Papers
2
Total Citations
4
H-Index
2
About
Mai Xu is a leading researcher at the intersection of autonomous driving perception and robotic tactile intelligence. Her work centers on multimodal fusion for 3D object detection and the development of large tactile-vision-language models for physical property inference. In her 2024 paper "NCFDet," Xu introduced a novel approach that leverages the neural collapse phenomenon to enhance point cloud features in multimodal fusion, significantly improving 3D object detection accuracy for autonomous systems navigating complex urban environments. This work has already garnered early citations, reflecting its potential impact on the field. In 2025, Xu advanced robotic perception with her work on a large tactile-vision-language model, enabling robots to infer physical properties through integrated sensory modalities—a breakthrough for dexterous manipulation. Her research bridges critical gaps between vision, touch, and language, pushing the boundaries of how machines understand and interact with the physical world. With her innovative contributions to both safety-critical autonomous driving and embodied AI, Xu is establishing herself as a rising star in robotics and computer vision.
Research Focus
Key Achievements
Top Papers
- 1
- 2