Profile

I am a Ph.D Candidate at National University of Singapore (NUS), working with Prof. Feng Mengling. Before Ph.D, I graduated from NUS with a master’s degree in electrical & computer engineering and did my research intern at IBM Research.

Research Interests: My research focuses on designing and applying AI models to address real-world healthcare challenges, with a long-term goal of building expert-level multimodal generalists to enhance clinical decision-making and patient care. My recent work concentrates on advancing multimodal large language models, leveraging their remarkable capacity for synergizing diverse modalities for reasoning and planning. Previously, I worked a lot on self-supervised learning models to improve label efficiency in the medical domain. I am also interested in time series modeling, and much of my work has involved time series data, as it is one of the most prevalent modalities in healthcare. I have published 10+ papers at the top international AI conferences and journals such as ICLR, ECCV, AAAI, Information Fusion, and TIST.

πŸ”₯ News

  • [03. 2025] Our paper on empowering MLLM for grounded ECG understanding now available at arXiv.
  • [06. 2024] Our paper on multi-stage contrastive learning is accepted in ECCV 2024.
  • [06. 2024] I was awarded for the Graduate Student Research Award AY2023/2024!
  • [01. 2024] Our paper on contrastive learning for time series applications is accepted in ICLR 2024.
  • [10. 2023] Our survey on LLM for healthcare applications is accepted in Information Fusion.
  • [05. 2022] Our paper about self-supervised learning for ECG is accepted in AAAI 2022.

πŸ“ Publications

Preprint 2025
sym

GEM: Empowering MLLM for Grounded ECG Understanding with Time Series and Images

Xiang Lan, Feng Wu, Kai He, Qinghao Zhao, Shenda Hong, Mengling Feng

Project Contributions

  • First Unified Multimodal ECG Model.
  • First High-granularity ECG Grounding Dataset.
  • Clinically Oriented Diagnostic System.

[Page][Code]

ICLR 2024
sym

Towards Enhancing Time Series Contrastive Learning: A Dynamic Bad Pair Mining Approach

Xiang Lan, Hanshu Yan, Shenda Hong, Mengling Feng

Project Contributions

  • First study to investigate the bad positive pair problem exists in time series contrastive learning.
  • A simple yet effective algorithm designed as a lightweight plug-in.
  • Enhancing the performance of existing state-of-the-art methods.

[Paper][Code]

ECCV 2024
sym

Learning the Unlearned: Mitigating Feature Suppression in Contrastive Learning

Jihai Zhang*, Xiang Lan*, Xiaoye Qu, Yu Cheng, Mengling Feng, Bryan Hooi
(*co-first author)

Project Contributions

  • First multistage contrastive learning framework.
  • First work to discuss and address feature suppression in both unimodal and multimodal contrastive learning.
  • Adaptable to various contrastive learning settings.

[Paper][Code]

AAAI 2022
sym

Intra-Inter Subject Self-Supervised Learning for Multivariate Cardiac Signals

Xiang Lan, Dianwen Ng, Shenda Hong, Mengling Feng

Project Contributions

  • First work that integrates medical knowledge into self-supervision to boost the performance of cardiac arrhythmias diagnosis.
  • Novel intra and inter subject self-supervision mechanism.
  • State-of-the-art performance.

[Paper][Code]

πŸŽ– Honors and Awards

  • 2024 Graduate Student Research Award, NUS
  • 2020 Championship, SG Healthcare AI Datathon 2020
  • 2020 2nd Runner-up, PhysioNet/Computing in Cardiology Challenge 2020

πŸ“– Educations

  • 2021.07 - now, Doctor of Philosophy, National University of Singapore.
  • 2018.07 - 2019.06, Master of Science, National University of Singapore.
  • 2014.07 - 2018.06, Bachelor of Science, University of Electronic Science and Technology of China.

🌍 Academic Services

I serve as a reviewer for ICLR, WWW, AAAI, KDD, TNNLS, TIST, Health Data Science.