prof_pic.jpeg

Jinmyeong Choi / 최진명

Email: jinmyeongchoi [at] snu [dot] ac [dot] kr / jinmyeoc [at] andrew [dot] cmu [dot] edu

Welcome! I’m Jinmyeong Choi, a master’s student at the Graduate School of Data Science at Seoul National University, advised by Prof. Taesup Kim.

I was also fortunate to spend time as a visiting scholar at Carnegie Mellon University, where I had the opportunity to work as a research intern at the Auton Lab with Prof. Artur Dubrawski.

My research lies at the intersection of time series foundation models, time series forecasting, and representation learning. I study how complex statistical properties of time series are encoded in learned representations, with a particular focus on non-stationarity and robust adaptation in real-world settings.

Outside of research, I enjoy swimming, running, and all kinds of sports, including futsal, soccer, basketball, badminton, and golf.

You can download my CV here.

news

Sep 30, 2026 Coming soon: our new preprint, “When Time Series Foundation Models Fail Under Cross-Channel Dependence.”
Sep 28, 2026 Our paper, “Interweaving Marginals into Multivariate Sample Paths: Training-Free Dependence Construction for Probabilistic Time Series Foundation Models,” was accepted to the FMTS Workshop at NeurIPS 2026🇦🇺.
Sep 24, 2026 Our Asilomar 2026 paper, “Towards Universal Representation-Based Process Control,” is now available on arXiv.
Sep 21, 2026 Our new paper, “Interweaving Marginals into Multivariate Sample Paths: Training-Free Dependence Construction for Probabilistic Time Series Foundation Models,” is now available on arXiv.
Jul 20, 2026 Our paper, “Towards Universal Representation-Based Process Control” was accepted to the Asilomar Conference on Signals, Systems, and Computers 2026!🇺🇸
Mar 01, 2026 Our paper, “Non-Stationarity in the Embedding Space of Time Series Foundation Models,” was accepted to the TSALM Workshop at ICLR 2026 🇧🇷

selected publications

  1. ICLR TSALM
    Non-Stationarity in the Embedding Space of Time Series Foundation Models
    Jinmyeong Choi, Brad Shook, and Artur Dubrawski
    In 1st ICLR Workshop on Time Series in the Age of Large Models, 2026
    arXiv
  2. ICLR
    Towards Robust Real-World Multivariate Time Series Forecasting: A Unified Framework for Dependency, Asynchrony, and Missingness
    Jinkwan Jang, Hyungjin Park, Jinmyeong Choi, and 1 more author
    In The Fourteenth International Conference on Learning Representations, 2026
    arXiv
  3. Asilomar
    Towards Universal Representation-Based Process Control
    Jinmyeong Choi, Taesup Kim, and Artur Dubrawski
    In 60th Asilomar Conference on Signals, Systems, and Computers, 2026
    arXiv
  4. NeurIPS FMTS
    Interweaving Marginals into Multivariate Sample Paths: Training-Free Dependence Construction for Probabilistic Time Series Foundation Models
    Jinmyeong Choi, Jinkwan Jang, Seul Lee, and 1 more author
    2026
    In NeurIPS 2026 Workshop on Foundation Models for Temporal Systems (FMTS)
    arXiv