Jiawen Zhang

Researcher @ Microsoft Research Asia

To build AI that holds up when the world changes.

About

I am a researcher at Microsoft Research Asia. My research focuses on foundation models and time series modeling, with an emphasis on cross-domain generalization, irregularly sampled and noisy data, and dependable deployment in real-world applications.

Before joining MSRA, I completed my PhD in Data Science and Analytics at The Hong Kong University of Science and Technology, co-advised by Prof. Jia Li and Prof. Xiaofang Zhou.

Research collaborations and discussions are always welcome. I am also hiring research interns. Please send your CV to jiawenzhang@microsoft.com if you are interested.

What I Work On

  • Foundation models for time series. Cross-domain pre-training, transfer, and evaluation for temporal data.
  • Time-series reasoning. Adaptive reasoning over temporal signals for interpretable, decision-ready predictions.
  • AI for real systems. Reliable modeling of messy, real-world data across healthcare and scientific applications.

Selected Projects

ARTIST Adaptive time-series reasoning via segment selection. ProbTS A benchmark for point and distributional forecasting across diverse horizons. ElasTST Elastic time-series transformer for robust varied-horizon forecasting.

Publications

See Google Scholar for the full publication list. (* denotes equal contribution.)

  • S. Messica, J. Zhang*, K. Li*, T. Tsiligkaridis, M. Zitnik. (2026). Adaptive Time Series Reasoning via Segment Selection. ICML 2026. [paper][code]
  • X. Hong, J. Zhang, W. Li, S. Lu, J. Li. (2025). Unify and Anchor: A Context-Aware Transformer for Cross-Domain Time Series Forecasting. arXiv. [paper]
  • Z. Zhang, J. Zhang, S. Zheng, Y. Gu, J. Bian. (2025). Does Cross-Domain Pre-Training Truly Help Time-Series Foundation Models? ICLR Workshop on Foundation Models in the Wild. [paper]
  • J. Zhang, S. Zheng, X. Wen, X. Zhou, J. Bian and J. Li. (2024). ElasTST: Towards Robust Varied-Horizon Forecasting with Elastic Time-Series Transformer. NeurIPS 2024. [paper][code]
  • J. Zhang, X. Wen, Z. Zhang, S. Zheng, J. Li and J. Bian. (2024). ProbTS: Benchmarking Point and Distributional Forecasting across Diverse Prediction Horizons. NeurIPS 2024. [paper][code]
  • J. Zhang, S. Zheng, W. Cao, J. Bian and J. Li. (2023). Warpformer: A Multi-scale Approach for Irregularly Sampled Multivariate Time Series. KDD 2023. [paper][code]
  • J. Zhang, J. Zhu, Y. Yang, W. Shi, C. Zhang and H. Wang. (2021). Knowledge-Enhanced Domain Adaptation in Few-Shot Relation Classification. KDD 2021. [paper][code]
© Jiawen Zhang Powered by Hexo