PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911679052251136 |
|---|---|
| author | Wu, Hao Xu, Fan Lu, Yuxu Zhao, Penghao Zhang, Fan Jia, Hao Liang, Yuxuan Gou, Ruijian Wen, Qingsong Wu, Xian Huang, Xiaomeng Gao, Yuan |
| author_facet | Wu, Hao Xu, Fan Lu, Yuxu Zhao, Penghao Zhang, Fan Jia, Hao Liang, Yuxuan Gou, Ruijian Wen, Qingsong Wu, Xian Huang, Xiaomeng Gao, Yuan |
| contents | Coupled spatiotemporal forecasting is important for predicting the future evolution of multiple interacting dynamical systems, such as in climate models. However, existing methods are severely constrained by the persistent bottleneck of compounding errors. In coupled systems, errors from each subsystem simulator propagate and amplify one another, a phenomenon we term Reciprocal Error Amplification, leading to a rapid collapse of long-range predictions. To address this challenge, we propose a universal framework called PnP-Corrector (Plug-and-Play Corrector). The core idea of our framework is to decouple the physical simulation from the error correction process: it freezes pre-trained physics simulation engines and exclusively trains a correction agent to proactively counteract the systematic biases emerging from the coupled system. Furthermore, we design an efficient predictive model architecture, DSLCast, to serve as the backbone of this framework. Extensive experiments demonstrate that our method significantly enhances the long-term stability and accuracy of coupled forecasting systems. For instance, in the challenging task of a 300-day global ocean-atmosphere coupled forecast, our PnP-Corrector framework reduces the prediction error of the baseline model by 29% and surpasses state-of-the-art models on several key metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_08935 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting Wu, Hao Xu, Fan Lu, Yuxu Zhao, Penghao Zhang, Fan Jia, Hao Liang, Yuxuan Gou, Ruijian Wen, Qingsong Wu, Xian Huang, Xiaomeng Gao, Yuan Artificial Intelligence Machine Learning Coupled spatiotemporal forecasting is important for predicting the future evolution of multiple interacting dynamical systems, such as in climate models. However, existing methods are severely constrained by the persistent bottleneck of compounding errors. In coupled systems, errors from each subsystem simulator propagate and amplify one another, a phenomenon we term Reciprocal Error Amplification, leading to a rapid collapse of long-range predictions. To address this challenge, we propose a universal framework called PnP-Corrector (Plug-and-Play Corrector). The core idea of our framework is to decouple the physical simulation from the error correction process: it freezes pre-trained physics simulation engines and exclusively trains a correction agent to proactively counteract the systematic biases emerging from the coupled system. Furthermore, we design an efficient predictive model architecture, DSLCast, to serve as the backbone of this framework. Extensive experiments demonstrate that our method significantly enhances the long-term stability and accuracy of coupled forecasting systems. For instance, in the challenging task of a 300-day global ocean-atmosphere coupled forecast, our PnP-Corrector framework reduces the prediction error of the baseline model by 29% and surpasses state-of-the-art models on several key metrics. |
| title | PnP-Corrector: A Universal Correction Framework for Coupled Spatiotemporal Forecasting |
| topic | Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2605.08935 |