Beatrice Casey
University of Notre Dame. bcasey6@nd.edu
Univeristy of Notre Dame
South Bend, Indiana 46617
Beatrice Casey is a PhD Candidate and Graduate Research Assistant in the Department of Computer Science and Engineering at the University of Notre Dame (ND) and is a recipient of the Arthur J. Schmitt Presidential Leadership Fellowship. She received her Bachelor of Science in Computer Science, with a minor in Physics, from Bucknell University where she was also a Presidential Fellow. She works under Dr. Joanna Cecilia da Silva Santos in the Security and Software Engineering research lab (S2E).
Beatrice’s research sits at the intersection of Security, Machine Learning, and Quantum Computing. Her work spans two areas: the security of ML model supply chains (including serialization vulnerabilities in model-sharing platforms like Hugging Face), and quantum-enhanced detection techniques for software and machine learning security. She is currently leading an NSF-funded project (as part of the Center for Quantum Technologies), Q-DART, which investigates the use of quantum techniques for real-time data tampering detection.
Check out her publications and get in touch below!
news
I’m excited to share that our paper, Towards Quantum-Based Detection of Adversarial Attacks in Natural Language Processing Systems, was accepted to the IEEE International Conference on Quantum Computing & Engineering (QCE) 2026! In this work, we explore a hybrid quantum-classical machine learning approach for detecting adversarial examples in natural language processing models, using projected quantum kernels with support vector machines. The paper will be presented at QCE in September 2026. See you in Toronto!
This past Friday, December 5th 2025, I successfully passed the Oral Candidacy Exam, officially becoming a PhD candidate. My dissertation proposal surrounds security in machine learning-based systems covering three major phases: model creation, storage, and usage. I would like to thank everyone who came to support me during my presentation, and my committee (Dr. Joanna Cecilia da Silva Santos, Dr. Taeho Jung, Dr. Peter Kogge, and Dr. Gail Kaiser) for their guidance and feedback. I am looking forward to continuing on with this work and enhancing security for machine learning-based systems.
Our paper, A Survey of Source Code Representations for Machine Learning-Based Cybersecurity Tasks, was accepted for publication in ACM Computing Surveys! In this work, we systematically review how source code is represented in ML-based cybersecurity research. Data representation is a key part of the ML pipeline and understanding what each representation offers in terms of feature information allows researchers to choose or create a representation that best suits their needs.
