ZHIYU LIU

ZHIYU LIU

B.Eng in Shanghai Jiao Tong University
M.Eng in University of California, Los Angeles

Welcome to my personal website, where I share my passion and expertise in CS related fields.

💬 About Me

I am a first-year Master of Engineering student at UCLA Samueli School of Engineering and a graduate of Shanghai Jiao Tong University from the ParisTech Elite Institute of Technology. My research interests include computer vision, natural language processing, and large language models. I am currently actively seeking full-time job opportunities.

You can find my CV here: Zhiyu Liu’s Curriculum Vitae.

Email / Wechat

📝 Publications

CEIP 2023
sym

Crowd counting model based on CNN and Transformer

Zhiyu Liu, Yongqing Qu

  • Supervised by Prof. Mark Vogelsberger from MIT, a combination of Convolutional Neural Networks (CNNs) and Vision Transformer (ViT) was used to tackle the crowd counting issue.
  • CNN do well in local feature extraction, while ViT’s attention mechanism is ideal for global feature extraction. Our approach using integration presents a better performance than the original CNN model.
SRSE 2024
sym

Intelligent Fault Diagnosis of Rolling Bearing based on Incremental Learning

Zhiyu Liu, Zhiyi Zhang, Mohamed Sallak, Siqi Qiu*

  • My graduation thesis, supervised by Prof. Siqi Qiu, proposes a VMD-enhanced incremental learning model to improve fault diagnosis in rolling bearings.
  • By addressing catastrophic forgetting and enhancing diagnostic accuracy, it provides an adaptable solution for evolving fault patterns in industrial environments.

🎖 Honors and Awards

  • 2024.06       Outstanding graduate of Shanghai Jiao Tong University (Top 15%)
  • 2023.10       Third-year university student B scholarship of excellence (Top 10%)
  • 2022.11       Merit Student in sophomore year (Top 4%)
  • 2022.10       Second-year university student B scholarship of excellence (Top 15%)
  • 2022.08       The Best Computer Vision Project (1st out of 12 groups) of Imperial College Data Science Online Summer School
  • 2022.07       The First Prize (1st out of 10 groups) of National University of Singapore (NUS) School of Computing (SOC) Summer Workshop
  • 2021.10       First-year university student C scholarship of excellence (Top 30%)

📖 Educations

💻 Experiences

  • 2025.06 - 2025.08       Interned as AI Algorihtm Engineer at Ant Group, responsible for AISDR unified foundation model, building a full pipeline from rule-based data synthesis to SFT/DPO/GRPO training, achieving significant FinEval improvements and accelerating model iteration with business alignment.
  • 2024.03 - 2024.06       Interned as Large Language Model Engineer at an AI unicorn company MINIMAX, responsible for the pre-training and supervised fine-tuning code of the company’s self-developed model ABAB, as well as the construction of long text datasets. I also expanded the code evaluation benchmark to enhance the model’s performance in the code generation task.
  • 2023.03 - 2024.03       Topography Measurement by EBSD Calibration. Supervised by Prof. Qiwei Shi. Patent application approved
  • 2023.06 - 2023.09       Referring Image Segmentation Research Internship. Supervised by Prof. Miaojing Shi
  • 2023.01 - 2023.03       Crowd Counting Model Research. Supervised by Prof. Mark Vogelsberger. Paper accepted for publication
  • 2022.12 - 2023.09       PRP (Participation in Research Program) project of Shanghai Jiao Tong University, No. 41: High frequency grammar points analysis in engineering French based on Natural Language Processing
  • 2022.07 - 2022.08       Imperial College Data Science Online Summer School
  • 2022.05 - 2022.07       National University of Singapore (NUS) School of Computing (SOC) Summer Workshop
  • 2022.02 - 2022.02       The Mathematical Contest in Modeling (MCM) Project - Problem C - Trading Strategies