Implicit Regularisation in Diffusion Models: An Algorithm-Dependent Generalisation Analysis

ICLR 2026 Conference Submission19483 Authors

Keywords: diffusion models, generalization, stability, algorithm dependent, learning theory

TL;DR: We propose a framework for analysing the generalisation properties of diffusion models. We use it to identify and analyse three key sources of implicit regularisation: early stopping, coarse sampler discretisation, and optimisation with SGD.

Abstract: The success of denoising diffusion models raises important questions regarding their generalisation behaviour, particularly in high-dimensional settings. Notably, it has been shown that when training and sampling are performed perfectly, these models memorise training data—implying that some form of regularisation is essential for generalisation. Existing theoretical analyses primarily rely on algorithm-independent techniques such as uniform convergence, heavily utilising model structure to obtain generalisation bounds. In this work, we instead leverage the algorithmic aspects that promote generalisation in diffusion models, developing a general theory of algorithm-dependent generalisation for this setting. Borrowing from the framework of algorithmic stability, we introduce the notion of score stability, which quantifies the sensitivity of score-matching algorithms to dataset perturbations. We derive generalisation bounds in terms of score stability, and apply our framework to several fundamental learning settings, identifying sources of regularisation. In particular, we consider denoising score matching with early stopping (denoising regularisation), sampler-wide coarse discretisation (sampler regularisation), and optimising with SGD (optimisation regularisation). By grounding our analysis in algorithmic properties rather than model structure, we identify multiple sources of implicit regularisation unique to diffusion models that have so far been overlooked in the literature.

Primary Area: generative models

Submission Number: 19483