Nan Xi, M.D., Ph.D.
Assistant Professor, Department of Computer Science VCU College of Engineering
Links
Research Focus
Research Focus
Dr. Xi's research focuses on empowering AI systems with the ability of visual reasoning akin to human (experts), with the goal of enhancing both the reliability and usability of these AI models, especially in medicine and healthcare applications.
Selected Articles
Chain-of-Look Spatial Reasoning for Dense Surgical Instrument Counting
Proceedings of the IEEE/CVF winter conference on applications of computer vision (WACV)Rishikesh Bhyri, Brian R Quaranto, Junsong Yuan, Peter C W Kim , Nan Xi
2026-03-06
We introduce Chain-of-Look, a novel visual reasoning framework that mimics the sequential human counting process by enforcing a structured visual chain, rather than relying on classic object detection that is unordered.
dFLMoE: Decentralized Federated Learning via Mixture of Experts for Medical Data Analysis
Proceedings of Computer Vision and Pattern Recognition (CVPR)Luyuan Xie, Tianyu Luan, Wenyuan Cai, Guochen Yan, Zhaoyu Chen, Nan Xi, Yuejian Fang, Qingni Shen, Zhonghai Wu, Junsong Yuan
2025-06-09
Decentralized federated learning framework
named dFLMoE that transmits each client’s knowledge to other clients and performs local decision-making on each client, effectively avoiding the knowledge damage caused by centralized server aggregation and eliminating the dependence on a central server.
Interaction-centric Spatio-Temporal Context Reasoning for Multi-person Video HOI Recognition
European Conference on Computer Vision (ECCV)Yisong Wang , Nan Xi *, Jingjing Meng, Junsong Yuan (* Correspondance)
2024-07-08
Interaction-centric Spatio-Temporal Context Reasoning for Multi-person Video HOI Recognition
Open Set Video HOI detection from Action- centric Chain-of-Look Prompting
Proceedings of the IEEE International Confer- ence on Computer Vision (ICCV)Nan Xi, Jingjing Meng, Junsong Yuan
2023-06-26
Open Set Video HOI detection from Action- centric Chain-of-Look Prompting
Chain-of-Look Prompting for Verb-centric Surgical Triplet Recognition in Endoscopic Videos
Proc. ACM International Conf. on Multimedia (ACM MM)Nan Xi, Jingjing Meng, Junsong Yuan
2023-06-12
Introducing the chain-of-look prompting and designing the underlying visual reasoning processes in endoscopic videos
for surgical triplet recognition, generating visual-semantic aware and spatio-temporal aware prompt features from VL models.
