PA-Net: Precipitation-Adaptive Mixture-of-Experts for Long-Tail Rainfall Nowcasting
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918388052262912 |
|---|---|
| author | Xiao, Xinyu Lei, Sen Liu, Eryun Xiang, Shiming Li, Hao Yuan, Cheng Qi, Yuan Jin, Qizhao |
| author_facet | Xiao, Xinyu Lei, Sen Liu, Eryun Xiang, Shiming Li, Hao Yuan, Cheng Qi, Yuan Jin, Qizhao |
| contents | Precipitation nowcasting is vital for flood warning, agricultural management, and emergency response, yet two bottlenecks persist: the prohibitive cost of modeling million-scale spatiotemporal tokens from multi-variate atmospheric fields, and the extreme long-tailed rainfall distribution where heavy-to-torrential events -- those of greatest societal impact -- constitute fewer than 0.1% of all samples. We propose the Precipitation-Adaptive Network (PA-Net), a Transformer framework whose computational budget is explicitly governed by rainfall intensity. Its core component, Precipitation-Adaptive MoE (PA-MoE), dynamically scales the number of activated experts per token according to local precipitation magnitude, channeling richer representational capacity toward the rare yet critical heavy-rainfall tail. A Dual-Axis Compressed Latent Attention mechanism factorizes spatiotemporal attention with convolutional reduction to manage massive context lengths, while an intensity-aware training protocol progressively amplifies learning signals from extreme-rainfall samples. Experiment on ERA5 demonstrate consistent improvements over state-of-the-art baselines, with particularly significant gains in heavy-rain and rainstorm regimes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_13818 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | PA-Net: Precipitation-Adaptive Mixture-of-Experts for Long-Tail Rainfall Nowcasting Xiao, Xinyu Lei, Sen Liu, Eryun Xiang, Shiming Li, Hao Yuan, Cheng Qi, Yuan Jin, Qizhao Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning Precipitation nowcasting is vital for flood warning, agricultural management, and emergency response, yet two bottlenecks persist: the prohibitive cost of modeling million-scale spatiotemporal tokens from multi-variate atmospheric fields, and the extreme long-tailed rainfall distribution where heavy-to-torrential events -- those of greatest societal impact -- constitute fewer than 0.1% of all samples. We propose the Precipitation-Adaptive Network (PA-Net), a Transformer framework whose computational budget is explicitly governed by rainfall intensity. Its core component, Precipitation-Adaptive MoE (PA-MoE), dynamically scales the number of activated experts per token according to local precipitation magnitude, channeling richer representational capacity toward the rare yet critical heavy-rainfall tail. A Dual-Axis Compressed Latent Attention mechanism factorizes spatiotemporal attention with convolutional reduction to manage massive context lengths, while an intensity-aware training protocol progressively amplifies learning signals from extreme-rainfall samples. Experiment on ERA5 demonstrate consistent improvements over state-of-the-art baselines, with particularly significant gains in heavy-rain and rainstorm regimes. |
| title | PA-Net: Precipitation-Adaptive Mixture-of-Experts for Long-Tail Rainfall Nowcasting |
| topic | Artificial Intelligence Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2603.13818 |