CV
Education
B.S., University of Electronic Science and Technology of China (UESTC), Expected 2025
- GPA: 3.95/4.00, Average Score: 91.21/100, Ranking: Top 8.7%, CET6: 582
- Leadership Roles: Chairman and Machine Learning Leader of School Technology Studio, Officer in the Student Union Federation
Experience
Research Assistant in State Key Laboratory of Internet of Things for Smart City, University of Macau Supervisor: Assistant Professor Zhenning Li Research Focus: Efficient perception and decision-making in autonomous driving, human-like understanding and modeling in autonomous driving.
- Research 1: Enhanced trajectory prediction through semantic scene annotation via LLM output and Chain-of-Thoughts Prompting. Proposed two new datasets for fine-tuning lightweight language models to address reasoning costs. This research has been submitted to the ECAI 2024 Conference on Artificial Intelligence - Co-first Author.
- Research 2: Developed a safety-aware trajectory prediction model integrating Responsibility-Sensitive Safety (RSS) metrics and introduced an Uncertainty-aware Graph Attention Network. This research has been submitted to IEEE T-RO, a top-tier robotics journal (editorial review passed) - Co-first Author.
- Research 3: Integrated cognitive insights into the trajectory prediction task, with a focus on perceptual safety and dynamic decision-making during driving, and introduced the MoCAD trajectory prediction dataset. This research has been accepted by the top-tier Artificial Intelligence conference IJCAI 2024.
- Research 4: Formulated a behavior-aware model for trajectory prediction, incorporating theories from human behavior and decision-making domains. This study has been accepted at the top-tier transportation conference TRB 2024 (Oral Presentation) - Co-first Author.
Main Projects
Web Security Detection System Based on Large Language Models and Graph Neural Networks
- Developed a web security detection system using LLMs and GNNs to achieve geographical IP location, security analysis, and visualization of web page sources.
- Integrated a street-level IP geolocation framework based on graph neural networks with the Amap API for IP localization and visualization, deployed fine-tuned Chat-GLM for security analysis.
- Authored engineering documentation and conducted comprehensive testing of the project, which is set to collaborate with enterprises to extend the system into a deployable product.
Integrated Course Design: Adolescent Mental Health Rescue SOS System Based on Affective Computing
- Developed a mental health detection and analysis system for adolescents, providing early warnings and assistance.
- Fine-tuned the BLIP2 model for multimodal sentiment analysis and used the Flask framework for backend development. This project was selected as an outstanding integrated course design project by the college.
“Dydai” Mutual Aid—Solving the “Last kilometer” Problem in Parcel Delivery
- Developed a WeChat mini-program platform based on mutual aid among residents for parcel collection to address the “last kilometer” challenge.
- Authored PRD documents and designed product prototypes. This project is open-sourced on GitHub and has won multiple awards in college computer design competitions.
Web Platform Data Intelligent Analysis System Based on NLP and Web Crawling
- Created a data visualization platform using NLP and web crawling techniques to analyze comment sections of mainstream internet platforms (Bilibili, Zhihu, Douban, etc.).
- Fine-tuned BERT for sentiment analysis and utilized Echarts and WordCloud for visualization. This project was successfully completed under the “Ginkgo Fruit Fund” at the University of Electronic Science and Technology.
Awards
- Chinese Colleges Computer Competition - Network Technology Challenge: National Grand Prize (First Place Nationally in Undergraduate Group)
- University of Electronic Science and Technology: Excellent Student Scholarship *2
- Blue Bridge Cup Software Design Python Group: Provincial Second Prize
- National Computer Design Competition: Provincial Third Prize
- National E-commerce “Innovation and Entrepreneurship” Competition: Provincial Third Prize
- University of Electronic Science and Technology “Ginkgo Fruit Fund”: One project successfully completed
- School of Information and Software Engineering, University of Electronic Science and Technology: Award for Outstanding Studio
Skills
- Language: Python (Pytorch for Neural Networks, Flask for Web), C/C++, Java, Golang, SQL
- Platform: Linux, Windows, Mac, Tencent Cloud
- Techniques: Deep Learning, Data Analysis, Academic Drawing, Product Design, Algorithm Design, System Architecture Design, Software Testing
- Abilities: Public Speaking and Presentation, Team Coordination and Organization, Ability to solve complex engineering problems, Quick learner of new knowledge, Proficient in productivity tools such as Office, Obsidian, and Notion.