Education

Xi’an Jiaotong–Liverpool University (XJTLU) — Suzhou, China
B.Sc. in Information and Computing Science, Sep 2022 – Present
GPA (WES): 3.53 / 4.0
Relevant courses: Computer Networks, Linear Algebra, Advanced Java Programming, Data Structures, Databases, Computer Systems, Software Engineering

Research Experience

Leveraging Large Language Models for CTF Challenge Solving — XJTLU Laboratory, Suzhou, China
Aug 2024 – Oct 2024 — Advisor: Prof. Wenjun Fan

• Built an automated penetration testing framework based on PentestGPT
• Combined decision-tree analysis, command execution, and LLM feedback loops to optimize CTF solving
• Research published at TrustCom 2024 (Y. Zou, Y. Hong, et al.)

Auditing Sensitive Information Leakage in Implicit LLM Reasoning — BUPT, Beijing, China
Jan 2025 – Sep 2025 — Advisor: Prof. Guoshun Nan

• Investigating implicit knowledge invocation and privacy risks in latent Chain-of-Thought reasoning
• Submitted to ICML 2026

AI-driven Automation for Security Alert Triage — DBAPPSecurity Central Research Institute, Hangzhou, China
May 2025 – Aug 2025 — Advisor: Shaoxing Ye

• Developed components of a security agent for malicious traffic analysis
• Designed behavioral analysis modules for fine-grained alert classification (successful/failed attacks, false positives, disguised traffic)
• Produced structured threat intelligence to support IDS filtering and attacker behavior detection

LLM-Driven Penetration Testing Framework — University of Washington Tacoma (collab), Suzhou, China
Aug 2025 – Present — Advisor: Prof. Wenjun Fan

• Initiating a self-designed, LLM-driven pentest framework beyond PentestGPT
• Early work: modular multi-agent prototype (recon + exploit synthesis) and small-scale evaluation setup

Work Experience

Operations & Security Intern, edscha (AnHui) — Hefei, China
Jan 2025 – Feb 2025

• Monitored core IT systems with Linux operations tools
• Automated monitoring and incident response using Shell/Python scripts
• Configured firewalls and vulnerability scanners

Publications

Leveraging Large Language Models for Challenge Solving in Capture-the-Flag

Y. Zou, Y. Hong, J. Xu, L. Liu and W. Fan, "Leveraging Large Language Models for Challenge Solving in Capture-the-Flag," 2024 IEEE 23rd International Conference on Trust, Security and Privacy in Computing and Communications (TrustCom), Sanya, China, 2024, pp. 1541-1550. doi: 10.1109/TrustCom63139.2024.00213.

Skills

Programming: Python, Java, C/C++, LaTeX
Cybersecurity: CTF (Web), penetration testing, malware traffic analysis
Languages: Chinese (native), English (fluent)

Service & Leadership

• Contributor to CTF-related research integrating Large Language Models
• Collaboration with academic and industry labs on cybersecurity and AI safety