Dr. Xiaoxue Ma
馬笑雪 PhD Graduate (2024) Ph.D (CityU, HK), M.Eng (HKUST, HK), B.Eng (BUPT, CN)About
Dr. Xiaoxue Ma is a PhD graduate in the Department of Computer Science, City University of Hong Kong.
Senior Lecturer in the Department of Electronic Engineering and Computer Science at Hong Kong Metropolitan University. Her research covers software testing and quality assurance, with current emphasis on log anomaly detection, autonomous driving system testing, and metamorphic testing. She and her collaborators examined how data resampling methods influence deep learning based log anomaly detection, finding that oversampling on raw data outperforms undersampling and feature space resampling. She also helped construct a unified benchmark for out of distribution detection in autonomous driving systems, and studied how imbalance loss functions affect deep learning based vulnerability detection. Her group applies this work to validating large language models, including tests of code generation stability. Her findings guide practitioners in building reliable test oracles and robust prediction models for modern automated systems.
Research Interests
Publications (33)
-
CoRR2026
-
APSEC2025
-
APSEC2025
-
CoRR2025
-
COMPSAC2024
-
COMPSAC2024
-
COMPSAC2024
-
Internetware2024
-
Int.2023ConferenceLearning Gaps in Project-based Requirements Engineering Education - A Case Study of Student ProjectsInt. Symp. Educational Technology (ISET)