Dr. Peiwei Hu (胡佩威)
Assistant Professor
School of Computer Science and Technology
Jilin University
Announcements
🎓 Graduate Student Recruitment (2026)
I am looking for motivated graduate students (Master's and PhD) to join my research group. Research areas include software security, LLM security and LLM for security. If you are interested, please send me an email with your CV.
About Me
Biography
I am an Assistant Professor at Jilin University. My research interests include improving the quality of open-source software and applying Large Language Models (LLMs) to traditional security areas. I have authored and co-authored papers in top-tier security conferences. I am also open to exploring other intriguing research topics.
Beyond academic research, I am an active contributor to the open-source community. I found 500+ defects/bugs in widely used open-source software. Among over 900 contributors to OpenSSL (25k+ stars on GitHub) worldwide, I have achieved the 24th highest ranking. I also rank 12th out of over 150 contributors to Rizin (2.7k+ stars on GitHub) and am a Linux Kernel contributor. These contributions reflect my commitment to improving software security and quality in real-world applications.
Research Interests
- Software Security
- LLM Security
- LLM for Security
Education
Ph.D. in Cybersecurity
Institute of Information Engineering, Chinese Academy of Sciences, 2020-2025
Advisor: Prof. Kai Chen
Awards:
- Director's Special Award, Institute of Information Engineering / 信息工程研究所所长特别奖
- National Scholarship / 国家奖学金
B.S. in Computer Science
Jilin University, 2016-2020
Publications
1. DeGPT: Optimizing Decompiler Output with LLM (NDSS 2024) [First Author]
2. AURC: Detecting Errors in Program Code and Documentation (USENIX Security 2023) [First Author]
3. TypeForge: Synthesizing and Selecting Best-Fit Composite Data Types for Stripped Binaries (IEEE S&P 2025) [Co-author]
4. Attention-Based Decompilation Through Neural Machine Translation (ICISC 2025) [Co-author]
5. Optimizing Decompiler Output by Eliminating Redundant Data Flow in Self-Recursive Inlining (ICSME 2024) [Co-author]
6. Boosting neural networks to decompile optimized binaries (ACSAC 2022) [Co-author]
7. The inconsistency of documentation: a study of online C standard library documents (Cybersecurity 2022) [Co-author]
8. Neutron: an attention-based neural decompiler (Cybersecurity 2021) [Co-author]
9. Semantics-recovering decompilation through neural machine translation (arXiv 2021) [Co-author]
10. Rtfm! automatic assumption discovery and verification derivation from library document for api misuse detection (CCS 2020) [Co-author]
Contact Information
peiwei@jlu.edu.cn
Office
Room A427, Wang Xianghao Building