EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Weformer: Adaptive spectral cross-variable transformer for long-horizon weather forecasting

Chunliang Wang, Xuzhang Shen, Jiuke Wang

Journal of King Saud University Computer and Information Sciences · 2026-09 · 38 (7)

DOI: 10.1007/s44443-026-01111-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of King Saud University Computer and Information Sciences, online erschienen am 18.08.2026. Verfasst von Chunliang Wang und 2 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Natural Science Foundation of China, National Key Research and Development Program of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Accurate long-term weather forecasting from multivariate sensor time series remains challenging due to the inherent non-stationarity of atmospheric observations acquired by heterogeneous environmental sensors and the complex physical coupling among meteorological variables. Existing general-purpose deep forecasting models, while successful on idealized benchmarks, suffer notable performance degradation on raw ground-station sensor data because they lack domain-specific mechanisms for cross-variable interaction and noise resistance. We propose Weformer, a Transformer-based architecture purpose-built for sensor-derived weather time series. Weformer introduces two core innovations: (1) a Frequency-driven Cross-Variable Rotary Position Embedding (CrossVarRoPE) that extracts dominant spectral patterns via the Fourier transform and injects adaptive, cross-variable-aware positional encoding into the attention mechanism, and (2) a Global Token mechanism that distills macroscopic environmental context through cross-attention, preventing overfitting to local high-frequency sensor noise. We provide rigorous theoretical analysis, including a noise-suppression bound, a capacity bound for the global token, and a forecasting error decomposition, to justify the design. To bridge the evaluation gap in current benchmarks, we curate seven new real-world datasets sourced from ground-station sensor networks and energy-grid monitoring systems, spanning diverse climate zones and temporal resolutions. Extensive experiments on eight datasets demonstrate that Weformer achieves the best overall performance among seven competitive baselines—the lowest global average MSE and MAE and the largest number of first-place results—under both standard and long-horizon settings (up to 2,880 steps, i.e., 20–120 days depending on the sampling resolution), while CrossVarRoPE serves as a portable plug-in that improves standard Transformer architectures in most evaluated settings. Code and data are available at https://github.com/sekiro1211/Weformer.

Förderung

  • National Natural Science Foundation of China
  • National Key Research and Development Program of China

Zitieren

APA

Wang, C., Shen, X., & Wang, J. (2026). Weformer: Adaptive spectral cross-variable transformer for long-horizon weather forecasting. Journal of King Saud University Computer and Information Sciences, 38(7). https://doi.org/10.1007/s44443-026-01111-9

Vancouver

Wang C, Shen X, Wang J. Weformer: Adaptive spectral cross-variable transformer for long-horizon weather forecasting. Journal of King Saud University Computer and Information Sciences. 2026;38(7). doi:10.1007/s44443-026-01111-9

BibTeX
@article{Wang2026,
  author = {Wang, Chunliang and Shen, Xuzhang and Wang, Jiuke},
  title = {Weformer: Adaptive spectral cross-variable transformer for long-horizon weather forecasting},
  journal = {Journal of King Saud University Computer and Information Sciences},
  year = {2026},
  volume = {38},
  number = {7},
  doi = {10.1007/s44443-026-01111-9}
}
Kostenlos starten →