Jinyi LI

Hi there! I’m Jinyi (Jerry) LI, welcome to my personal website. In here, you can discover something about me.

I am currently an undergraduate student in South China University of Technology. I have visited Lady Margaret’s Hall, University of Oxford during the summer 2023 and participated in the AI programme. With great pleasure, I appeared in the official video posted by LMH.

I worked as an intern with Prof. WANG at The Hong Kong University of Science and Technology (Guangzhou).

My current research projects include Meta Learning related works. My interests include Computer Vision, Multi-Modaling, Large Language Models and Meta Learning.

My latest paper is available on ArXiv. Check out PCToolkit: A Unified Plug-and-Play Prompt Compression Toolkit of Large Language Models.

Experience

University of Oxford

• 2023.7.17 - 2023.8.4

Artificial Intelligence and Machine Learning: Advanced Applications of Neural Networks and Deep Learning

Oxford Grade: 83 Letter Grade: A+

I learnt a lot about Artificial Intelligence and Machine Learning. Specifically, I learnt Generative Models including Auto Regression, Auto Encoder, Variation Auto Encoder, Generative Adversarial Network and Diffusion Model; Reinforcement Learning including DQN, Actor and Critic and DDPG methods; Graph Neural Network including GCN. My tutor Prof. Khan spoke very well of me.

Hong Kong University of Science and Technology (Guangzhou)

• 2024.2.1 - 2024.6.7

Internship in Artificial Intellengce Thrust, Information Hub

I worked with Prof. Hao WANG, doing researches on the topic of Large Language Models. We focused on Prompt Compression things and Multi-modal things.

Projects

• Student Research Program (SRP)

Surface Defect Detection of Industrial Products based on Computer Vision, supervised by Prof. YU. (Participating)

• Gesture Recognition Based on Cross-modal Supervision of Wi-Fi Beamforming Technique. (Team Leader, Participating)

• Large Language Model (LLM)-Readable Context Compression, supervised by Prof. WANG. (Participating)

• Completed an Autoencoder-like network for Hand-writing Formula Recognition. Designed a Motion Detector by Python using differential methods. Designed a Scheduler System by Qt Creator in C++ Course Design. Designed a Student Management System with the use of SQLite 3 in Database Course. Designed a Pseudo Eco-sphere in Data Structure Course Design. More projects are shown in project page.

Honors & Awards

• The Dongfeng Enterprise Scholarship, 2023

• The Second Prize Scholarship, 2022

• Merit Student, 2022&2023 (Twice)

• Excellent League Member, 2022&2023 (Twice)

• Excellent Student Cadre, 2023

• Mathematical Contest in Modeling and Interdisciplinary Contest in Modeling (MCM/ICM): Honorable Mention, 2023

• Asia and Pacific Mathematical Contest in Modeling (APMCM): Second Prize, 2023

Acknowledgement

Special thanks to Prof. YU who supported me when I firstly started my research.

Prof. XU is my supervisor for Bachelor’s degree and will be my supervisor during Ph.D.