Informed Forecasting: Leveraging Auxiliary Knowledge to Boost LLM Performance on Time Series Forecasting

TL;DR

This paper introduces a novel framework that enhances Large Language Models' ability to perform accurate time series forecasting by integrating structured temporal knowledge, demonstrating significant improvements over baseline methods in real-world datasets.

Contribution

The paper presents a new cross-domain knowledge transfer method that systematically infuses LLMs with temporal information to improve forecasting accuracy.

Findings

Knowledge-informed forecasting outperforms naive baselines.

Enhanced models show better generalization in real-world datasets.

Structured temporal information boosts LLM performance.

Abstract

With the widespread adoption of Large Language Models (LLMs), there is a growing need to establish best practices for leveraging their capabilities beyond traditional natural language tasks. In this paper, a novel cross-domain knowledge transfer framework is proposed to enhance the performance of LLMs in time series forecasting -- a task of increasing relevance in fields such as energy systems, finance, and healthcare. The approach systematically infuses LLMs with structured temporal information to improve their forecasting accuracy. This study evaluates the proposed method on a real-world time series dataset and compares it to a naive baseline where the LLM receives no auxiliary information. Results show that knowledge-informed forecasting significantly outperforms the uninformed baseline in terms of predictive accuracy and generalization. These findings highlight the potential of knowledge transfer strategies to bridge the gap between LLMs and domain-specific forecasting tasks.