Recently, the School of Artificial Intelligence held a high-level academic lecture in Meeting Room 711, Science and Technology Comprehensive Building at Jiaojiang Campus. Professor Xu Changsheng from the Institute of Automation, Chinese Academy of Sciences was invited to give a cutting-edge lecture: Research and Applications of Multimodal Large Models in the Open World.

Professor Xu Changsheng, Distinguished Young Scholar (NSFC) and IEEE Fellow, has long been engaged in multimedia artificial intelligence research with fruitful achievements and numerous awards. Focusing on the existing challenges of multimodal large models in the open world, the lecture elaborates on two core issues: multimodal data correlation modeling and cross-domain & few-shot generalization via transfer. Xu has introduced Libra, a self-developed decoupled visual multimodal model, explained a series of transfer algorithms for guaranteeing model generalization capability and analyzed approaches to constructing an ideal visual system. His lecture has sorted out frontier research directions of multimodal large models for teachers and students present.

The lecture has closely followed the latest hot topics in the development of the AI discipline, broadening the research horizons of faculty and students. The school will continue to invite top experts for academic exchanges, helping faculty and students keep pace with AI frontiers and consolidate their capabilities for scientific research and innovation.

At the Lecture