RIS CUTTING EDGE FOURM
Toward Physics-native Intelligence: World Model, Data Structure and Training Algorithm
Toward Physics-native Intelligence: World Model, Data Structure and Training Algorithm
Toward Physics-native Intelligence: World Model, Data Structure and Training Algorithm
- RISTA CUTTING EDGE FORUM Season 5 Session 12
- Time：8pm - 9pm, August 27th, 2026.
- Speaker guest：Prof.Shengbo Li
- Chair guest：Prof.Yongpeng Wu
As artificial intelligence advances beyond vision and language processing toward embodied systems—including autonomous driving and robotics—prevailing deep learning paradigms are confronting fundamental challenges. This talk presents a systematic overview of physics-native intelligence (Phi), covering its core design principles and methodologies for constructing its world model, data structure, and training algorithm. Four representative physics-native algorithms are elaborated: (1) RAD: A neural network optimizer engineered with built-in symplectic preservation to guarantee long-term training stability. (2) DACER: A reinforcement learning algorithm that equips neural network policies with the capacity to model multi-modal action distributions. (3) RACS: A ternary iteration framework for addressing safety constraints, which jointly learns the feasible region and optimal policy with provable monotonicity and convergence guarantees. (4) BOOM: An algorithm enabling bidirectional co-improvement with reliance on world model. By internalizing planning capabilities directly into the policy, it substantially enhances sample efficiency.
Shengbo Li is a Professor at School of Vehicle and Mobility, and School of Artificial Intelligence, Tsinghua University. Before joining Tsinghua University, he has worked at Stanford University, University of Michigan, and UC Berkeley. His research has established a systematic framework for physics-native intelligence (Phi), and its world models, data structure, and training algorithms. On its basis, he and his team have contributed to absolute safety guarantees and ternary iterative mechanism for reinforcement learning, symplectic neural network optimizer, multimodal latent world model, geometry-aware Bayesian filter, as well as their applications in autonomous driving and robotics. His important awards include National Leading Talents in Sci. and Tech. Innovation in China, Youth Sci. & Tech Award of Ministry of Education, and Youth Sci. & Tech. Innovation Leader in Transportation Sector, National Sci. & Tech. Progress Award in China (Second Prize), National Award for Technological Invention in China (Second Prize), Grand Prize of Science and Technology Award of China in Automotive Industry, Natural Science Award of Chinese Association of Automation (First Prize). He also serves as the director of Technical Committee on AI of SAE-China, deputy director of Technical Committee on Vehicle Control and Intelligence of CAA, and the leader of AI working group in China Industry Innovation Alliance for ICVs.
Related
