Feb, 2024
LSTPrompt: 大型语言模型作为零样本时间序列预测器之 短时 - 长时提示
LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting
Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. Aditya Prakash
TL;DRTime-series forecasting (TSF) finds broad applications in real-world scenarios, and this paper proposes LSTPrompt, a novel approach for prompting Large Language Models (LLMs) in zero-shot TSF tasks, which consistently outperforms existing prompting methods and achieves competitive results compared to foundation TSF models.