- UPL-TTA: Uncertainty-aware pseudo label guided fully test time adaptation for fetal brain segmentation, IPMI 2023
- FPL+: Filtered Pseudo Label-based Unsupervised Cross-Modality Adaptation for 3D Medical Image Segmentation, IEEE TMI (accepted)
Research Experience
Currently open to research collaboration and visiting/internship opportunities related to LLM post-training, test-time adaptation, medical image analysis, and foundation models.
Education
PhD student at Monash University; Master's degree from the University of Electronic Science and Technology of China (UESTC), supervised by Prof. Guotai Wang and Prof. Shaoting Zhang; Member of HiLab.
Background
Research interests: Post-training adaptation of medical AI models, ranging from lightweight deep learning networks to large-scale foundation models. Research focuses on domain adaptation, test-time training, and self-supervised learning to enhance the reasoning-time robustness and generalizability of visual, language, and multi-modal systems in real-world medical scenarios.