CC-Time: Cross-Model and Cross-Modality Time Series Forecasting

TL;DR

CC-Time introduces a novel approach combining pre-trained language models with cross-modality and cross-model learning to significantly improve time series forecasting accuracy, especially in limited data scenarios.

Contribution

It proposes a new framework that integrates PLMs with time series models through cross-modality learning and a fusion block, enhancing forecasting performance.

Findings

Achieves state-of-the-art accuracy on nine datasets.

Effective in both full-data and few-shot learning.

Demonstrates the potential of PLMs in time series analysis.

Abstract

With the success of pre-trained language models (PLMs) in various application fields beyond natural language processing, language models have raised emerging attention in the field of time series forecasting (TSF) and have shown great prospects. However, current PLM-based TSF methods still fail to achieve satisfactory prediction accuracy matching the strong sequential modeling power of language models. To address this issue, we propose Cross-Model and Cross-Modality Learning with PLMs for time series forecasting (CC-Time). We explore the potential of PLMs for time series forecasting from two aspects: 1) what time series features could be modeled by PLMs, and 2) whether relying solely on PLMs is sufficient for building time series models. In the first aspect, CC-Time incorporates cross-modality learning to model temporal dependency and channel correlations in the language model from both time series sequences and their corresponding text descriptions. In the second aspect, CC-Time further proposes the cross-model fusion block to adaptively integrate knowledge from the PLMs and time series model to form a more comprehensive modeling of time series patterns. Extensive experiments on nine real-world datasets demonstrate that CC-Time achieves state-of-the-art prediction accuracy in both full-data training and few-shot learning situations.