English

Stabilizing Self-Consuming Diffusion Models with Latent Space Filtering

Machine Learning 2025-11-18 v1

Abstract

As synthetic data proliferates across the Internet, it is often reused to train successive generations of generative models. This creates a ``self-consuming loop" that can lead to training instability or \textit{model collapse}. Common strategies to address the issue -- such as accumulating historical training data or injecting fresh real data -- either increase computational cost or require expensive human annotation. In this paper, we empirically analyze the latent space dynamics of self-consuming diffusion models and observe that the low-dimensional structure of latent representations extracted from synthetic data degrade over generations. Based on this insight, we propose \textit{Latent Space Filtering} (LSF), a novel approach that mitigates model collapse by filtering out less realistic synthetic data from mixed datasets. Theoretically, we present a framework that connects latent space degradation to empirical observations. Experimentally, we show that LSF consistently outperforms existing baselines across multiple real-world datasets, effectively mitigating model collapse without increasing training cost or relying on human annotation.

Keywords

Cite

@article{arxiv.2511.12742,
  title  = {Stabilizing Self-Consuming Diffusion Models with Latent Space Filtering},
  author = {Zhongteng Cai and Yaxuan Wang and Yang Liu and Xueru Zhang},
  journal= {arXiv preprint arXiv:2511.12742},
  year   = {2025}
}

Comments

Accepted by AAAI-26

R2 v1 2026-07-01T07:40:00.796Z