<ul><li>Researchers propose TimeCMA, an intuitive framework for Multivariate Time Series Forecasting (MTSF) via cross-modality alignment.</li><li>TimeCMA combines large language models (LLMs) with time series data to achieve improved forecasting performance.</li><li>The framework uses a dual-modality encoding approach to obtain disentangled time series embeddings and robust prompt embeddings.</li><li>TimeCMA outperforms existing methods in MTSF according to extensive experiments on eight real datasets.</li></ul>

TimeCMA: Towards LLM-Empowered Multivariate Time Series Forecasting via Cross-Modality Alignment

Discover more