Dual-Forecaster: A Multimodal Time Series Model Integrating Descriptive and Predictive Texts

TL;DR

Dual-Forecaster is a novel multimodal time series model that integrates both historical descriptive texts and future predictive texts, significantly improving forecasting accuracy by leveraging advanced cross-modality techniques.

Contribution

It introduces a pioneering approach combining descriptive and predictive textual information with new cross-modality alignment methods for enhanced time series forecasting.

Findings

Outperforms or matches state-of-the-art models on 15 datasets

Demonstrates the effectiveness of integrating textual information

Highlights the importance of dual textual insights in forecasting

Abstract

Most existing single-modal time series models rely solely on numerical series, which suffer from the limitations imposed by insufficient information. Recent studies have revealed that multimodal models can address the core issue by integrating textual information. However, these models focus on either historical or future textual information, overlooking the unique contributions each plays in time series forecasting. Besides, these models fail to grasp the intricate relationships between textual and time series data, constrained by their moderate capacity for multimodal comprehension. To tackle these challenges, we propose Dual-Forecaster, a pioneering multimodal time series model that combines both descriptively historical textual information and predictive textual insights, leveraging advanced multimodal comprehension capability empowered by three well-designed cross-modality alignment techniques. Our comprehensive evaluations on fifteen multimodal time series datasets demonstrate that Dual-Forecaster is a distinctly effective multimodal time series model that outperforms or is comparable to other state-of-the-art models, highlighting the superiority of integrating textual information for time series forecasting. This work opens new avenues in the integration of textual information with numerical time series data for multimodal time series analysis.