CMoS: Rethinking Time Series Prediction Through the Lens of Chunk-wise Spatial Correlations

Haotian Si, Changhua Pei, Jianhui Li, Dan Pei, Gaogang Xie
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:55493-55510, 2025.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-si25a, title = {{CM}o{S}: Rethinking Time Series Prediction Through the Lens of Chunk-wise Spatial Correlations}, author = {Si, Haotian and Pei, Changhua and Li, Jianhui and Pei, Dan and Xie, Gaogang}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {55493--55510}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/si25a/si25a.pdf}, url = {https://proceedings.mlr.press/v267/si25a.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T CMoS: Rethinking Time Series Prediction Through the Lens of Chunk-wise Spatial Correlations %A Haotian Si %A Changhua Pei %A Jianhui Li %A Dan Pei %A Gaogang Xie %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-si25a %I PMLR %P 55493--55510 %U https://proceedings.mlr.press/v267/si25a.html %V 267 %X 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.
APA
Si, H., Pei, C., Li, J., Pei, D. & Xie, G.. (2025). CMoS: Rethinking Time Series Prediction Through the Lens of Chunk-wise Spatial Correlations. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:55493-55510 Available from https://proceedings.mlr.press/v267/si25a.html.

Related Material