Foundation Model for Science

Our research in foundation models for science focuses on developing large multimodal models that can process and comprehend complex scientific data. By integrating diverse data types, including texts, audio, images, and 3D structures, our models aim to uncover novel patterns and assist in scientific discovery. These models are built to advance scientific understanding and bridge gaps between theoretical and applied research, fostering cross-disciplinary innovation.

Foundation Model for Robotics

Our robotics research leverages foundation models to create intelligent robotic systems capable of real-world problem-solving. By utilizing multimodal inputs, including audio, vision, and haptics, our models enable robots to perform complex tasks in dynamic environments. This approach aims to push the boundaries of robotics, enabling them to interact with humans and their surroundings, ultimately advancing the pursuit of AGI in robotics.

Center for Foundation Models and Generative AI

The Center for Foundation Models and Generative AI (CFMG) aims to synergize cutting-edge AI technologies with human civilization. Led by distinguished members of the National Academy of Sciences and National Academy of Engineering, the CFMG and focuses on promoting high-impact, multidisciplinary generative AI research across the broad Midwest region of the United States.