David Yuan

\(\bullet\) Email: yuandawei@gdust.edu.cn (wu.xiguanghua2014@gmail.com) \(\bullet\) GitHub: davidyuan666 \(\bullet\) Google Scholar: Profile \(\bullet\) ModelScope: ModelScope

\(\S\) Research Interests

\(\bullet\) Software Engineering - Code Analysis, Software Testing, and Maintenance
\(\bullet\) Large Language Models (LLM) - Applications in Software Engineering
\(\bullet\) Web3 - Smart Contract Security and Blockchain Technology
\(\bullet\) Embedded Systems - IoT and Real-time Systems

Macau University of Science and Technology
Ph.D. in Artificial Intelligence

University of Science and Technology of China
Master of Software Engineering

Nanjing University of Posts and Telecommunications
Bachelor of Network Engineering

2025 - Present Guangdong University of Science and Technology, Dongguan, Guangdong - Assistant Professor

May 2024 - 2025 Vesync, Shenzhen - Senior Algorithm Engineer

June 2023 - May 2024 BitAuto(易車), Beijing - Senior Algorithm Engineer

Dec 2022 - May 2023 Oujian Group, Shanghai - Algorithm Team Leader

Dec 2016 - Aug 2020 United Imaging Healthcare, Shanghai - Software Engineer

2025 - Present JavaEE - Teaching Course

\(\bullet\) Do Pretrained Language Models Indeed Understand Software Engineering Tasks?
IEEE Transactions on Software Engineering (TSE), 2023 [Link]
\(\bullet\) Java Code Clone Detection by Exploiting Semantic and Syntax Information From Intermediate Code-Based Graph
IEEE Transactions on Reliability (TR), 2023 [Link]
\(\bullet\) Optimizing smart contract vulnerability detection via multi-modality code and entropy embedding
Journal of Systems and Software (JSS), 2023 [Link]
\(\bullet\) Meta-Learning for Multi-Family Android Malware Classification
ACM Transactions on Software Engineering and Methodology (TOSEM), 2024 [Link]
\(\bullet\) Code Context-Based Reviewer Recommendation
Frontiers of Computer Science (FCS), 2023 [Link]
\(\bullet\) UNA: Improving Automated PL-NL System by A Unified Neural Architecture
IEEE Transactions on Reliability (TR), 2025 [Link]
\(\bullet\) UI2HTML: utilizing LLM agents with chain of thought to convert UI into HTML code
Automated Software Engineering, 2025 [Link]

Journal Reviewer