Enhancing LLM Reasoning for Time Series Classification by Tailored Thinking and Fused Decision

TL;DR

This paper introduces ReasonTSC, a novel framework that enhances large language model reasoning for time series classification by combining multi-turn reasoning with fused decision strategies, leading to improved accuracy and correction of errors.

Contribution

The paper presents a new approach, ReasonTSC, that effectively leverages LLM reasoning for TSC by integrating domain-specific models and structured multi-turn reasoning, which was underexplored before.

Findings

ReasonTSC outperforms existing baselines in TSC tasks.

It can identify and correct false predictions from plug-in models.

Extensive experiments validate its effectiveness and robustness.

Abstract

The reasoning capabilities of large language models (LLMs) have significantly advanced their performance by enabling in-depth understanding of diverse tasks. With growing interest in applying LLMs to the time series domain, this has proven nontrivial, as evidenced by the limited efficacy of straightforwardly adapting text-domain reasoning techniques. Although recent work has shown promise in several time series tasks, further leveraging advancements in LLM reasoning remains under-explored for time series classification (TSC) tasks, despite their prevalence and significance in many real-world applications. In this paper, we propose ReasonTSC, a novel framework designed to effectively leverage LLM reasoning for time series classification through both a multi-turn reasoning and a fused decision-making strategy tailored to TSC. Rather than straightforwardly applying existing reasoning techniques or relying solely on LLMs' built-in reasoning capabilities, ReasonTSC first steers the model to think over the essential characteristics of time series data. Next, it integrates predictions and confidence scores from plug-in classifiers, e.g., domain-specific time series models, as in-context examples. Finally, ReasonTSC guides the LLM through a structured reasoning process: it evaluates the initial assessment, backtracks to consider alternative hypotheses, and compares their merits before arriving at a final classification. Extensive experiments and systematic ablation studies demonstrate that ReasonTSC consistently outperforms both existing time series reasoning baselines and plug-in models, and is even capable of identifying and correcting plug-in models' false predictions.