About Me

I am Yuchang Sun. I am currently working in DAIL, Alibaba as Senior Algorithm Engineer.

I received the Ph.D. degree from The Hong Kong University of Science and Technology (HKUST) in 2024, supervised by Prof. Jun Zhang, and the B.Eng. degree from Beijing Institute of Technology (BIT) in 2020. During my Ph.D., I was fortunate to visit EURECOM, advised by Prof. Marios Kountouris and to intern at Alibaba, mentored by Yuexiang Xie and led by Dr. Yaliang Li.

If you are interested in my research, do not hesitate to contact me via hiyuchang@outlook.com or yuchang_0816 (Wechat).

Research Interests

  • Test-time computation of large language models (LLMs)
  • Active learning (AL)
  • Federated learning (FL) and distributed learning

News

  • 10/12/2024: Our paper on active data querying in FL was accepted to AAAI 2025! Thanks to Xinran, Dr. Tao Lin, and Prof. Zhang! The paper and code are public now.

  • 26/11/2024: We release the source code for our paper on FL with selective layer fine-tuning! See the repo.

  • 23/11/2024: I attended the Graduation Ceremony of HKUST and received the Ph.D. degree. Thanks to my family, friends, and supervisors!

  • 22/11/2024: Our paper on collaboration in cross-silo FL was accepted to TMC! Thanks to my co-authors!

  • 29/08/2024: Our paper on FL with selective layer fine-tuning is public at ArXiv. Welcome to discuss with me!

  • 08/08/2024: I joined DAIL, Alibaba as a full-time senior algorithm engineer.

  • 25/06/2024: I received HKUST RedBird Academic Excellence Award for Continuing PhD Students in 2023-24.

  • 02/05/2024: I passed my Ph.D. defense exam! I am a “Doctor” now!

  • 01/04/2024: We release the codes for MimiC! See the repo! Welcome to cite the paper if you find it useful!

  • 01/02/2024: Our paper on FL is accepted by IEEE TMC! See Feature Matching Data Synthesis for Non-IID Federated Learning. Congrats to Zijian!

  • 25/01/2024: One new paper on collaboration in FL is public at ArXiv.

  • News before 2024 can be found here.

Selected Publications

The following selected publications are some of my recent works. For a full list of publications, please refer to this page or my Google Scholar page.

Learn How to Query from Unlabeled Data Streams in Federated Learning (Paper)

  • Authors: Y. Sun, X. Li, T. Lin, J. Zhang
  • Accepted to AAAI 2025 (CCF A).
  • Code is available at here.

MimiC: Combating Client Dropouts in Federated Learning by Mimicking Central Updates (Paper)

  • Authors: Y. Sun, Y. Mao, J. Zhang
  • Accepted to IEEE Transactions on Mobile Computing (CCF A).
  • This is one of my personally favourite ideas! A chinese introduction is online here.
  • Code is available at here.

Semi-Decentralized Federated Edge Learning with Data and Device Heterogeneity (Paper)

  • Authors: Y. Sun, J. Shao, Y. Mao, J.H. Wang, J. Zhang
  • Accepted to IEEE Transactions on Network and Service Management, vol. 20, no. 2, Jun. 2023.

Selected Awards and Honors

  • HKUST RedBird Academic Excellence Award for Continuing PhD Students, AY 2023/24

  • Outstanding Graduates in Beijing (Top 4%), Jul. 2020

  • National Scholarship (Top 2%), AY 2018/19

  • DWIN Scholarship (Top 2%), AY 2017/18