Welcome to Hongwei Wu’s personal website!

About Me

I am currently pursuing my Ph.D. in Computer Science at Purdue University under the guidance of Prof. Antonio Bianchi. I completed my Master of Science in Computer Science at Binghamton University, SUNY, where I was recognized as a Distinguished Graduate Student. Prior to that, I earned my Bachelor of Science in Information Security from Renmin University of China. My research focuses on program analysis, binary analysis, and reverse engineering, with a particular emphasis on leveraging large language models and symbolic execution to verify patch safety, generate human-readable descriptions of binary patches, and enhance vulnerability management for both modern and legacy embedded systems.

Program Contributions

DARPA AI Cyber Challenge (AIxCC): Spearheaded the development of Artiphishell with team Shellphish — an LLM-based Cyber Reasoning System that autonomously discovers and patches vulnerabilities. Won a $2M semifinal award and placed 5th in the final at DEF CON 33 (out of 42 initial teams). [blog]
DARPA Assured Micropatching (AMP): Applied and extended VeriBin to verify binary-level micropatches on legacy embedded systems (Cummins engine ECUs, NASA lunar rovers, power-grid controllers), enhancing angr’s symbolic execution across ARM, PowerPC, and other embedded architectures.

Publications

Patch2Text: Generating Human Readable Descriptions of Binary Patches from Stripped Binaries
Hongwei Wu, Han Dai, Antonio Bianchi
In submission, 2026.
D-LiFT: Improving LLM-based Decompiler Backend via Code Quality-driven Fine-tuning
Muqi Zou, Hongyu Cai, Hongwei Wu, Zion Leonahenahe Basque, Arslan Khan, Berkay Celik, Dave (Jing) Tian, Antonio Bianchi, Ruoyu (Fish) Wang, and Dongyan Xu
In submission, 2026.
VeriBin: Adaptive Verification of Patches at the Binary Level [paper][code]
Hongwei Wu, Jianliang Wu, Ruoyu Wu, Ayushi Sharma, Aravind Machiry, and Antonio Bianchi
In Proceedings of the Network and Distributed System Security Symposium (NDSS), 2025.