Wei Sun’s Personal Website
I am a PhD researcher in Computer Science at KU Leuven, supervised by Prof. Marie-Francine Moens. My research connects medical natural language processing with multilingual large language models, with a focus on efficient adaptation, knowledge editing, and explainable AI.
Expected graduation · January 2027.
Previously, I completed my master's degree at Aalto University under the supervision of Prof. Pekka Marttinen, and my bachelor's degree at Macau University of Science and Technology under Prof. Yanyan Liang. My work brings together method development and clinical applications through collaborations with European research projects and hospitals.
Research interests
- Medical language understanding: Clinical information extraction, temporal relations, and automated medical coding.
- Multilingual language models: Knowledge editing, cross-lingual transfer, and inference-time intervention.
- Efficient & explainable AI: Parameter-efficient fine-tuning, evidence-based explanations, and weakly supervised argument mining.
Selected research
All publicationsCross-lingual Representation Learning via Centroid Intervention Fusion
Fuses multilingual intervention projections into a shared operator to improve cross-lingual transfer without updating model parameters.
Event Interaction in Low-Rank Bottlenecks for Temporal Relation Extraction
Models event-pair interactions inside low-rank bottlenecks for temporal relation extraction with Adapter and LoRA fine-tuning.
Mitigating Negative Interference in Multilingual Sequential Knowledge Editing through Null-Space Constraints
LangEdit constrains sequential knowledge updates in null spaces to reduce interference across languages and previously edited knowledge.
Generating Explanations in Medical Question-Answering by Expectation Maximization Inference over Evidence
Jointly learns evidence selection, answer prediction, and explanation generation using expectation maximization over medical evidence.
Research projects
My contributions to collaborative European research projects.
Medical language models & multilingual information extraction
Led research on medical language model pretraining and information extraction, from domain knowledge integration and parameter-efficient adaptation to multilingual temporal relation modeling and evaluation.
Evidence-based explanations for medical AI
Led method development for medical evidence retrieval, question answering, and explanation generation. Developed weakly supervised argument mining to structure claims, evidence, and their relations.
Clinical information extraction with limited supervision
Led low-resource information extraction research for Finnish clinical text, using medical ontologies to construct weak supervision and develop domain-adapted named entity recognition models.
Get in touch
For research inquiries and collaboration, please contact me by email.
s872078724@gmail.com