CSE - Software Security Analysis using Machine Learning & Quantum Computing

Our lab's research projects are at the intersection of security and software engineering.

The S2E research lab (https://s2e-lab.github.io/)  has several main ongoing research projects related to:

• Relying on the power of Large Language Models (LLMs), such as ChatGPT and GitHub Copilot, to generate unit tests for security features.
• Enhancing the code generated by LLMs such that it is not only functionally correct, but also free of vulnerabilities.
• Exploring multi-agent architectures for software security analysis.
• Evaluating whether the safety guardrails in code LLMs hold when prompts are written in languages other than English.
• Building benchmarks and datasets that measure both the correctness and the security of LLM-generated code across programming languages, including low-resource ones.
• Designing guardrails for code LLMs used in CS courses, so students get scaffolded hints rather than complete solutions.
• Developing novel program analysis techniques to identify software vulnerabilities within code commits (i.e., verify whether a code change will introduce a new vulnerability in production).
• Using quantum computing to scale exploit generation and vulnerability detection
• Developing novel and secure model serialization formats. 

When applying, please submit both (1) your CV / Resume and (2) Transcript (an unofficial transcript is acceptable as long as the GPA and courses are clearly listed).