-
Speaker: Prof. Chun-Yi Lee, Department of Computer Science and Information Engineering, National Taiwan University
-
Topic: World Models and World-Action Models: Learning to Predict, Plan, and Act
-
Participants: All third-year and senior students, faculty and staff members, and physicians affiliated with Chang Gung
-
Venue: AI Lecture Hall, 11th Floor, Management Building
-
Date and Time: Tuesday, October 6, 2026, 1:00-3:00 PM
-
Online Meeting Link: https://reurl.cc/DYKy9e or scan the QR code on the poster
-
No registration is required. Please join the Teams meeting room at the scheduled time.
About the Speaker:
Chun-Yi Lee is a Professor in the Department of Computer Science and Information Engineering at National Taiwan University and the Director of Elsa Lab. He received his Ph.D. in Electrical Engineering from Princeton University. His research interests include deep reinforcement learning, intelligent robotics, computer vision, and generative AI. He is a recipient of the 2026 Academia Sinica Scholar Award and the K. T. Li Breakthrough Award.
Lecture Abstract:
World models learn to predict how an environment evolves and are rapidly becoming a foundation for physical AI. This talk will review recent advances, ranging from model-based reinforcement learning, such as DreamerV3, and generative interactive world models, including Genie and NVIDIA Cosmos, to self-supervised, prediction-based models such as Meta's V-JEPA and V-JEPA 2.
The talk will then introduce world-action models, which combine learned world models with actions for planning and control. Through examples such as V-JEPA 2-AC and Cosmos Policy, the speaker will explain how these models enable zero-shot robotic planning. The talk will conclude with a discussion of the major open challenges and future opportunities in this field.
