abstract: Diffusion models and recursive reasoners are both iterative, but they carry information across iterations differently. We add a persistent hidden state...