English

A Survey on Cache Methods in Diffusion Models: Toward Efficient Multi-Modal Generation

Machine Learning 2025-11-04 v3 Artificial Intelligence Computer Vision and Pattern Recognition

Abstract

Diffusion Models have become a cornerstone of modern generative AI for their exceptional generation quality and controllability. However, their inherent \textit{multi-step iterations} and \textit{complex backbone networks} lead to prohibitive computational overhead and generation latency, forming a major bottleneck for real-time applications. Although existing acceleration techniques have made progress, they still face challenges such as limited applicability, high training costs, or quality degradation. Against this backdrop, \textbf{Diffusion Caching} offers a promising training-free, architecture-agnostic, and efficient inference paradigm. Its core mechanism identifies and reuses intrinsic computational redundancies in the diffusion process. By enabling feature-level cross-step reuse and inter-layer scheduling, it reduces computation without modifying model parameters. This paper systematically reviews the theoretical foundations and evolution of Diffusion Caching and proposes a unified framework for its classification and analysis. Through comparative analysis of representative methods, we show that Diffusion Caching evolves from \textit{static reuse} to \textit{dynamic prediction}. This trend enhances caching flexibility across diverse tasks and enables integration with other acceleration techniques such as sampling optimization and model distillation, paving the way for a unified, efficient inference framework for future multimodal and interactive applications. We argue that this paradigm will become a key enabler of real-time and efficient generative AI, injecting new vitality into both theory and practice of \textit{Efficient Generative Intelligence}.

Keywords

Cite

@article{arxiv.2510.19755,
  title  = {A Survey on Cache Methods in Diffusion Models: Toward Efficient Multi-Modal Generation},
  author = {Jiacheng Liu and Xinyu Wang and Yuqi Lin and Zhikai Wang and Peiru Wang and Peiliang Cai and Qinming Zhou and Zhengan Yan and Zexuan Yan and Zhengyi Shi and Chang Zou and Yue Ma and Linfeng Zhang},
  journal= {arXiv preprint arXiv:2510.19755},
  year   = {2025}
}

Comments

22 pages,2 figures

R2 v1 2026-07-01T07:00:07.757Z