pith. sign in

arxiv: 2502.15637 · v2 · pith:NY2AY2XCnew · submitted 2025-02-21 · 💻 cs.LG · cs.AI· stat.ML

Mantis: Lightweight Foundation Model for Time Series Classification

Vasilii Feofanov , Songkang Wen , Shifeng Xie , Simon Roschmann , Marius Alonso , Hongbo Guo , Romain Ilbert , Malik Tiomoko
show 5 more authors
Quentin Bouniot Zeynep Akata Lujia Pan Jianfeng Zhang Ievgen Redko
This is my paper
classification 💻 cs.LG cs.AIstat.ML
keywords foundationmantisapplicationclassificationdemonstratedomainsexistingintroduce
0
0 comments X
read the original abstract

While foundation models have revolutionized various domains, their application to time series classification remains rather under-explored, with existing literature predominantly focused on forecasting. To bridge this gap, we introduce \textbf{Mantis}, a transformer-based foundation model pre-trained exclusively on synthetic data via self-supervised contrastive learning. We demonstrate that effective tokenization is critical to unlocking the full potential of transformers, proposing a novel token generator unit. Furthermore, we introduce an enhanced test-time methodology that bridges the performance gap between Mantis and strong specialized approaches by leveraging intermediate-layer representations, self-ensembling, and cross-model embedding fusion. Extensive experiments demonstrate that Mantis establishes a new state-of-the-art, outperforming existing foundation models across four diverse dataset collections covering various application domains.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Beyond IID: How General Are Tabular Foundation Models, Really?

    cs.LG 2026-06 unverdicted novelty 7.0

    Tabular foundation models excel on tiny- to medium-sized IID data but are outperformed by traditional tree-based and deep learning models on non-IID, large, and high-dimensional datasets, based on evaluations across 1...

  2. TelecomTS: A Multi-Modal Observability Dataset for Time Series and Language Analysis

    cs.AI 2025-10 conditional novelty 7.0

    TelecomTS is a new observability dataset from 5G networks that preserves absolute scale and supports multi-modal tasks, showing that current time series and language models struggle with abrupt noisy dynamics.

  3. Modular Retrieval-Augmented Generalization for Human Action Recognition

    eess.SP 2026-04 unverdicted novelty 6.0

    MoRA is a new retrieval-augmented module for IMU-based human activity recognition that uses uncertainty-adaptive fusion of retrieved motion patterns to improve model performance.

  4. COMODO: Cross-Modal Video-to-IMU Distillation for Efficient Egocentric Human Activity Recognition

    cs.CV 2025-03 unverdicted novelty 6.0

    COMODO is a cross-modal self-supervised distillation framework that uses a frozen video encoder and dynamic instance queue to align video and IMU embeddings, improving IMU-based egocentric HAR to match supervised performance.