Wenxuan Zeng

Wenxuan Zeng

Student at Peking University

Peking University

Biography

I am currently a first-year master student at Peking University (PKU) in Institute for Artificial Intelligence and School of Software and Microelectronics. I am now advised by Prof. Meng Li and Prof. Runsheng Wang, and devoted to efficient and secure deep learning system.

My main interested research topics include (but not limited to):

  • Efficient AI: model compression and acceleration
  • Secure AI: privacy-preserving deep learning
  • Large language model (LLM), CNN and Vision Transformers.

Download my resumé (not avaliable temporarily).

Interests
  • Efficient and secure deep learning
  • CV, NLP and graph learning
  • Music (fingerstyle guitar and singing)🎸
  • Badminton🏸 Fitness
Education
  • Master in School of Software and Microelectronics, 2023-Present

    Peking University

  • BSc in School of Software Engineering, 2019-2023

    University of Electronic Science and Technology of China

Research Theme

Machine Learning and Deep Learning

ML and DL build the AI world!

Model Compression and Acceleration

Make heavy models lighter and faster!

AI Security and Privacy

Preserve the pricacy of model and data!

Experience

 
 
 
 
 
Institute for Artificial Intelligence, Peking University
Current Scientific Research Lab
Sep 2022 – Present Beijing

Researches include:

  • Efficient and secure multi-modal deep learning, advised by Prof. Meng Li and Prof. Runsheng Wang
  • Devoted to building better AI systems with high performance, efficiency and security.
 
 
 
 
 
Knowledge Works Research Laboratory, Fudan University
Research Intern
Apr 2022 – Jul 2022 Remote

Researches include:

  • Study factuality and faithfulness in natural language generation (NLG), advised by Prof. Yanghua Xiao
  • One paper about factual error correction is accpeted by AAAI 2023
 
 
 
 
 
Sichuan Key Laboratory of Network and Data Security, UESTC
Research Intern
Apr 2021 – Apr 2022 Chengdu

Researches include:

  • Graph Neural Networks (GNN) and IP Geolocation, advised by Prof. Fan Zhou
  • One paper is accepted by KDD 2022

Accomplish­ments 荣誉奖项

Annual Symposium and Tech Day at PKU 北京大学人工智能研究院科技节
China Collegiate Computer Contest 中国高校计算机大赛
“Tencent” Scholarship Special Prize 腾讯特等奖学金
China Collegiate Computer Design Competition 中国大学生计算机设计大赛
China Collegiate Computer Contest 中国高校计算机大赛
“Shi Qiang” Scholarship First Prize 世强一等奖学金
“Zhong Gong Cup” Sichuan Collegiate Computer Competition 四川省大学生计算机作品赛
Pan-Pearl River Delta Collegiate Computer Competition 泛珠三角大学生计算机作品赛

Talks

Recent Publications

Quickly discover relevant content by filtering publications.
(2023). MPCViT: Searching for Accurate and Efficient MPC-Friendly Vision Transformer with Heterogeneous Attention. ICCV 2023.

PDF Cite DOI

(2023). CoPriv: Network/Protocol Co-Optimization for Communication-Efficient Private Inference. NeurIPS 2023.

PDF Cite

(2023). Converge to the truth: Factual error correction via iterative constrained editing. AAAI 2023.

PDF Cite Code DOI

(2022). Connecting the Hosts: Street-Level IP Geolocation with Graph Neural Networks. KDD 2022.

PDF Cite Video DOI

Contact