LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters

TL;DR

This paper introduces LLM4TS, a novel framework that adapts pre-trained Large Language Models for multivariate time-series forecasting, effectively handling limited data scenarios through a specialized two-stage fine-tuning and multi-scale data integration.

Contribution

The paper presents a new two-stage fine-tuning strategy and a two-level aggregation method to adapt LLMs for time-series forecasting, addressing challenges of data heterogeneity and multi-scale temporal information.

Findings

Outperforms state-of-the-art models in full-shot scenarios.

Achieves top performance in few-shot scenarios.

Demonstrates effectiveness over unsupervised representation learning methods.

Abstract

Multivariate time-series forecasting is vital in various domains, e.g., economic planning and weather prediction. Deep train-from-scratch models have exhibited effective performance yet require large amounts of data, which limits real-world applicability. Recently, researchers have leveraged the representation learning transferability of pre-trained Large Language Models (LLMs) to handle limited non-linguistic datasets effectively. However, incorporating LLMs with time-series data presents challenges of limited adaptation due to different compositions between time-series and linguistic data, and the inability to process multi-scale temporal information. To tackle these challenges, we propose LLM4TS, a framework for time-series forecasting with pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the time-series alignment stage to align LLMs with the nuances of time-series data, and the forecasting fine-tuning stage for downstream time-series forecasting tasks. Furthermore, our framework features a novel two-level aggregation method that integrates multi-scale temporal data within pre-trained LLMs, enhancing their ability to interpret time-specific information. In experiments across 7 time-series forecasting datasets, LLM4TS is superior to existing state-of-the-art methods compared with trained-from-scratch models in full-shot scenarios, and also achieves the highest rank in few-shot scenarios. In addition, evaluations compared with different unsupervised representation learning approaches highlight LLM4TS's effectiveness with representation learning in forecasting tasks. Ablation studies further validate each component's contribution to LLM4TS and underscore the essential role of utilizing LLM's pre-trained weights for optimal performance. The code is available at https://github.com/blacksnail789521/LLM4TS.