Assistant Professor Yaoyao Liu has been awarded a grant through the NVIDIA Academic Grant Program. NVIDIA, a world leader in accelerated computing and AI, established the program to advance academic research by providing world-class computing access and resources to researchers. Liu has received 32,000 A100 GPU-hours on Brev, an AI and machine learning platform that empowers developers to run, build, train, deploy, and scale AI models with GPU in the cloud.
For his project, "Multi-modal Generative Models for Large-scale Continual Learning," Liu aims to build AI systems that can continuously learn and adapt over time, rather than being trained once and becoming outdated.
"Today's AI models often 'forget' previously learned knowledge when acquiring new information, which limits their effectiveness in dynamic, real-world environments," said Liu. "Our approach enables AI to incorporate new knowledge, such as segmenting new types of organs and tumors in CT scans, adapting to evolving driving scenarios, or understanding emerging concepts across text, images, and 3D data, without losing what it has already learned."
According to Liu, by allowing the system to expand its capabilities in a structured and efficient manner, this project makes AI more reliable, scalable, and practical for real-world applications where information is constantly evolving.
Liu's research lies at the intersection of computer vision and machine learning. He is especially interested in building intelligent visual systems—computer-based technologies that extract and interpret visual data in real time—that are continual and data-efficient. He earned his PhD in computer science at Max Planck Institute for Informatics and his bachelor’s degree in electronic information engineering at Tianjin University.