Are Time-Series Foundation Models Ready for E-Nose Data? An Empirical Assessment of Their Embeddings

2026-06-25

AI Summary

This study presents the first systematic evaluation of time series foundation models (TSFMs) for electronic nose gas sensing data, focusing on gas identification and concentration prediction tasks. The authors propose a novel paradigm that involves fine-tuning representative TSFMs—such as Chronos-2 and MOMENT—and fusing their learned embeddings with features from task-specific models. Experimental results demonstrate that fine-tuning is essential for effective TSFM performance in this domain, and that the proposed embedding fusion strategy substantially improves prediction accuracy. These findings validate the potential of TSFMs in gas sensing applications while also highlighting their current limitations.

Abstract

Inspired by advances in natural language processing and computer vision, "time-series foundation models" (TSFMs) have recently been introduced with the promise of strong generalization across diverse time-series tasks, including forecasting, classification, and anomaly detection, as well as across domains such as healthcare, climate science, and manufacturing. However, their utility for gas-sensing data remains largely unexplored. To address this gap, this paper systematically evaluates recent TSFMs on electronic nose (E-Nose) data. In particular, we investigate whether embeddings produced by representative TSFMs, including Chronos-2 and MOMENT, provide effective representations for gas identification and concentration prediction. Specifically, we show that fine-tuning is necessary to achieve satisfactory performance on E-Nose data, and fusing TSFM embeddings with representations learned by specialized predictive models can further improve the performance, suggesting both the potential and limitations of current TSFMs for gas-sensing applications.

Problem

• time-series foundation models
• E-Nose
• gas sensing
• embeddings
• representation

Innovation

• time-series foundation models
• E-Nose
• embeddings
• fine-tuning
• representation fusion