See you at ICSE 2026!
Our journal first article, “Evaluating and Improving the Robustness of Security Attack Detectors Generated by LLMs,” has been accepted to the IEEE/ACM International Conference on Software Engineering (ICSE) 2026!
In this work, we investigate how LLMs generate security detectors, proposing a pipeline that combines RAG and self-ranking to improve robustness in the generated code.
I’ll be in Rio de Janeiro between April 12-18 to present this work, connect with the research community, and explore new directions at the intersection of LLMs, deep learning, NLP, security, and code generation.