LLM-Mixer: Multiscale Mixing in LLMs for Time Series Forecasting

TL;DR

LLM-Mixer introduces a multiscale decomposition approach combined with pre-trained large language models to enhance time series forecasting accuracy across various datasets and horizons.

Contribution

The paper proposes a novel framework that integrates multiscale time-series decomposition with frozen LLMs, enabling improved forecasting of complex temporal patterns.

Findings

Outperforms recent state-of-the-art models in forecasting accuracy

Effective in capturing both short-term fluctuations and long-term trends

Demonstrates scalability across multivariate and univariate datasets

Abstract

Time series forecasting remains a challenging task, particularly in the context of complex multiscale temporal patterns. This study presents LLM-Mixer, a framework that improves forecasting accuracy through the combination of multiscale time-series decomposition with pre-trained LLMs (Large Language Models). LLM-Mixer captures both short-term fluctuations and long-term trends by decomposing the data into multiple temporal resolutions and processing them with a frozen LLM, guided by a textual prompt specifically designed for time-series data. Extensive experiments conducted on multivariate and univariate datasets demonstrate that LLM-Mixer achieves competitive performance, outperforming recent state-of-the-art models across various forecasting horizons. This work highlights the potential of combining multiscale analysis and LLMs for effective and scalable time-series forecasting.