معرفی
Ying Xu is a researcher at the Department of Computer Technology and Informatics, Norwegian University of Science and Technology (NTNU), serving as a Guest researcher with primary affiliation to NTNU. She recently completed her doctoral thesis at NTNU in 2025 titled "Generalizing Deepfake Detection: Towards Robust, Interpretable and Fair Models", establishing her expertise in multimedia security and artificial intelligence.
Her research centers on deepfake detection with critical focus on model robustness against evolving generative techniques, algorithmic fairness across demographic groups, and interpretability of detection mechanisms. She pioneers approaches including pairwise interaction modeling and supervised contrastive learning to address cross-dataset generalization challenges and mitigate biases in synthetic media analysis. Her work bridges computer vision, machine learning ethics, and digital forensics to develop trustworthy detection systems.
Analysis of her 2022-2025 publications reveals consistent innovation in adversarial machine learning for multimedia security, with increasing emphasis on societal implications of deepfake technology. She integrates fairness metrics with technical detection frameworks while advancing explainability in deep learning architectures, positioning her work at the intersection of technical rigor and ethical AI development.
Scientific awards: No awards documented in provided materials.
Advising and grants: Current information indicates no formal advisees or grant leadership roles. Her research appears conducted within NTNU's institutional framework without specified external funding mentions.
Labs and teams: While affiliated with NTNU's Department of Computer Technology and Informatics, no specific laboratory or research team affiliations are detailed in available documentation.
