Recursive Self-Improvement (RSI) enables intelligence systems to autonomously refine their capabilities. This paper explores the application of RSI in text-to-image diffusion models, addressing the challenge of training collapse caused by synthetic data. We identify two key factors contributing to this collapse: the lack of perceptual alignment and the accumulation of generative hallucinations. To mitigate these issues, we propose three strategies: (1) a prompt construction and filtering pipeline designed to facilitate the generation of perceptual aligned data, (2) a preference sampling method to identify human-preferred samples and filter out generative hallucinations, and (3) a distribution-based weighting scheme to penalize selected samples with hallucinatory errors. Our extensive experiments validate the effectiveness of these approaches.
@article{arxiv.2502.09963,
title = {Generating on Generated: An Approach Towards Self-Evolving Diffusion Models},
author = {Xulu Zhang and Xiaoyong Wei and Jinlin Wu and Jiaxin Wu and Zhaoxiang Zhang and Zhen Lei and Qing Li},
journal= {arXiv preprint arXiv:2502.09963},
year = {2025}
}