Kyungmin Jeon

M.S Student

Seoul National University & Seoul National University Hospital, South Korea

Research Topics

Ultrasound, Vision-Language Models, AI in Radiology

News!

[Aug. 2026] Our paper "Reformulating Radiology Report Error Detection as Sentence-Level Vision-Language Learning" has been accepted to EMNLP 2026 Findings.
[Feb. 2026] Our paper "Temporal Inversion for Learning Interval Change in Chest X-Rays" has been accepted to CVPR 2026.
[Sep. 2025] Started Master's degree at iRAIL in Seoul National University.
[Jul. 2025] Our abstract "Evaluating Generative Models for Open-Ended Medical Diagnosis in Realistic Clinical Scenarios" has been accepted to AMAI workshop on MICCAI 2025.
[Oct. 2024] Our paper "Harnessing Generative AI for Comprehensive Evaluation of Medical Imaging AI" has been accepted to GenAI4Health workshop on NeurIPS 2024.
[Sep. 2024] Our abstract "Evaluating Open and Closed-Source Language and Vision-Language Models for Multicenter Image-Based Diagnosis in Radiology: A Comparative Study with Human Performance" has been accepted to RSNA 2024.
[Oct. 2023] Our paper "Lesion in-and-out painting for medical image augmentation" has been accepted to DGM4H workshop on NeurIPS 2023.
[Dec. 2022] Our team ranked 5th place in the CDM AI Challenge ver.2 held by SNUH MAIC.
[Dec. 2022] Our team ranked 3rd place in the Sleep AI Challenge ver.3 held by SNUH MAIC.
[Nov. 2022] Our team ranked 2nd place in the Medical Synthetic Data AI Challenge held by Urban Corporation.
[Jun. 2022] Started intership at innovative Radiology Artificial Intelligence Lab (iRAIL), Seoul National University.

Ongoing Projects

Evaluation of Diagnostic Capability of VLMs (Sep. 2025 - Present)

Built a multi-modal evaluation framework to benchmark open-weight vision–language models on clinical diagnosis tasks.

Artifact Classification in Lung Ultrasound (Jan. 2026 - Present)

Developed a deep learning pipeline for lung ultrasound classification aimed at clinical deployment.

Automated Breast Ultrasound Foundation Models (Apr. 2026 - Present)

Designed a 3D vision encoder to learn transferable representations from automated breast ultrasound volumes.

Educations

M.S. in Seoul National University

Interdisciplinary Program in Bioengineering & Integrated Major in Innovative Medical Science
Supervised by Prof. Chang Min Park & Co-advised by Prof. Dongheon Lee
Sep. 2025 - Present
GPA: 4.2 / 4.3

B.S. in Dankook University

Publications

International Journal

International Conference

Dabin Min, Kyungmin Jeon, Dongheon Lee†, Chang Min Park†

Empirical Methods in Natural Language Processing (EMNLP) Findings, 2026

Hanbin Ko, Kyungmin Jeon, Doowoong Choi, Chang Min Park

Computer Vision and Pattern Recognition (CVPR), 2026

International Workshop

Kyungmin Jeon, Gihun Cho, Dabin Min, Jiyoung Lee, Donguk Kim, Chang Min Park.

Applications of Medical AI (AMAI) on MICCAI 2025 (Oral Presentation)

Yisak Kim, Seunghyun Jang, Soyeon Kim, Kyungmin Jeon, Chang Min Park

GenAI for Health Potential, Trust, and Policy Compliance (GenAI4Health) on NeurIPS, 2024

Yisak Kim, Kyungmin Jeon, Soyeon Kim, Chang Min Park

Deep Generative Models for Health (DGM4H) on NeurIPS, 2023

Academic Services

Workshop Reviewer
- - GenAI for Health Potential, Trust, and Policy Compliance (GenAI4Health) Workshop on NeurIPS 2025
  - GenAI for Health Agentic Systems, Clinical Trust, and Future Potential (GenAI4Health) Workshop on NeurIPS 2026

Page updated

Google Sites

Report abuse