Daeun Kyung

Daeun Kyung

Ph.D. Student · KAIST AI

About

I am a Ph.D. student at KAIST AI, advised by Prof. Edward Choi. My research focuses on large language models (LLMs) and multimodal learning, particularly in healthcare applications. I am interested in building and evaluating AI systems, from curating high-quality datasets to developing models that work reliably in real-world settings. Most recently, I have been exploring how LLMs can enable more realistic and grounded clinical interactions. As part of this effort, I developed PatientSim, a persona-driven framework for simulating realistic doctor-patient conversations grounded in real medical records.

Large Language Models Conversational Agents Multimodal Learning Generative Models Healthcare AI

News

2026.03 Released a new demo for PatientSim.
2026.01 Papers accepted at ICLR 2026 (ProPerSim), ICLR 2026 Workshop (DialSim), and ICASSP 2026 (ECG-Agent).
2025.09 PatientSim accepted as Spotlight at NeurIPS 2025 Datasets and Benchmarks.
2025.01 Paper on EHR-conditional CXR prediction accepted at CHIL 2025.
2024.09 EHRCon accepted as Spotlight at NeurIPS 2024 Datasets and Benchmarks.
2023.06 Received Best Student Paper Award at ICASSP 2023 for PerX2CT.

Publications

* denotes equal contribution

NeurIPS
PatientSim: A Persona-Driven Simulator for Realistic Doctor-Patient Interactions

Daeun Kyung, Hyunseung Chung, Seongsu Bae, Jiho Kim, Jae Ho Sohn, Taerim Kim, Soo Kyung Kim, Edward Choi

NeurIPS 2025 Datasets and Benchmarks Spotlight

CHIL
Towards Predicting Temporal Changes in a Patient's Chest X-ray Images based on Electronic Health Records

Daeun Kyung, Junu Kim, Tackeun Kim, Edward Choi

Proc. of Conference on Health, Inference, and Learning (CHIL) 2025

Scientific Reports
Significantly improving zero-shot X-ray pathology classification via fine-tuning pre-trained image-text encoders

Jongseong Jang*, Daeun Kyung*, Seung Hwan Kim, Honglak Lee, Kyunghoon Bae, Edward Choi

Scientific Reports, 14, 23199 (2024)

NeurIPS
EHRXQA: A Multi-Modal Question Answering Dataset for Electronic Health Records with Chest X-ray Images

Seongsu Bae*, Daeun Kyung*, Jaehee Ryu, Eunbyeol Cho, Gyubok Lee, Sunjun Kweon, Jungwoo Oh, Lei Ji, Eric I-Chao Chang, Tackeun Kim, Edward Choi

NeurIPS 2023 Datasets and Benchmarks

ICASSP
Perspective Projection-Based 3D CT Reconstruction from Biplanar X-rays

Daeun Kyung*, Kyungmin Jo*, Jaegul Choo, Joonseok Lee, Edward Choi

ICASSP 2023 (Oral) Best Student Paper Award

2026

ICLR Workshop
DialSim: A Real-Time Simulator for Evaluating Long-Term Dialogue Understanding of Conversational Agents

Jiho Kim, Woosog Chay, Hyeonji Hwang, Daeun Kyung, Hyunseung Chung, Eunbyeol Cho, Yohan Jo, Edward Choi

Workshop on Memory for LLM-Based Agentic Systems @ ICLR 2026

ICLR
ProPerSim: Developing Proactive and Personalized AI Assistants through User-Assistant Simulation

Jiho Kim, Junseong Choi, Woosog Chay, Daeun Kyung, Yeonsu Kwon, Yohan Jo, Edward Choi

Proc. of International Conference on Learning Representations (ICLR) 2026

ICASSP
ECG-Agent: On-Device Tool-Calling Agent for ECG Multi-Turn Dialogue

Hyunseung Chung, Jungwoo Oh, Daeun Kyung, Jiho Kim, Yeonsu Kwon, Min-Gyu Kim, Edward Choi

