Yuzhu Mao

Yuzhu Mao

AI Researcher & Engineer

About me

I am currently working as a researcher and senior engineer at Alipay.com Co., Ltd., focusing on multi-modal foundation models and their deployment on mobile devices. Prior to joining Alipay, I worked as a research assistant in the Smart Sensing and Robotics Group at the Tsinghua-Berkeley Shenzhen Institute (TBSI), where I explored large pre-trained models and their applications in resource-limited environments.

I hold a BEng degree from Wuhan University and an MS degree from Tsinghua University, where I was supervised by Prof. Wenbo Ding. During my student years, my research focused on artificial intelligence, with an emphasis on federated learning systems. I was also fortunate to receive invaluable feedback and guidance from Prof. Yang Liu (Tsinghua University), Prof. Linqi Song (City University of Hong Kong), and Prof. Tian Lan (George Washington University, USA).

News

  • Had one paper accepted by TMLR (12/2024).
  • Had one paper accepted by MobiCom 2024 Picasso Workshop (07/2024).
  • Had one paper accepted by ICLR 2024 Workshop on LLM Agents (04/2024).
  • Joined Tsinghua-Berkeley Shenzhen Institute as a research assistant (09/2023).
  • Had one another paper accepted by TMC (11/2023).
  • Had one paper accepted by TMC (06/2023).
  • Completed the Master’s defense (05/2023).
  • Submitted the Master’s thesis (04/2023).
  • Had one survey paper accepted by JFI (01/2023).
  • Joined Meituan UAV Lab as a research intern (07/2022).
  • Had one paper accepted by ACM TIST (01/2022).
  • Delivered a presentation at TBSI Workshop on Learning Theory (WOLT) and received the Best Poster Award (07/2021).
  • Joined Tencent Robotics-X Lab as a research intern (06/2021).
  • Enrolled at Tsinghua-Berkeley Shenzhen Institute as a master’s student in the data science and information technology track (09/2020).
  • Graduated with honors from Wuhan University (06/2020).

Publications

  • Mao, Y., Ping, S., Zhao, Z., Liu, Y., & Ding, W. (2024). Enhancing parameter efficiency and generalization in large-scale models: A regularized and masked low-rank adaptation approach. Transactions on Machine Learning Research (TMLR), Accepted [paper]
  • Ping, S.*, Mao, Y.*, Liu, Y., Zhang, X. P., & Ding, W. (2024). FL-TAC: Enhanced fine-tuning in federated learning via low-rank, task-specific adapter clustering. In ICLR 2024 Workshop on Large Language Model (LLM) Agents [paper]
  • Li, J., Zhao, C., Mao, Y., Chen, X., Ding, W., Qu, X., & Wang, J. (2024). FormerReckoning: Physics inspired transformer for accurate inertial navigation. In the 7th International Workshop on Physics Embedded AI Solutions in Mobile Computing (MobiCom Picasso Workshop 2024)
  • Mao, Y., Zhao, Z., Yang, M., Liang, L., Liu, Y., Ding, W., Lan, T., & Zhang, X. P. (2023). SAFARI: Sparsity-enabled federated learning with limited and unreliable communications. IEEE Transactions on Mobile Computing (TMC). DOI: 10.1109/TMC.2023.3296624 [paper]
  • Zhao, Z.*, Mao, Y.*, Shi, Z., Liu, Y., Lan, T., Ding, W., & Zhang, X. P. (2023). AQUILA: Communication efficient federated learning with adaptive quantization in device selection strategy. IEEE Transactions on Mobile Computing (TMC). DOI: 10.1109/TMC.2023.3332901 [paper]
  • Zhao, Z.*, Mao, Y.*, Liu, Y., Song, L., Ouyang, Y., Chen, X., & Ding, W. (2023). Towards efficient communications in federated learning: A contemporary survey. Journal of the Franklin Institute (JFI), 360(12), 8669-8703. DOI: 10.1016/j.jfranklin.2022.12.053 [paper]
  • Mao, Y., Zhao, Z., Yan, G., Liu, Y., Lan, T., Song, L., & Ding, W. (2022). Communication-efficient federated learning with adaptive quantization. ACM Transactions on Intelligent Systems and Technology (TIST), 13(4), 1-26. DOI: 10.1145/3510587 [paper][video]

(* indicates equal contribution)

Education Background

  • MS in Data Science and Information Technology, Tsinghua University, China (2020-2023)
  • BEng in Information Security, Wuhan University, China (2016-2020)

Work Experience

  • Researcher & Senior Engineer (07/2024 - Present), Alipay.com, Hangzhou, China
  • Research Intern (06/2022 - 09/2022), Meituan UAV Lab, Shenzhen, China
  • Research Intern (06/2021 - 09/2021), Tencent Robotics-X Lab, Shenzhen, China

Research Interests

  • Generalized and Interpretable Artificial Intelligence
  • Large Pre-Trained Models (Prompting, Fine-Tuning, RAG, Evaluation)
  • Machine Learning Systems

Reviewer Services

  • Conferences: UbiComp
  • Journals: TIST, TMC

Contact

  • Email: myz20@tsinghua.org.cn
  • Address: Information Building 1101A, Tsinghua Shenzhen International Graduate School, Shenzhen, China