Recent advances in lightweight time series forecasting models suggest the inherent simplicity of time series forecasting tasks. In this paper, we present CMoS, a super-lightweight time series forecasting model. Instead of learning the embedding of the shapes, CMoS directly models the spatial correlations between different time series chunks. Additionally, we introduce a Correlation Mixing technique that enables the model to capture diverse spatial correlations with minimal parameters, and an optional Periodicity Injection technique to ensure faster convergence. Despite utilizing as low as 1% of the lightweight model DLinear's parameters count, experimental results demonstrate that CMoS outperforms existing state-of-the-art models across multiple datasets. Furthermore, the learned weights of CMoS exhibit great interpretability, providing practitioners with valuable insights into temporal structures within specific application scenarios.
@article{arxiv.2505.19090,
title = {CMoS: Rethinking Time Series Prediction Through the Lens of Chunk-wise Spatial Correlations},
author = {Haotian Si and Changhua Pei and Jianhui Li and Dan Pei and Gaogang Xie},
journal= {arXiv preprint arXiv:2505.19090},
year = {2025}
}
Comments
Accepted by Forty-second International Conference on Machine Learning (ICML'25)