About me

Hi there 👋 I’m Jingmin Wei. I received my M.S. degree in Computer Science at University of Southern California (USC) in 2024. I am very fortunate to work at IRIS Computer Vision Lab under the guidance of Prof. Ram Nevatia, collaborating closely with Zhaoheng Zheng. My research interest lies in the area of robustness and efficiency, including learning robust visual representation, focusing on strengthening models’ resilience against adversarial attacks, and parameter-efficient fine-tuning, aiming to achieve greater efficacy in multi-modal optimization.

Prior to USC, I completed my bachelor degree at the School of Artificial Intelligence and Automation, Huazhong Univ. of Science and Technology (a Top 10 university in China). Advised by Prof. Sheng Zhong, I focused on research about building emotion assess system based on facial language analysis and understanding.

Research Interests

• Computer Vision
• Multi-modal Machine Learning
• Robustness and Trustworthy ML
• Parameter-efficient Fine-tuning
• Multi-object Detection and Tracking

Publications

Large Language Models are Good Prompt Learners for Low-Shot Image Classification
Zhaoheng Zheng, Jingmin Wei, Xuefeng Hu, Haidong Zhu, and Ram Nevatia
CVPR 2024
Paper Code

GDN: A Stacking Network Used for Skin Cancer Diagnosis
Jingmin Wei, Haoyang Shen, Ziyi Wang, Ziqian Zhang
International Conference on Signal Processing Systems, 2022
Paper

Projects

Learning Robust visual Represtation (LR²)
Advisor: Prof. Ram Nevatia, Zhaoheng Zheng
Research Program

Language-Driven 3D Stylization
Advisor: Prof. Yue Wang
Course Project, Score A
Code

Multi-modal Masked Adapter
Advisor: Prof. Jesse Thomason
Course Project, Score A
Code

Emotion Assessment System Based on Facial Language Understanding
Advisor: Prof. Sheng Zhong
Excellent College Students Research Program

Education

Awards

• Outstanding Graduate, 2022
• Excellent College Students Research Program, 2022
• Scholarship for Academic Excellence (Top 15%), 2021
• Scholarship for Scientific and Technological Innovation, 2021
• Social Scholarship for "Shenzhen Stock Exchange" (The Only One in School), 2021
• National First Prize, Big Data Competition (Top 0.3%), 2021

Technical Skills