English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Multivariate time-series forecasting is vital in various domains, e.g.,economic planning and weather prediction. Deep train-from-scratch models haveexhibited effective performance yet require large amounts of data, which limitsreal-world applicability. Recently, researchers have leveraged therepresentation learning transferability of pre-trained Large Language Models(LLMs) to handle limited non-linguistic datasets effectively. However,incorporating LLMs with time-series data presents challenges of limitedadaptation due to different compositions between time-series and linguisticdata, and the inability to process multi-scale temporal information. To tacklethese challenges, we propose LLM4TS, a framework for time-series forecastingwith pre-trained LLMs. LLM4TS consists of a two-stage fine-tuning strategy: the\textit{time-series alignment} stage to align LLMs with the nuances oftime-series data, and the \textit{forecasting fine-tuning} stage for downstreamtime-series forecasting tasks. Furthermore, our framework features a noveltwo-level aggregation method that integrates multi-scale temporal data withinpre-trained LLMs, enhancing their ability to interpret time-specificinformation. In experiments across 7 time-series forecasting datasets, LLM4TSis superior to existing state-of-the-art methods compared withtrained-from-scratch models in full-shot scenarios, and also achieves anaverage improvement of 6.84% in MSE in few-shot scenarios. In addition,evaluations compared with different self-supervised learning approacheshighlight LLM4TS's effectiveness with representation learning in forecastingtasks.

LLM4TS: Aligning Pre-Trained LLMs as Data-Efficient Time-Series Forecasters