About Me
I am currently a Research Fellow (Postdoc) at Nanyang Technological University (NTU), working under the supervision of Professor Jie Zhang. Previously, I was a Ph.D. candidate jointly trained at South China University of Technology (SCUT) and Institute for Infocomm Research (I2R), A*STAR. My doctoral study was supervised by Professor Qianli Ma (2021–2025) at SCUT and Principal Scientist Min Wu (2024–2025) at I2R, A*STAR. My research primarily focuses on deep learning for time series analysis, with current interests in time series classification and time series foundation models. I am particularly interested in developing unified, scalable representation-learning frameworks that improve generalization and robustness in time-series classification.
Recent News
- 01/2026: One paper titled “Learning Recursive Multi-Scale Representations for Irregular Multivariate Time Series Forecasting” was accepted by ICLR 2026.
- 12/2025: One paper titled “CompleMatch: Boosting Time-Series Semi-Supervised Classification with Temporal-Frequency Complementarity” was accepted by IEEE Transactions on Pattern Analysis and Machine Intelligence.
- 11/2025: One paper titled “A Unified Shape-Aware Foundation Model for Time Series Classification” was accepted by AAAI 2026.
- 05/2025: Three papers were accepted by ICML 2025.
- 10/2024: One paper titled “A survey on time-series pre-trained models” was accepted by IEEE Transactions on Knowledge and Data Engineering.
- 09/2024: One paper titled “Knowledge-Empowered Dynamic Graph Network for Irregularly Sampled Medical Time Series” was accepted by NeurIPS 2024.
- 05/2024: One paper titled “Incremental Sequence Labeling: A Tale of Two Shifts” was accepted by ACL 2024 (Findings).
- 12/2023: One paper titled “Diffusion Language-Shapelets for Semi-supervised Time-Series Classification” was accepted by AAAI 2024.
- 09/2023：One paper titled “Scale-teaching: Robust Multi-scale Training for Time Series Classification with Noisy Labels” was accepted by NeurIPS 2023.
- 04/2023：One paper titled “CTW: Confident Time-Warping for Time-Series Label-Noise Learning” was accepted by IJCAI 2023.
- 03/2023：One paper titled “Category-aware optimal transport for incomplete data classification” was accepted by Information Sciences 2023.
- 12/2022：One paper titled “Temporal-frequency co-training for time series semi-supervised learning” was accepted by AAAI 2023.
- 09/2019：One paper titled “Mechanical fault diagnosis and prediction in IoT based on multi-source sensing data fusion” was accepted by Simulation Modelling Practice and Theory 2020.