Proc. of ICASSP 2026

2025

NeurIPS
PatientSim: A Persona-Driven Simulator for Realistic Doctor-Patient Interactions

Daeun Kyung, Hyunseung Chung, Seongsu Bae, Jiho Kim, Jae Ho Sohn, Taerim Kim, Soo Kyung Kim, Edward Choi

NeurIPS 2025 Datasets and Benchmarks Spotlight

CHIL
Towards Predicting Temporal Changes in a Patient's Chest X-ray Images based on Electronic Health Records

Daeun Kyung, Junu Kim, Tackeun Kim, Edward Choi

Proc. of Conference on Health, Inference, and Learning (CHIL) 2025

2024

NeurIPS
EHRCon: Dataset for Checking Consistency between Unstructured Notes and Structured Tables in Electronic Health Records

Yeonsu Kwon*, Jiho Kim*, Gyubok Lee, Seongsu Bae, Daeun Kyung, Wonchul Cha, Tom Pollard, Alistair Johnson, Edward Choi

NeurIPS 2024 Datasets and Benchmarks Spotlight

Scientific Reports
Significantly improving zero-shot X-ray pathology classification via fine-tuning pre-trained image-text encoders

Jongseong Jang*, Daeun Kyung*, Seung Hwan Kim, Honglak Lee, Kyunghoon Bae, Edward Choi

Scientific Reports, 14, 23199 (2024)

2023

NeurIPS
EHRXQA: A Multi-Modal Question Answering Dataset for Electronic Health Records with Chest X-ray Images

Seongsu Bae*, Daeun Kyung*, Jaehee Ryu, Eunbyeol Cho, Gyubok Lee, Sunjun Kweon, Jungwoo Oh, Lei Ji, Eric I-Chao Chang, Tackeun Kim, Edward Choi

NeurIPS 2023 Datasets and Benchmarks

ICASSP
Perspective Projection-Based 3D CT Reconstruction from Biplanar X-rays

Daeun Kyung*, Kyungmin Jo*, Jaegul Choo, Joonseok Lee, Edward Choi

ICASSP 2023 (Oral) Best Student Paper Award

IEEE Access
Unsupervised Domain Adaptation for 3D Point Clouds by Searched Transformations

Dongmin Kang, Yeongwoo Nam, Daeun Kyung, Jonghyun Choi

IEEE Access, Vol. 10, pp. 56901-56913

Experience

Oct 2022 — Apr 2023

LG AI Research

Research Intern · Seoul, Korea

Developed a multimodal image-text framework for chest X-ray/report alignment, introducing a fine-tuning strategy with random sentence sampling and loss relaxation to improve zero-shot pathology classification. Achieved a 4.3% average increase in macro AUROC across four datasets, surpassing board-certified radiologists. Published in Scientific Reports.

Jan 2020 — Aug 2021

GIST Computer Vision Lab

Undergraduate Research Intern · Advisor: Jonghyun Choi

Conducted research on 3D point cloud domain adaptation. Proposed a search-based transformation approach for unsupervised domain adaptation, published in IEEE Access (2022).

Education

Aug 2021 — Present (Expected Aug 2026)

KAIST

Ph.D. in Artificial Intelligence (Integrated Master's/Doctoral)

Kim Jaechul Graduate School of AI · Advisor: Prof. Edward Choi

Mar 2017 — Aug 2021

GIST

B.S. in Electrical Engineering and Computer Science

Gwangju Institute of Science and Technology

Awards

Best Student Paper Award

ICASSP 2023

EECS Outstanding Bachelor's Research Award

GIST · Nov 2020

Services

Reviewer: ICLR (2025–2026), ICASSP (2023–2026), ML4H (2024–2025), CHIL (2024–2025), COLM (2025), ARR (2025 Feb), NeurIPS Datasets & Benchmarks (2024)

Teaching: KoSAIM Summer School Hands-on Session — Numpy and Pytorch Intro (Aug 2023)

Get in Touch

I'm always happy to discuss research collaborations, healthcare AI, or anything in between.