DualSG: A Dual-Stream Explicit Semantic-Guided Multivariate Time Series Forecasting Framework

2025-07-29

AI Summary

Existing MTSF methods directly employ LLMs for end-to-end numerical forecasting, degrading accuracy and contradicting their intrinsic semantic modeling capability; alternatively, latent-space cross-modal alignment suffers from semantic-temporal misalignment. Method: We propose DualSG, a dual-stream framework that decouples semantic guidance from numerical prediction: an LLM serves as a lightweight semantic guidance module to generate trend-aware natural language captions, while a caption-guided fusion mechanism enables explicit, interpretable semantic-temporal alignment. A dedicated time-series encoder handles numerical modeling, relieving the LLM of extrinsic tasks. Contribution/Results: DualSG achieves significant improvements over 15 state-of-the-art baselines across multiple real-world datasets, demonstrating superior accuracy, interpretability, and generalizability—validating the efficacy of explicit semantic guidance in MTSF.

Technology Category

• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Time-Series/Data Streams
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Multivariate Time Series Forecasting plays a key role in many applications. Recent works have explored using Large Language Models for MTSF to take advantage of their reasoning abilities. However, many methods treat LLMs as end-to-end forecasters, which often leads to a loss of numerical precision and forces LLMs to handle patterns beyond their intended design. Alternatively, methods that attempt to align textual and time series modalities within latent space frequently encounter alignment difficulty. In this paper, we propose to treat LLMs not as standalone forecasters, but as semantic guidance modules within a dual-stream framework. We propose DualSG, a dual-stream framework that provides explicit semantic guidance, where LLMs act as Semantic Guides to refine rather than replace traditional predictions. As part of DualSG, we introduce Time Series Caption, an explicit prompt format that summarizes trend patterns in natural language and provides interpretable context for LLMs, rather than relying on implicit alignment between text and time series in the latent space. We also design a caption-guided fusion module that explicitly models inter-variable relationships while reducing noise and computation. Experiments on real-world datasets from diverse domains show that DualSG consistently outperforms 15 state-of-the-art baselines, demonstrating the value of explicitly combining numerical forecasting with semantic guidance.

Problem

• Improving multivariate time series forecasting precision with LLMs
• Addressing latent space alignment issues in multimodal methods
• Enhancing interpretability via explicit semantic guidance in forecasting

Innovation

• Dual-stream framework with explicit semantic guidance
• Time Series Caption for interpretable LLM prompts
• Caption-guided fusion reduces noise and computation