Decoding Time Series with LLMs: A Multi-Agent Framework for Cross-Domain Annotation

2024-10-22

arXiv.org

AI Summary

To address the scarcity of high-quality cross-domain annotations for time series, this paper proposes TESSA, a novel multi-agent framework featuring a pioneering collaborative architecture of general-purpose and domain-specific agents to jointly model cross-domain knowledge transfer and few-shot domain adaptation. TESSA integrates large language models, time-series feature extraction, cross-domain representation alignment, and few-shot prompt learning to enable joint decoding of temporal patterns and textual semantics. Experiments on multiple synthetic and real-world datasets demonstrate that TESSA achieves a 23.5% improvement in general annotation accuracy and attains a domain-specific annotation F1-score exceeding 0.89—substantially outperforming state-of-the-art methods. Its core contributions are: (1) a dual-agent collaborative generation paradigm; (2) a unified cross-domain–domain optimization mechanism; and (3) joint time-series–semantic decoding capability.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Multiagent Systems: Teamwork
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling

Abstract

Time series data is ubiquitous across various domains, including manufacturing, finance, and healthcare. High-quality annotations are essential for effectively understanding time series and facilitating downstream tasks; however, obtaining such annotations is challenging, particularly in mission-critical domains. In this paper, we propose TESSA, a multi-agent system designed to automatically generate both general and domain-specific annotations for time series data. TESSA introduces two agents: a general annotation agent and a domain-specific annotation agent. The general agent captures common patterns and knowledge across multiple source domains, leveraging both time-series-wise and text-wise features to generate general annotations. Meanwhile, the domain-specific agent utilizes limited annotations from the target domain to learn domain-specific terminology and generate targeted annotations. Extensive experiments on multiple synthetic and real-world datasets demonstrate that TESSA effectively generates high-quality annotations, outperforming existing methods.

Problem

• Automating high-quality time series annotations across domains
• Bridging general and domain-specific knowledge for annotation
• Improving annotation accuracy in mission-critical applications

Innovation

• Multi-agent system for time series annotation
• Combines general and domain-specific annotation agents
• Leverages cross-domain and target domain data