Kanghui Ning

I’m a second-year Ph.D. student in the School of Computing at the University of Connecticut (UConn), under the supervision of Professor Dongjin Song. Currently, I am a research intern at Morgan Stanley Machine Learning Research in New York.

My primary research goal is to build reliable and interpretable AI systems. Currently, I focus on AI for time series analysis, with particular interests in:

Time Series Reasoning interpretable & trustworthy reasoning systems

Time Series Foundation Models forecasting, retrieval augmentation, model routing

Multi-modal Time Series LLM-empowered analysis & understanding

Previously, I obtained my bachelor’s degree from UESTC and my master’s degree from HUST.

If you share similar research interests or would like to collaborate, feel free to reach out at kanghui.ning@uconn.edu — I’m always open to discussions and potential collaborations.

news

Jun 15, 2026	📄 Our paper TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Models was accepted by the ICML 2026 Workshop on AI Forecasting!
Jun 01, 2026	💼 I started my research internship at Morgan Stanley Machine Learning Research in New York.
May 01, 2026	🎓 I was honored to receive the Predoctoral Fellowship (Platinum) from the School of Computing at UConn!
Oct 01, 2025	🏆 Our paper Towards Interpretable and Trustworthy Time Series Reasoning: A BlueSky Vision was accepted by the ICDM 2025 BlueSky track, and received the Best BlueSky Paper Award (3rd Place)!
Sep 18, 2025	📄 Our paper TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster was accepted by NeurIPS 2025!
Aug 01, 2025	🎓 I was honored to have been selected as a Fellow in the Entrepreneurship Fellowship Program at UConn, looking forward to the one-year journey!
May 15, 2025	🎓 I was honored to receive the Predoctoral Honorable Mention from the School of Computing at UConn!
May 01, 2025	📄 Our tutorial Multi-modal Time Series Analysis: A Tutorial and Survey was accepted by KDD 2025!

selected publications

ICML-WS

TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Models

Kanghui Ning, Yushan Jiang, Kashif Rasul, Anderson Schneider, Yuriy Nevmyvaka, and Dongjin Song

In ICML Workshop on AI Forecasting, 2026

arXiv Bib Code

@inproceedings{ning2026timerouter,
  title = {TimeRouter: Efficient and Adaptive Routing of Time-Series Foundation Models},
  author = {Ning, Kanghui and Jiang, Yushan and Rasul, Kashif and Schneider, Anderson and Nevmyvaka, Yuriy and Song, Dongjin},
  booktitle = {ICML Workshop on AI Forecasting},
  year = {2026},
}

ICDM

Towards Interpretable and Trustworthy Time Series Reasoning: A BlueSky Vision

Kanghui Ning, Zijie Pan, Yushan Jiang, Anderson Schneider, Yuriy Nevmyvaka, and Dongjin Song

In IEEE International Conference on Data Mining (ICDM), BlueSky Track, 2025

Best BlueSky Paper (3rd Place) arXiv Bib

Best BlueSky Paper Award (3rd Place) at ICDM 2025

@inproceedings{ning2025bluesky,
  title = {Towards Interpretable and Trustworthy Time Series Reasoning: A BlueSky Vision},
  author = {Ning, Kanghui and Pan, Zijie and Jiang, Yushan and Schneider, Anderson and Nevmyvaka, Yuriy and Song, Dongjin},
  booktitle = {IEEE International Conference on Data Mining (ICDM), BlueSky Track},
  year = {2025},
}

NeurIPS

TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster

Kanghui Ning, Zijie Pan, Yu Liu, Yushan Jiang, James Y. Zhang, Kashif Rasul, Anderson Schneider, Lintao Ma, Yuriy Nevmyvaka, and Dongjin Song

In Advances in Neural Information Processing Systems (NeurIPS), 2025

arXiv Bib Code

@inproceedings{ning2025tsrag,
  title = {TS-RAG: Retrieval-Augmented Generation based Time Series Foundation Models are Stronger Zero-Shot Forecaster},
  author = {Ning, Kanghui and Pan, Zijie and Liu, Yu and Jiang, Yushan and Zhang, James Y. and Rasul, Kashif and Schneider, Anderson and Ma, Lintao and Nevmyvaka, Yuriy and Song, Dongjin},
  booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
  year = {2025},
}

KDD

Multi-modal Time Series Analysis: A Tutorial and Survey

Yushan Jiang^*, Kanghui Ning^*, Zijie Pan^*, Xuyang Shen, Jingchao Ni, Wenchao Yu, Anderson Schneider, Haifeng Chen, Yuriy Nevmyvaka, and Dongjin Song

In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2025

arXiv Bib Code

@inproceedings{jiang2025multimodal,
  title = {Multi-modal Time Series Analysis: A Tutorial and Survey},
  author = {Jiang, Yushan and Ning, Kanghui and Pan, Zijie and Shen, Xuyang and Ni, Jingchao and Yu, Wenchao and Schneider, Anderson and Chen, Haifeng and Nevmyvaka, Yuriy and Song, Dongjin},
  booktitle = {ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD)},
  year = {2025},
}

experience

Morgan Stanley — Research Intern, Machine Learning Research
Jun 2026 – Present · New York, NY
University of Connecticut — Research Assistant, DSIS Lab
Aug 2024 – Present · Storrs, CT
Ant Group — Research Intern
May 2024 – Aug 2024 · Hangzhou, China
ByteDance — Research Intern, Corporate Strategy (Project Voyager), AI4Bioinformatics
Aug 2023 – Dec 2023 · Beijing, China
ByteDance — Research Intern, Applied Machine Learning, AI4Science Group
Jul 2022 – Aug 2023 · Beijing, China