MAP4TS: A Multi-Aspect Prompting Framework for Time-Series Forecasting with Large Language Models

TL;DR

MAP4TS introduces a multi-aspect prompting framework that integrates classical time-series analysis techniques with large language models to improve forecasting accuracy across diverse datasets.

Contribution

The paper presents a novel prompting framework that explicitly incorporates statistical and temporal features into LLM-based time-series forecasting, outperforming existing methods.

Findings

MAP4TS outperforms state-of-the-art LLM-based forecasting methods on eight datasets.

Prompt-aware design improves performance stability and accuracy.

GPT-2 with structured prompts surpasses larger models like LLaMA in long-term forecasting.

Abstract

Recent advances have investigated the use of pretrained large language models (LLMs) for time-series forecasting by aligning numerical inputs with LLM embedding spaces. However, existing multimodal approaches often overlook the distinct statistical properties and temporal dependencies that are fundamental to time-series data. To bridge this gap, we propose MAP4TS, a novel Multi-Aspect Prompting Framework that explicitly incorporates classical time-series analysis into the prompt design. Our framework introduces four specialized prompt components: a Global Domain Prompt that conveys dataset-level context, a Local Domain Prompt that encodes recent trends and series-specific behaviors, and a pair of Statistical and Temporal Prompts that embed handcrafted insights derived from autocorrelation (ACF), partial autocorrelation (PACF), and Fourier analysis. Multi-Aspect Prompts are combined with raw time-series embeddings and passed through a cross-modality alignment module to produce unified representations, which are then processed by an LLM and projected for final forecasting. Extensive experiments across eight diverse datasets show that MAP4TS consistently outperforms state-of-the-art LLM-based methods. Our ablation studies further reveal that prompt-aware designs significantly enhance performance stability and that GPT-2 backbones, when paired with structured prompts, outperform larger models like LLaMA in long-term forecasting tasks.