LLM-Prompt: Integrated Heterogeneous Prompts for Unlocking LLMs in Time Series Forecasting

2025-06-21

AI Summary

Existing LLM-based time series forecasting approaches face two key bottlenecks: the absence of a unified prompting paradigm and the neglect of modality discrepancies between textual and temporal data. To address these, we propose LLM-Prompt—a novel heterogeneous unified prompting framework specifically designed for time series forecasting. It synergistically integrates learnable soft prompts with structured hard prompts and introduces a cross-modal semantic alignment mechanism to enable deep fusion of time-series signals and textual representations within a shared semantic space. Our method comprises three core components: time-series projection, multi-granularity textual encoding, and alignment-aware optimization. Extensive experiments across six general-purpose and three carbon-emission forecasting benchmarks demonstrate significant improvements in long-horizon prediction accuracy and few-shot generalization capability. LLM-Prompt establishes an interpretable, scalable paradigm for empowering LLMs in time series modeling.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Prompt Engineering / Prompting
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Time series forecasting aims to model temporal dependencies among variables for future state inference, holding significant importance and widespread applications in real-world scenarios. Although deep learning-based methods have achieved remarkable progress, they still exhibit suboptimal performance in long-term forecasting and data-scarce scenarios. Recent research demonstrates that large language models (LLMs) achieve promising performance in time series forecasting. However, we find existing LLM-based methods still have shortcomings: (1) the absence of a unified paradigm for textual prompt formulation and (2) the neglect of modality discrepancies between textual prompts and time series. To address this, we propose LLM-Prompt, an LLM-based time series forecasting framework integrating multi-prompt information and cross-modal semantic alignment. Specifically, we first construct a unified textual prompt paradigm containing learnable soft prompts and textualized hard prompts. Second, to enhance LLMs' comprehensive understanding of the forecasting task, we design a semantic space embedding and cross-modal alignment module to achieve cross-modal fusion of temporal and textual information. Finally, the transformed time series from the LLMs are projected to obtain the forecasts. Comprehensive evaluations on 6 public datasets and 3 carbon emission datasets demonstrate that LLM-Prompt is a powerful framework for time series forecasting.

Problem

• Lack of unified textual prompt paradigm for LLMs in forecasting
• Modality gap between textual prompts and time series data
• Suboptimal LLM performance in long-term and data-scarce scenarios

Innovation

• Unified textual prompt paradigm with learnable soft prompts
• Semantic space embedding for cross-modal alignment
• Cross-modal fusion of temporal and textual information