SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement
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_ | 1866917238555017216 |
|---|---|
| author | Qiu, Xiangfei Liu, Xvyuan Shen, Tianen Wu, Xingjian Cheng, Hanyin Yang, Bin Hu, Jilin |
| author_facet | Qiu, Xiangfei Liu, Xvyuan Shen, Tianen Wu, Xingjian Cheng, Hanyin Yang, Bin Hu, Jilin |
| contents | Time series forecasting is important in many fields that require accurate predictions for decision-making. Patching techniques, commonly used and effective in time series modeling, help capture temporal dependencies by dividing the data into patches. However, existing patch-based methods fail to dynamically select patches and typically use all patches during the prediction process. In real-world time series, there are often low-quality issues during data collection, such as missing values, distribution shifts, anomalies and white noise, which may cause some patches to contain low-quality information, negatively impacting the prediction results. To address this issue, this study proposes a robust time series forecasting framework called SEER. Firstly, we propose an Augmented Embedding Module, which improves patch-wise representations using a Mixture-of-Experts (MoE) architecture and obtains series-wise token representations through a channel-adaptive perception mechanism. Secondly, we introduce a Learnable Patch Replacement Module, which enhances forecasting robustness and model accuracy through a two-stage process: 1) a dynamic filtering mechanism eliminates negative patch-wise tokens; 2) a replaced attention module substitutes the identified low-quality patches with global series-wise token, further refining their representations through a causal attention mechanism. Comprehensive experimental results demonstrate the SOTA performance of SEER. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_00589 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement Qiu, Xiangfei Liu, Xvyuan Shen, Tianen Wu, Xingjian Cheng, Hanyin Yang, Bin Hu, Jilin Machine Learning Time series forecasting is important in many fields that require accurate predictions for decision-making. Patching techniques, commonly used and effective in time series modeling, help capture temporal dependencies by dividing the data into patches. However, existing patch-based methods fail to dynamically select patches and typically use all patches during the prediction process. In real-world time series, there are often low-quality issues during data collection, such as missing values, distribution shifts, anomalies and white noise, which may cause some patches to contain low-quality information, negatively impacting the prediction results. To address this issue, this study proposes a robust time series forecasting framework called SEER. Firstly, we propose an Augmented Embedding Module, which improves patch-wise representations using a Mixture-of-Experts (MoE) architecture and obtains series-wise token representations through a channel-adaptive perception mechanism. Secondly, we introduce a Learnable Patch Replacement Module, which enhances forecasting robustness and model accuracy through a two-stage process: 1) a dynamic filtering mechanism eliminates negative patch-wise tokens; 2) a replaced attention module substitutes the identified low-quality patches with global series-wise token, further refining their representations through a causal attention mechanism. Comprehensive experimental results demonstrate the SOTA performance of SEER. |
| title | SEER: Transformer-based Robust Time Series Forecasting via Automated Patch Enhancement and Replacement |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2602.00589 |