PhD Dissertation Proposal by ZHU Jiawen | Toward Generalist Anomaly Detectors

Toward Generalist Anomaly Detectors

ABOUT THE TALK

Anomaly Detection (AD) is essential across manufacturing, healthcare, and security applications, yet existing deep learning–based approaches remain highly specialized: they require dataset-specific training and struggle to generalize to unseen anomaly types or new domains. Real-world anomalies are rare, heterogeneous, and inherently unpredictable, motivating the development of generalist anomaly detectors capable of operating across diverse settings without task-specific retraining or large amounts of target data. This dissertation aims to advance the generalization capability of AD models through a progressive research trajectory spanning three increasingly challenging supervision regimes, culminating in a unified pathway from open-set supervised AD to few-shot generalist AD and ultimately zero-shot AD.

SPEAKER BIOGRAPHY

Jiawen Zhu is currently a fourth-year PhD candidate at Singapore Management University and supervised by Prof. Guansong Pang. She has published and presented multiple papers at top venues, including CVPR and ICCV. Her research interests lie in the field of computer vision and open-world learning, particularly in generalist anomaly detection.