Recent advances in visual generative models have enabled increasingly realistic and flexible content creation, raising important questions about controllability and trustworthiness. In this talk, Dr.Lo will present her recent efforts toward shaping and securing visual generative AI, including fine-grained image and video generation, balancing generative fidelity and diversity, explainable deepfake detection, and image immunization against unauthorized manipulation. Together, her works explore how generative AI can better reflect human intent while enabling more reliable interactions with visual content.

Ling Lo is an Assistant Professor in the Department of Computer Science at National Tsing Hua University (NTHU). She received her Ph.D. and B.S. degrees in Electronics Engineering from National Yang Ming Chiao Tung University (NYCU). She was also a visiting scholar at the University of California, Merced, and the University of Tokyo. Her research lies at the intersection of computer vision and machine learning, with a focus on generative AI for multimedia, trustworthy visual content creation, and affective computing. Her work has been published in leading venues, including CVPR, ICCV, AAAI, ACM MM, IEEE TMM, ACM TOMM, and IEEE TAFFC.
She received the Best Paper Award at IEEE ICME 2021.