Xing Su

About Me

I am a Ph.D. student in Computer Science and Technology at Nanjing University, advised by Prof. Fengyuan Xu in the COSEC Group, led by Prof. Sheng Zhong. I expect to graduate in March 2027.

My research focuses on AI for Security (AI4Sec), combining large language models with program analysis to understand programs (e.g., smart contracts) and real-world (on-chain) attacks. I am particularly interested in AI agents for cybersecurity, tool-assisted vulnerability investigation, and blockchain security.

I am seeking academic and industry research positions, including postdoctoral opportunities, and welcome research collaborations. Please feel free to contact me.

Publications

First-author papers

ASE 2026CCF-A

Trace-Driven Automated PoC Synthesis for On-Chain Attacks

Xing Su, Hao Wu, Hanzhong Liang, Yunlin Jiang, Yuxi Cheng, Yating Liu, Fengyuan Xu, and Sheng Zhong.

41st IEEE/ACM International Conference on Automated Software Engineering, 2026.

Paper Website

FSE 2025CCF-A

DiSCo: Towards Decompiling EVM Bytecode to Source Code using Large Language Models

Xing Su, Hanzhong Liang, Hao Wu, Ben Niu, Fengyuan Xu, and Sheng Zhong.

Proceedings of the ACM on Software Engineering, 2(FSE), Article FSE103, 2025.

Paper Website

Collaborative papers

JCRD 2022CCF-T1

Security Testing of Visual Perception Module in Autonomous Driving System

Hao Wu, Hao Wang, Xing Su, Minghao Li, Fengyuan Xu, and Sheng Zhong.

Journal of Computer Research and Development, 59(5): 1133–1147, 2022. In Chinese.

Paper

WWW 2021CCF-A

DAPter: Preventing User Data Abuse in Deep Learning Inference Services

Hao Wu, Xuejin Tian, Yuhang Gong, Xing Su, Minghao Li, and Fengyuan Xu.

Proceedings of the Web Conference, 2021, pp. 1017–1028.

Paper

Preprints

Revealing Adversarial Smart Contracts through Semantic Interpretation and Uncertainty Estimation

Yating Liu, Xing Su, Hao Wu, Sijin Li, Yuxi Cheng, Fengyuan Xu, and Sheng Zhong.

arXiv:2509.18934, 2025.

arXiv

ConneX: Automatically Resolving Transaction Opacity of Cross-Chain Bridges for Security Analysis

Hanzhong Liang, Yue Duan, Xing Su, Xiao Li, Yating Liu, Yulong Tian, Fengyuan Xu, and Sheng Zhong.

arXiv:2511.01393, 2025.

arXiv

Experience

Qi-Anxin · Institute of Applied Sciences

Dec 2024 – Jun 2026

Vulnerability Research & Exploitation Engineer Intern

LLM-assisted firmware auditing: combined program context with model-based reasoning to identify authentication logic, assess sanitizer bypassability, and prioritize taint-analysis alarms.
Security intelligence: contributed to automated collection and structured analysis of public disclosures on Twitter, turning unstructured posts into usable security intelligence.

Technical Skills

Blockchain: EVM bytecode and source code analysis, transaction trace analysis, attack reproduction.
AI & machine learning: LLM-assisted code analysis and generation, graph neural networks.
Cybersecurity: Vulnerability research, taint analysis, firmware auditing, fuzzing.
Program analysis: Control-flow and data-flow analysis, symbolic reasoning, decompilation.

Service

Academic (sub)reviewing (2025–2026). Participated in reviewing for USENIX Security, IEEE S&P, KDD, IEEE TIFS, Computers & Security, and Chinese Journal of Computers.
Teaching assistant. Cybersecurity & Privacy, taught in English, Fall 2022.
Community service. Helped organize cybersecurity courses at the Jiangsu graduate summer school (2024–2026); defense committee member for the vulnerability analysis track at DataCon 2024.

Awards

DeFiHackLabs open-source contribution
Contributed 33 attack PoCs in two days; all accepted and merged. Awarded $900 in bounties.
2025
National First Prize
China Undergraduate Mathematical Contest in Modeling
2016–2017
Honorable Mention
Mathematical Contest in Modeling
2017–2018