SOURCE-LINKED INTELLIGENCE

Double Descent and Malign Overfitting in Diffusion Models

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Conventional wisdom in deep learning holds that overparameterization---having more parameters $p$ than training samples $n$---is benign: larger models generalize better and, even without regularization, interpolating models generalize well, the test error following a double-descent curve. One might expect the same benign overfitting for diffusion models, whose training reduces to regression, i.e. to minimizing a quadratic score-matching loss. Yet the opposite is observed: overfitting here is catastrophic, driving the model into a memorization regime. We resolve this paradox by combining experi

Read original source ↗ Open in workspace

recordType

paper

region

Global