Hi! I am Lingzhi Wang (Lexus). I am an incoming tenure-track Assistant Professor in the Department of Integrated Information Technology at the University of South Carolina, starting August 2026. I received my PhD in Computer Science from Northwestern University, advised by Prof. Yan Chen, and my Bachelor’s degree from Tsinghua University in 2020.

I will be recruiting PhD students in AI-driven cybersecurity for Spring/Fall 2027. More details about my group and recruiting will be shared soon. Stay tuned!

My research focuses on applying advanced machine learning and artificial intelligence techniques to strengthen both cyber offense and defense. Specifically, I explore the following questions:

• How can cutting-edge AI methods be integrated into practical security and network systems such as Security Operation Centers (SOCs) and Extended Detection and Response (XDR) systems?
• How can we model human knowledge and reasoning with LLMs to carry out complex offensive cybersecurity tasks such as penetration testing and red teaming?
• How can we build higher-quality research infrastructure for cybersecurity, including datasets, benchmarks, testbeds, and knowledge bases, to support rigorous and reproducible research?
• As LLMs and autonomous agents are increasingly deployed in real-world production environments, how can we identify, assess, and mitigate the emerging security risks they introduce?

News 📰

• 2026.6 I’m thrilled to share that I will be joining the Department of Integrated Information Technology at the University of South Carolina as a tenure-track Assistant Professor starting August 2026. I will be recruiting PhD students in AI-driven cybersecurity for Fall 2027 — stay tuned!
• 2026.3 I will be joining Obsidian Security as a Research Intern this spring, working on security monitoring and governance for multi-agent AI systems.
• 2026.2 I presented a tutorial on using our Aurora system for automated cyberattack emulation and benchmark dataset generation at PRISM’26 (co-located with NDSS’26). Stay tuned for materials and slides!
• 2026.2: Our Work-In-Progress paper Building Next-Generation Datasets for Provenance-Based Intrusion Detection has been accepted and presented at Workshop on Attack Provenance, Reasoning, and Investigation for Security in the Monitored Environment (PRISM), co-located with NDSS’26. Here is the paper.
• 2026.1: I started my internship in USC Information Sciences Institute on the SPHERE Research Infrastructure project.
• 2025.11: Our paper From Sands to Mansions: Actionable, Customizable, and Causality-Preserving Cyberattack Emulation with LLM-powered Symbolic Planning has been accepted by the 24th International Conference on Applied Cryptography and Network Security (ACNS 2026).
• 2025.9: Our paper GraphFaaS: Serverless GNN Inference for Burst-Resilient, Real-Time Intrusion Detection has been accepted to the Workshop on ML for Systems at NeurIPS 2025.
• 2025.9: Our paper Incorporating Gradients to Rules: Towards Online, Adaptive Provenance-based Intrusion Detection has been accepted by TDSC.
• 2025.6.23: I began my internship as a research intern at SRI.
• 2025.4 Our paper PentestAgent: Incorporating LLM Agents to Automated Penetration Testing has been accepted by AsiaCCS’25.
• 2025.2: I’ll be attending NDSS’25 in San Diego to present our work Incorporating Gradients to Rules: Towards Lightweight, Adaptive Provenance-based Intrusion Detection. See you in San Diego!
• 2024.12.6: I passed my PhD prospectus.
• 2024.8.30 Our paper Incorporating Gradients to Rules: Towards Lightweight, Adaptive Provenance-based Intrusion Detection has been accepted by NDSS’25.