FIRFormer : Transformer With Frequency‐Aware Fusion and Recalibrated Attention for Time Series Forecasting
Expert Systems · 2026-09 · 43 (9)
DOI: 10.1111/exsy.70378 ↗Kurz gefasst
Fachartikel in der Zeitschrift Expert Systems, online erschienen am 26.07.2026. Verfasst von Shijun Chen und 2 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Youth Innovation Technology Project of Higher School in Shandong Province.
Abstract (Originalsprache)
The effectiveness of Transformers in time series forecasting largely depends on their ability to extract and utilise key features. The model not only needs to capture temporal patterns but also focus its attention on the most informative inputs, thereby constructing sharp and discriminative attention distributions to highlight critical information and improve forecasting accuracy. However, despite their strengths in modelling global dependencies, conventional Transformers remain limited in capturing predictive key patterns and forming highly distinctive attention weights. To address this limitation, we propose FIRFormer, a framework that enhances input representation and reconstructs the attention mechanism to better capture important temporal dynamics. Specifically, we introduce the Dynamic Frequency Selector module to adaptively filter frequency components most relevant to the forecasting objective, reducing redundancy and improving input efficiency. Subsequently, we introduce the Biased Integration Window module, which integrates a temporal decay mechanism into sliding aggregation to strengthen the correlations within local contexts, achieves precise fusion of temporal dynamic patterns and further emphasises the modelling value of recent critical information. Finally, we propose the Attention Recalibration module, which employs a dual‐branch attention refinement strategy. It enhances attention selectivity from both the local feature perspective and the global subspace perspective, promoting sharper and more focused attention distributions. FIRFormer enhances the perception of critical dynamic patterns and achieves performance improvements across multiple real‐world forecasting tasks.
Förderung
- National Natural Science Foundation of China
- Youth Innovation Technology Project of Higher School in Shandong Province
Zitieren
APA
Chen, S., Wang, H., & Zhang, F. (2026). FIRFormer : Transformer With Frequency‐Aware Fusion and Recalibrated Attention for Time Series Forecasting. Expert Systems, 43(9). https://doi.org/10.1111/exsy.70378
Vancouver
Chen S, Wang H, Zhang F. FIRFormer : Transformer With Frequency‐Aware Fusion and Recalibrated Attention for Time Series Forecasting. Expert Systems. 2026;43(9). doi:10.1111/exsy.70378
BibTeX
@article{Chen2026,
author = {Chen, Shijun and Wang, Hua and Zhang, Fan},
title = {FIRFormer : Transformer With Frequency‐Aware Fusion and Recalibrated Attention for Time Series Forecasting},
journal = {Expert Systems},
year = {2026},
volume = {43},
number = {9},
doi = {10.1111/exsy.70378}
}
EUVIMED