Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
Feijie (Harli) Wu's Homepage
[go: Go Back, main page]

About

Short Bio

I am currently a Ph.D. student advised by Prof. Jing Gao in the Elmore Family School of Electrical and Computer Engineering at Purdue University. Before joining Purdue in 2022, I worked with Prof. Song Guo at the Pervasive Edge Intelligence Laboratory (PEILab). I obtained my B.Sc. degree in Computing (major) and Applied Mathematics (minor) from The Hong Kong Polytechnic University in 2020. During my doctoral studies, I have conducted research in collaboration with leading technology companies, including Alibaba and Apple. I have served as a reviewer for a number of prestigious conferences, such as ICML, NeurIPS, ICLR, and KDD. I am a recipient of the Bilsland Dissertation Fellowship and the Best Student Paper Award from ACM BSCI'19.

Research Interest

Large Language Models

Federated Learning

Large-scale Optimization

AI Agents

Multi-agent System

News

  • [Dec. 2025] I receive the Bilsland Dissertation Fellowship from Purdue University.
  • [Dec. 2025] Our paper on privacy-preserving split federated learning is accepted by KDD'26.
  • [Oct. 2025] A new paper on LLM training for tool use comes out. Check out the paper on arXiv.
  • [Oct. 2025] I have created an account on X. Feel free to follow me.
  • [Jul. 2025] Our paper on LLM copyright protection with regularized selective unlearning is accepted by COLM'25.
  • [May 2025] I join Apple AIML as an intern this summer.
  • [Jan. 2025] Our paper on federated RLHF is accepted by ICLR'25.
  • [Oct. 2024] I receive a travel grant from SenSys'24.
  • [Sept. 2024] Our paper on model-heterogeneous federated learning is accepted by NeurIPS'24.
  • [Sept. 2024] Our paper on LLM copyright protection is accepted by EMNLP'24.
  • [Sept. 2024] Our paper on heterogeneous multi-modal federated learning is accepted by SenSys'24.

* indicates equal contribution.

* indicates equal contribution.