Seongsu Bae

About

Seongsu Bae

Ph.D. Candidate @ KAIST AI

I'm interested in semantic interfaces — AI that expands what humans can understand and do, not just what it can automate. I build them for healthcare and study how to rigorously measure them. Advised by Edward Choi.

photo

Seoul

photo

Tokyo

photo

NeurIPS

photo

Jeju

photo

Beijing

photo

Lab

Selected Publications

All 14

ListenCare: Encounter-Grounded Audio Question Answering for Long-Form Clinical Conversation Speech

Seongsu Bae*, Chaeeun Shim*, Sungbae Park, Edward Choi

ICML 2026 Workshop on Machine Learning for Audio

KorMedMCQA-V: A Multimodal Benchmark for Evaluating Vision-Language Models on Korean Medical Licensing Exam

Byungjin Choi*, Seongsu Bae*, Sunjun Kweon, Edward Choi

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

Graph-Text Multi-Modal Pre-training for Medical Representation Learning

Sungjin Park, Seongsu Bae, Jiho Kim, Tackeun Kim, Edward Choi

CHIL 2022

Question Answering for Complex Electronic Health Records Database using Unified Encoder-Decoder Architecture

Seongsu Bae, Daeyoung Kim, Jiho Kim, Edward Choi

ML4H 2021 (Oral Spotlight)

Experience

Microsoft Research Asia

Research Intern · Mentors: Eric Chang and Lei Ji

Worked on text-to-image generation and multimodal question answering.

Oct 2022 – Apr 2023

Education

KAIST AI

Ph.D. Candidate · Advisor: Edward Choi

Sep 2022 – Present

KAIST AI

M.S. · Advisor: Edward Choi

Sep 2020 – Aug 2022

Talks

W&B KoreaEvaluation & Benchmark

2025

Stanford MedAIEHRXQAvideo

2024

Microsoft ResearchEHRXQA

2023