← Back to all news
arXiv Machine Learning By Druv Pai, Dejan Stancevic, Yi Ma, Luca Ambrogioni

Information-Theoretic Scaling Laws for Generative Diffusion Training

Read the original on arXiv Machine Learning →

The paper introduces an information-theoretic measure called model information to quantify how much a diffusion model’s outputs depend on its specific training dataset. By analyzing how this measure scales with model size and dataset size, the authors identify two distinct regimes: a nonlinear regime where memorization transitions to imperfect generalization, and a linear regime where generalization improves and sample quality rises. They propose a scaling law that captures these regimes and demonstrate that it can predict memorization and sample quality beyond the fitted data range.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

arXiv Machine Learning
Jun 10

The Emergence of Reproducibility and Generalizability in Diffusion Models

arXiv:2310. 05264v5 Announce Type: replace Abstract: In this work, we investigate an intriguing and prevalent phenomenon of diffusion models which we term as "consistent model reproducibility": given the same starting noise input and a deterministic sampler, different diffusion models often yield remarkably similar outputs.

By Huijie Zhang, Jinfan Zhou, Yifu Lu, Minzhe Guo, Peng Wang, Liyue Shen, Qing Qu
diffusionfine-tuningsafety
More like this →
arXiv AI
Sep 3

Language Diffusion Models are Associative Memories Capable of Retrieving Unseen Data

The paper investigates when language diffusion models, specifically Uniform-based Discrete Diffusion Models (UDDMs), shift from memorizing training data to generalizing to new data. It shows that UDDMs act as associative memories, forming basins of attraction around stored examples without requiring an explicit energy function. By measuring token recovery and conditional entropy, the authors identify a sharp transition governed by training set size, where memorization (vanishing entropy) gives way to generalization (finite entropy).

By Bao Pham, Mohammed J. Zaki, Luca Ambrogioni, Dmitry Krotov, Matteo Negri
diffusion
More like this →
arXiv Machine Learning
Oct 9

Beyond Sample Copying: Structural Memorization in Diffusion Models

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.

By Tim Kaiser, Markus Kollmann
diffusion
More like this →