Multiscale Time-Series Modeling

Overview

Multiscale time-series modeling studies how to represent and predict signals whose patterns unfold over different time spans, from brief events and local fluctuations to cycles and long-term trends. Common approaches use wavelet or frequency decomposition, hierarchical patches and pooling, multi-branch convolutional or Transformer and state-space encoders, and adaptive gates or expert routing to combine scales; some also handle irregular sampling and cross-variable or spatial structure. These methods support forecasting, classification, anomaly detection, imputation, and generation in settings such as sensors, health, climate, traffic, and finance, where missing a scale can obscure useful dynamics. A central challenge is choosing scales that adapt to changing patterns and sampling rates while preserving detail, generalizing across datasets and horizons, and keeping computation and uncertainty reliable.

Latest papers

CardsList

132 papers10 papers in the last four weeks, down 55% on the four weeks before. 0.1% of all new papers.

  1. How (Not) to Hybridize Neural and Mechanistic Models for Epidemiological Forecasting

    Feb 6, 2026Yiqi Su, Ray Lee, Jiaming Cui +1Epidemic ForecastingNon-Stationary Time Series Forecasting

  2. COMET: Codebook-based Online-adaptive Multi-scale Embedding for Time-series Anomaly Detection

    Feb 2, 2026Jinwoo Park, Hyeongwon Kang, Seung Hun Han +1Unsupervised Anomaly DetectionTime-Series Anomaly Detection

  3. Spectral Text Fusion: A Frequency-Aware Approach to Multimodal Time-Series Forecasting

    Feb 2, 2026Huu Hiep Nguyen, Minh Hoang Nguyen, Dung Nguyen +1Multivariate Time Series ForecastingFrequency-Domain Time Series Modeling

  4. Advanced Global Wildfire Activity Modeling with Hierarchical Graph ODE

    Jan 4, 2026Fan Xu, Wei Gong, Hao Wu +6Wildfire ForecastingSpatiotemporal GNNs

  5. HieraMix: A Hierarchical MLP-Mixer for Large-Scale Traffic Forecasting

    Nov 26, 2025Yongyao Wang, Xie Yu, Jingyuan Wang +2Intelligent Transportation SystemsSpatiotemporal Forecasting

  6. Kairos: Toward Adaptive and Parameter-Efficient Time Series Foundation Models

    Sep 30, 2025Kun Feng, Shaocheng Lan, Yuchen Fang +6Rotary Positional EmbeddingsZero-Shot Time Series Forecasting

  7. User Trajectory Prediction Unifying Global and Local Temporal Information

    Aug 4, 2025Wei Hao, Bin Chong, Ronghua Ji +1Human Trajectory PredictionTrajectory Prediction

  8. DMSC: Dynamic Multi-Scale Coordination Framework for Time Series Forecasting

    Aug 3, 2025Haonan Yang, Jianchao Tang, Zhuo Li +1Time Series ForecastingMulti-Scale Feature Fusion

  9. FISHER: A Foundation Model for Multi-Modal Industrial Signal Comprehensive Representation

    Jul 22, 2025Pingyi Fan, Anbai Jiang, Shuwei Zhang +9Representation LearningSelf-Supervised Time Series Representation Learning

  10. A State-of-the-Art Review of Computational Models for Analyzing Longitudinal Wearable Sensor Data in Healthcare

    Jul 31, 2024Paula LagoWearable SensingMultiscale Time-Series Modeling

  11. Fast, Interpretable, and Deterministic Time Series Classification With a Bag-of-Receptive-Fields

    Nov 29, 2023Francesco Spinnato, Riccardo Guidotti, Anna Monreale +1Time Series ClassificationInterpretable ML