Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918436228038656 |
|---|---|
| author | Liu, Yong Su, Xingjian Wang, Shiyu Zhang, Haoran Liu, Haixuan Wang, Yuxuan Ye, Zhou Xiang, Yang Wang, Jianmin Long, Mingsheng |
| author_facet | Liu, Yong Su, Xingjian Wang, Shiyu Zhang, Haoran Liu, Haixuan Wang, Yuxuan Ye, Zhou Xiang, Yang Wang, Jianmin Long, Mingsheng |
| contents | We introduce Timer-S1, a strong Mixture-of-Experts (MoE) time series foundation model with 8.3B total parameters, 0.75B activated parameters for each token, and a context length of 11.5K. To overcome the scalability bottleneck in existing pre-trained time series foundation models, we perform Serial Scaling in three dimensions: model architecture, dataset, and training pipeline. Timer-S1 integrates sparse TimeMoE blocks and generic TimeSTP blocks for Serial-Token Prediction (STP), a generic training objective that adheres to the serial nature of forecasting. The proposed paradigm introduces serial computations to improve long-term predictions while avoiding costly rolling-style inference and pronounced error accumulation in the standard next-token prediction. Pursuing a high-quality and unbiased training dataset, we curate TimeBench, a corpus with one trillion time points, and apply meticulous data augmentation to mitigate predictive bias. We further pioneer a post-training stage, including continued pre-training and long-context extension, to enhance short-term and long-context performance. Evaluated on the large-scale GIFT-Eval leaderboard, Timer-S1 achieves state-of-the-art forecasting performance, attaining the best MASE and CRPS scores as a pre-trained model. Timer-S1 is released to facilitate further research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_04791 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling Liu, Yong Su, Xingjian Wang, Shiyu Zhang, Haoran Liu, Haixuan Wang, Yuxuan Ye, Zhou Xiang, Yang Wang, Jianmin Long, Mingsheng Artificial Intelligence We introduce Timer-S1, a strong Mixture-of-Experts (MoE) time series foundation model with 8.3B total parameters, 0.75B activated parameters for each token, and a context length of 11.5K. To overcome the scalability bottleneck in existing pre-trained time series foundation models, we perform Serial Scaling in three dimensions: model architecture, dataset, and training pipeline. Timer-S1 integrates sparse TimeMoE blocks and generic TimeSTP blocks for Serial-Token Prediction (STP), a generic training objective that adheres to the serial nature of forecasting. The proposed paradigm introduces serial computations to improve long-term predictions while avoiding costly rolling-style inference and pronounced error accumulation in the standard next-token prediction. Pursuing a high-quality and unbiased training dataset, we curate TimeBench, a corpus with one trillion time points, and apply meticulous data augmentation to mitigate predictive bias. We further pioneer a post-training stage, including continued pre-training and long-context extension, to enhance short-term and long-context performance. Evaluated on the large-scale GIFT-Eval leaderboard, Timer-S1 achieves state-of-the-art forecasting performance, attaining the best MASE and CRPS scores as a pre-trained model. Timer-S1 is released to facilitate further research. |
| title | Timer-S1: A Billion-Scale Time Series Foundation Model with Serial Scaling |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2603.04791 |