---
title: "FreqLLM: Frequency-Aware Large Language Models for Time Series Forecasting"
authors:
  - "Shunnan Wang"
  - "Min Gao"
  - "Zongwei Wang"
  - "Yibing Bai"
  - "Feng Jiang"
  - "Guansong Pang"
doi: "10.24963/ijcai.2025/377"
url: "https://ijcai.org/proceedings/2025/377"
pdf: "https://ijcai.org/proceedings/2025/0377.pdf"
venue: "Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence"
year: 2025
pages: "3389-3397"
track: "Main Track"
keywords:
  - "Machine Learning: ML: Regression"
  - "Machine Learning: ML: Time series and data streams"
  - "Data Mining: DM: Mining spatial and/or temporal data"
---

# FreqLLM: Frequency-Aware Large Language Models for Time Series Forecasting

Shunnan Wang, Min Gao, Zongwei Wang, Yibing Bai, Feng Jiang, Guansong Pang

*Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence*, pages 3389-3397. DOI: [10.24963/ijcai.2025/377](https://doi.org/10.24963/ijcai.2025/377)

## Abstract

Large Language Models (LLMs) have recently shown promise in Time Series Forecasting (TSF) by effectively capturing intricate time-domain dependencies. However, our preliminary experiments reveal that standard LLM-based approaches often fail to capture global correlations, limiting predictive performance. We found that embedding frequency-domain signals smooths weight distributions and enhances structured correlations by clearly separating global trends (low-frequency components) from local variations (high-frequency components). Building on these insights, we propose FreqLLM, a novel framework that integrates frequency-domain semantic alignment into LLMs to refine prompts for improved time series analysis. By bridging the gap between frequency signals and textual embeddings, FreqLLM effectively captures multi-scale temporal patterns and provides more robust forecasting results. Extensive experiments on benchmark datasets demonstrate that FreqLLM outperforms state-of-the-art TSF methods in both accuracy and generalization. The code is available at https://github.com/biya0105/FreqLLM.