Fengbei Liu

Postdoctoral Researcher at Cornell Tech, Cornell University

Computer vision · Medical image analysis · Label-efficient learning

About

I am a Postdoctoral Researcher at Cornell Tech, Cornell University, working with Prof. Mert R. Sabuncu. I obtained my Ph.D. from the Australian Institute for Machine Learning at the University of Adelaide, advised by Prof. Gustavo Carneiro and Prof. Mark Jenkinson.

My research lies at the intersection of computer vision and medical image analysis. I develop weakly supervised and label-efficient deep learning methods, with a current focus on medical radiology and cardiovascular imaging.

Recent publications

  1. 2026
  2. 2026
    MAdam: Metric-Aware Multi-Objective AdamF Liu, R Saluja, S Kwak, R Wang, R Deng, H Kim, JC Paetzold, ...arXiv preprint arXiv:2606.03904
  3. 2026
    HyperCT: Low-Rank Hypernet for Unified Chest CT AnalysisF Liu, S Kwak, H Phung, NB Nizam, I Richter, N Uriel, H Averbuch-Elor, ...MIDL 2026
  4. 2026
    RNED: Rotary Number Encoding and Decoding for Medical VLMsF Liu, S Kwak, N Nizam, I Richter, A Beecy, J Raikhelkar, D Estrin, ...CVPR 2026, 13722-13731
  5. 2026
    From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level GroundingY Liu, Y Ji, A Le, J Zhu, J Pan, C Peng, J Deng, F Liu, J WuCVPR 2026
  6. 2026
    Beyond Machine Interpretation: Learning from Expert Over-Reads Improves ECG DiagnosisS Kwak, F Liu, NB Nizam, I Richter, N Uriel, PM Okin, MR SabuncuMIDL 2026

For the complete and current publication list, see Google Scholar.

Experience

2024–present
Postdoctoral Researcher, Cornell Tech, Cornell University
2020–2023
Ph.D. in Computer Vision and Medical Imaging, University of Adelaide
2019–2020
Honours in Computer Science (First Class), University of Adelaide
2015–2018
Bachelor in Computer Science, University of Adelaide

Academic service

Conferences
CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, BMVC, MICCAI
Journals
IEEE TPAMI