Multimodal Conditioned Diffusive Time Series Forecasting
This work addresses the limitation of existing diffusion-based approaches that overlook multimodal information in time series data, offering a novel method for improved forecasting in various domains.
The paper tackles the problem of time series forecasting by proposing a multimodal conditioned diffusion model (MCD-TSF) that leverages timestamps and texts as extra guidance, achieving state-of-the-art performance on real-world benchmark datasets across eight domains.
Diffusion models achieve remarkable success in processing images and text, and have been extended to special domains such as time series forecasting (TSF). Existing diffusion-based approaches for TSF primarily focus on modeling single-modality numerical sequences, overlooking the rich multimodal information in time series data. To effectively leverage such information for prediction, we propose a multimodal conditioned diffusion model for TSF, namely, MCD-TSF, to jointly utilize timestamps and texts as extra guidance for time series modeling, especially for forecasting. Specifically, Timestamps are combined with time series to establish temporal and semantic correlations among different data points when aggregating information along the temporal dimension. Texts serve as supplementary descriptions of time series' history, and adaptively aligned with data points as well as dynamically controlled in a classifier-free manner. Extensive experiments on real-world benchmark datasets across eight domains demonstrate that the proposed MCD-TSF model achieves state-of-the-art performance.