Extreme multi-label classification, with active contributions to LibMultiLabel, a library for multi-class and multi-label classification.
Efficient machine learning, focusing on accelerating large language model (LLM) inference and enabling sparse computation.
In addition, I help maintain LIBSVM and LIBLINEAR, well-established libraries for support vector machines and large-scale linear models.
Before joining MBZUAI, I spent three years working as a machine learning engineer at Taiwan Semiconductor Manufacturing Company (TSMC) and Yahoo, where I gained hands-on experience developing machine learning models and conducting data analysis for industrial applications.
I received my M.S. in Statistics from National Chengchi University, Taiwan, in 2020, where I was fortunate to be advised by Professor Chiu-Hsing Weng, and my B.S. in Mathematics from National Central University, Taiwan, in 2018.
Our work on hyperparameter search for tree based linear methods in extreme multi-label text classification has been accepted for presentation at the CIKM 2026 Applied Research Track!
May 29, 2026
Our work on probability estimation for extreme multi-label learning has been accepted for presentation at the ECML PKDD 2026 Research Track!
Oct 14, 2025
Our ICDM 2025 paper was selected as a Best Paper Award candidate! 🎉
Aug 26, 2025
My work on Revisiting One-Versus-One and One-Versus-Rest has been accepted as a regular paper at IEEE ICDM 2025.
@inproceedings{KTC25a,title={Revisiting one-versus-one and one-versus-rest: insights into imbalanced multi-class classification},author={Chen, Kuan-Ting and Lin, Chih-Jen},booktitle={Proceedings of the IEEE International Conference on Data Mining (ICDM)},pages={150-158},url={https://www.csie.ntu.edu.tw/~cjlin/papers/ovo-ovr/OVO-OVR.pdf},year={2025},}
Extreme Multi-Label
Probability Estimation in Tree-Based Linear Methods for Extreme Multi-Label Learning
Sheng-Wei Chen, Kuan-Ting Chen, Guan-Ting Chen, and 2 more authors
In Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML/PKDD), 2026
@inproceedings{SWC26b,author={Chen, Sheng-Wei and Chen, Kuan-Ting and Chen, Guan-Ting and Liu, Yaxu and Lin, Chih-Jen},title={Probability Estimation in Tree-Based Linear Methods for Extreme Multi-Label Learning},year={2026},booktitle={Proceedings of the Joint European Conference on Machine Learning and Knowledge Discovery in Databases (ECML/PKDD)},pages={},url={https://www.csie.ntu.edu.tw/~cjlin/papers/tree_prob_issue/tree_prob_issue.pdf},}
Extreme Multi-Label
On the Influence of Hyperparameters in Tree-Based Linear Methods for Extreme Multi-Label Text Classification: Insights for Efficient and Effective Search
Kuan-Ting Chen, Hung-Chih Chiang, and Chih-Jen Lin
In Proceedings of the 35th ACM International Conference on Information and Knowledge Management (CIKM), 2026
@inproceedings{KTC26a,author={Chen, Kuan-Ting and Chiang, Hung-Chih and Lin, Chih-Jen},title={On the Influence of Hyperparameters in Tree-Based Linear Methods for Extreme Multi-Label Text Classification: Insights for Efficient and Effective Search},booktitle={Proceedings of the 35th ACM International Conference on Information and Knowledge Management (CIKM)},pages={},url={https://www.csie.ntu.edu.tw/~cjlin/papers/tree_hyperparams_search/tree_hyperparams_search.pdf},year={2026},}
