About Me
I am a Ph.D. student in the Artificial Intelligence Thrust at the Hong Kong University of Science and Technology (Guangzhou), supervised by Dr. Jintai Chen. My current doctoral research focuses on ECG signal analysis, rule mining for tabular data, and drug discovery.
I earned my Master's degree in Artificial Intelligence from Xiamen University, where I was advised by Prof. Min Jiang. During my master's studies, I focused on building trustworthy AI algorithms, specifically enhancing their security, robustness, interpretability, and fairness.
I hold a Bachelor's degree in Computer Science and Technology from Guangzhou University of Chinese Medicine, where I was advised by Prof. Wu Zhou. My academic journey has been driven by a passion for bridging machine learning theory with practical applications in high-stakes domains, particularly in clinical medicine.
Education
Ph.D. in Artificial Intelligence
The Hong Kong University of Science and Technology (Guangzhou)
Jan. 2026 - Present
Master of Artificial Intelligence
Xiamen University
Sep. 2022 - Jun. 2025
Bachelor of Computer Science
Guangzhou University of Chinese Medicine
Sep. 2018 - Jun. 2022
Research Experience
AI for Healthcare and Tabular Data Mining
Developing advanced deep learning models for precise ECG classification and automated rule mining from clinical trial tabular data.
Security and Robustness of AI Algorithms
This research reveals how attackers can subtly and efficiently alter AI outputs, like image captioning and object detection, without internal model access.
Interpretability and Fairness of AI Algorithms
This research underscores the need for fairness in ML algorithms, examining how subgraph patterns cause bias in GNNs and suggesting mitigation strategies.
AI Algorithms for Clinical Tumor Characterization
This research exposes the shortcomings of conventional mathematical models in DWI and DCE-MRI for tumor analysis, proposing an AI model with spatiotemporal attention for direct prediction.
Publications
Publication Summary: total: 14 papers including 8 first-author papers
(* denotes corresponding author, co-first authors are equally contributed.)
-
Fading the Digital Ink: A Universal Black-Box Attack Framework for 3DGS Watermarking Systems. (2026)
In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI). CCF-A
-
Ask, Attend, Attack: An Effective Decision-Based Black-Box Targeted Attack for Image-to-Text Models. (2024)
In Proceedings of the Neural Information Processing Systems (NeurIPS). CCF-A
-
VisionLaw: Inferring Interpretable Intrinsic Dynamics from Visual Observations via Bilevel Optimization. (2026)
In Proceedings of the International Conference on Learning Representations (ICLR). CCF-A
-
Multi-View Subgraph Neural Networks: Self-Supervised Learning with Scarce Labeled Data. (2024)
IEEE Transactions on Neural Networks and Learning Systems (TNNLS). JCR-Q1
-
Cross-Task Attack: A Self-Supervision Generative Framework Based on Attention Shift. (2024)
In Proceedings of the International Joint Conference on Neural Networks (IJCNN). CCF-C
-
Generating diagnostic and actionable explanations for fair graph neural networks. (2024)
In Proceedings of the AAAI Conference on Artificial Intelligence (AAAI). CCF-A
-
IVIM using convolutional neural networks predicts microvascular invasion in HCC. (2022)
European Radiology. JCR-Q1
-
An attention-based deep learning model for predicting microvascular invasion of hepatocellular carcinoma using an intra-voxel incoherent motion model of diffusion-weighted magnetic resonance imaging. (2021)
Physics in Medicine and Biology. JCR-Q1
-
An attention based deep learning model for direct estimation of pharmacokinetic maps from DCE-MRI images. (2021)
In 2021 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). CCF-B
-
Inducing Implicit Focus through Structured Irrelevance for Robust Neural PDE Solvers. (2025)
In Proceedings of the International Conference on Machine Intelligence and Data Science (MIND). Accepted
-
DualCAM++: Dual-saliency Guided Sample Selection for Plug-and-Play Backdoor Enhancement. (2025)
In Proceedings of the International Conference on Machine Intelligence and Data Science (MIND). Accepted
-
Black-Box Watermark Removal Using Diffusion Models and Self-Attention Mechanisms. (2025)
In Proceedings of the International Conference on Machine Intelligence and Data Science (MIND). Accepted
-
Adversarial Attacks Boosted by Gradient-Evolutionary Multiform Optimization. (2026)
In Proceedings of IEEE Conference on Computer Vision and Pattern Recognition(CVPR). CCF-AUnder Review
-
Text as a Compass: Semantic-Navigated 3D Bone Collapse Prediction via Orthogonal Micro-Volume Primitives. (2026)
In Proceedings of the Medical Image Computing and Computer Assisted Intervention (MICCAI). CCF-BUnder Review
Academic Service
- Program Committee for AAAI Conference on Artificial Intelligence (AAAI)
- Program Committee for Computer Vision and Pattern Recognition (CVPR)
- Reviewer for IEEE Transactions on Circuits and Systems for Video Technology (TCSVT)
- Reviewer for IEEE Transactions on Neural Networks and Learning Systems (TNNLS)
- Reviewer for IEEE Computational Intelligence Magazine (CIM)
- Reviewer for IEEE Transactions on Evolutionary Computation (TEVC)
- Reviewer for IEEE Transactions on Cognitive and Developmental Systems (TCDS)
Honors & Awards
- Second Prize in the National Traditional Chinese Medicine College Programming Contest, 2021
- Third Prize in the China Collegiate Programming Contest - Guangdong Provincial, 2021
- Provincial Project in the College Students' Innovative Entrepreneurial Training Plan Program, 2021
- Second Prize in the Blue Bridge Cup National Software and Information Technology Talents Competition, 2020
- Provincial Excellence Award in the China Undergraduate Mathematical Contest in Modeling (Guangdong), 2020
- Second Prize in Comprehensive Scholarship of Guangzhou University of Chinese Medicine, 2020
Work Experience
Medical Imaging Algorithm Engineer
Conducted research on Deep Neural Networks (DNN) to accelerate the computation of pharmacokinetic (PK) parameter maps in Dynamic Contrast-Enhanced Magnetic Resonance Imaging (DCE-MRI), enhancing the efficiency and accuracy of diagnostic tools.
Skills
- English Level: IELTS score (Overall = 6.5, Listening = 7, Reading = 8, Writing = 6, Speaking = 5.5)
- Programming Skill: Python, C++, Unity, Pytorch, Tensorflow