About Me
Currently, I am a Senior Applied Scientist at Amazon, working on physical AI, agentic workflows, and model customization. Previously, I was a Senior Researcher at Tencent in Singapore, where I worked on 3D model generation for computer games, and a Research Scientist at Motional, focusing on semantic scene understanding for autonomous driving. Before that, I was a Research Fellow at the National University of Singapore (NUS), where I also earned my Ph.D. under the supervision of Prof. Marcelo H Ang Jr and co-advised by Prof. Gim Hee Lee. My research interests include physical AI, agentic workflows, model customization, deep learning, computer vision, and 3D learning. I am particularly interested in translating advances in foundation models into practical, reliable, and adaptable intelligent systems for real-world applications.
Publications
Inject, Encode, Navigate: Feature-Wise Time Injection for Rectified Flow with Bidirectional Feature Encoding
Xiaogang Wang, Debasish Mishra, Jia Hao Toh, Wangsheng Liu and Yash Shah
IROS, 2026
Understanding the Limitations of Medical Reasoning in Large Language Models
Bill Cai, Xiaogang Wang, Ujjwal Ratan and Yash Shah
Machine Learning for Health (ML4H), 2025
Paper
BIMS-PU: Bi-Directional and Multi-Scale Point Cloud Upsampling
Yechao Bai*, Xiaogang Wang*, Marcelo H. Ang Jr and Daniela Rus
RA-L + IROS presentation, 2022
Paper
Cascaded Refinement Network for Point Cloud Completion with Self-supervision
Xiaogang Wang, Marcelo H Ang Jr and Gim Hee Lee
T-PAMI, 2021
Paper