Geometric AI · Neurotechnology · Wearable Health
Learning with geometry,
for signals that matter.
My research explores geometry-aware learning for neural signals, matrix manifolds, and health intelligence, with an interest in both mathematical foundations and practical applications.
About
I received my M.S. degree in Artificial Intelligence from Jiangnan University in June 2026 under the supervision of Rui Wang. Since September 2025, I have been a visiting student in the Medical Artificial Intelligence Laboratory at Westlake University, advised by Yefeng Zheng, who will also supervise my doctoral studies at Westlake University.
Since March 2026, I have been an intern at the Ant Group Healthcare Lab, where I work on algorithms for wearable devices under the mentorship of Gengyan Zhao and Le Lv.
News
- I graduated from Jiangnan University with an M.S. degree in Artificial Intelligence.
- Three papers were accepted to MICCAI 2026.
- I was fortunate to receive a Ph.D. admission offer from Westlake University.
- I was honored to be selected as a Gold Reviewer for ICML 2026.
- CorSW was accepted to KDD 2026.
- I started an internship at the Ant Group Healthcare Lab, working on wearable health intelligence under the mentorship of Gengyan Zhao and Le Lv.
- RHOP was accepted to ICLR 2026.
- I was honored to receive the National Scholarship.
- I started as a visiting student in the Medical Artificial Intelligence Laboratory at Westlake University, advised by Yefeng Zheng.
- GyroAtt was accepted to NeurIPS 2025.
- CorAtt was accepted to IJCAI 2025.
Selected Publications
A selection of work on geometry, neural signals, and scalable learning.
A Sliced-Wasserstein Framework on Correlation Matrices for EEG Decoding
A sliced-Wasserstein framework on pullback Euclidean metric manifolds, evaluated for EEG domain generalization.
Riemannian High-Order Pooling for Brain Foundation Models
A plug-and-play Riemannian high-order pooling head for EEG foundation models.
Towards a General Attention Framework on Gyrovector Spaces for Matrix Manifolds
An attention framework across SPD, SPSD, and Grassmannian gyrovector spaces.
A Correlation Manifold Self-Attention Network for EEG Decoding
A self-attention method defined directly on full-rank correlation manifolds for EEG representation learning.
Full Publication List
-
KDD 2026
-
MICCAI 2026Riemannian Batch Normalization on Correlation Manifolds for EEG Decoding
-
MICCAI 2026Hierarchical Hyperbolic Self-Attention Network for Medical Image Segmentation
-
MICCAI 2026MedPruner: Training-Free Hierarchical Token Pruning for Efficient 3D Medical Image Understanding in Vision-Language Models
-
ICLR 2026
-
PRCV 2025
-
NeurIPS 2025
-
IJCAI 2025
-
IJCAI 2024
-
ERA 2024
Honors and Awards
Education
M.S. in Artificial Intelligence
B.E. in Data Science and Big Data Technology