Search papers, labs, and topics across Lattice.
This paper introduces TimesX, a novel benchmark for multimodal time series forecasting that addresses critical limitations in existing benchmarks, including generalization issues and data leakage. By leveraging a diverse set of real-world time series and accompanying textual contexts generated through an automated pipeline, the study reveals that many previously successful forecasting methods struggle with TimesX. Notably, simple ensemble techniques that utilize the rich contextual information significantly outperform established baselines, highlighting the importance of context in forecasting accuracy.
Simple ensemble methods leveraging rich textual context can outperform state-of-the-art multimodal forecasting approaches on a new benchmark, TimesX, revealing hidden vulnerabilities in existing evaluations.
We introduce a new context-enriched, multimodal time series forecasting benchmark, TimesX. TimesX contains a wide selection of high-quality real-world time series with diverse domains and textual contexts obtained from an automated data generation pipeline, which helps address three main issues of existing multimodal forecasting benchmarks: (1) poor generalization due to the small scale and synthetic nature of benchmark data, (2) very limited types of textual contexts in the benchmarks, and (3) an inability to mitigate data leakage in evaluation. We conduct a thorough empirical study of zero-shot multimodal forecasting approaches on TimesX. Our results suggest that many approaches that perform well on existing benchmarks may fail on TimesX. In contrast, simple ensemble methods that leverage rich textual context accompanying time-series can outperform strong baselines on TimesX.
Google Research