Empowering VLMs for Few-Shot Multimodal Time Series Classification via Tailored Agentic Reasoning

TL;DR

This paper introduces MarsTSC, a novel multimodal time series classification framework using vision-language models with agentic reasoning, a dynamic knowledge bank, and test-time updates, achieving superior few-shot performance.

Contribution

The paper presents the first VL agentic reasoning framework for few-shot multimodal time series classification, incorporating a self-evolving knowledge bank and a test-time update strategy.

Findings

MarsTSC outperforms classical and foundation models on 12 benchmarks.

It improves few-shot classification accuracy across 6 VLM backbones.

The framework provides interpretable rationales grounded in human-readable features.

Abstract

In this paper, we propose the first VL$\underline{\textbf{M}}$ $\underline{\textbf{a}}$gentic $\underline{\textbf{r}}$easoning framework for few-$\underline{\textbf{s}}$hot multimodal $\underline{\textbf{T}}$ime $\underline{\textbf{S}}$eries $\underline{\textbf{C}}$lassification ($\textbf{MarsTSC}$), which introduces a self-evolving knowledge bank as a dynamic context iteratively refined via reflective agentic reasoning. The framework comprises three collaborative roles: i) Generator conducts reliable classification via reasoning; ii) Reflector diagnoses the root causes of reasoning errors to yield discriminative insights targeting the temporal features overlooked by Generator; iii) Modifier applies verified updates to the knowledge bank to prevent context collapse. We further introduce a test-time update strategy to enable cautious, continuous knowledge bank refinement to mitigate few-shot bias and distribution shift. Extensive experiments across 12 mainstream time series benchmarks demonstrate that $\textbf{MarsTSC}$ delivers substantial and consistent performance gains across 6 VLM backbones, outperforming both classical and foundation model-based time series baselines under few-shot conditions, while producing interpretable rationales that ground each classification decision in human-readable feature evidence.