Seoul National University
M.S. in Data Science
Advisor: Hyunwoo Park
Sungho Bae
M.S. Student, Graduate School of Data Science, Seoul National University
Seoul, Republic of Korea
Hello, 안녕하세요 👋
I am a M.S. student at the Graduate School of Data Science, Seoul National University, where I work in the Data Intelligence and Learning (DIAL) Lab under Prof. Hyunwoo Park.
My current interest is agentic AI. In particular, I work on using on-policy self-distillation and on-policy reinforcement learning to make long-horizon, tool-using agents more reliable. I am also interested in generative methods such as diffusion and flow matching, and in how they carry across multiple modalities.
You can reach me at sunghobae@snu.ac.kr, or take a look at my CV.
News
- Jul 2026
- Two papers accepted at ICML 2026 workshops, FAGEN (Failure Modes in Agentic AI) and AI for Science. ✨
- May 2026
- One paper accepted to ACM SIGKDD 2026 (oral presentation). 🔥
- Dec 2025
- One paper accepted at the NeurIPS 2025 Workshop on AI for Science (Best Poster Award). 🏅
- Sep 2024
- Started my M.S. at the Graduate School of Data Science, Seoul National University. 🎓
- Aug 2024
- Received the Outstanding Undergraduate Thesis Award from GIST. 🎖️
Selected Publications
View all publications →Sungho Bae denotes my name · * denotes equal contribution
-
Redefining Maritime Anomaly Detection via Equation-Grounded Synthetic Anomalies
ACM SIGKDD, 2026Oral
-
AISFlow: Boundary-Informed Flow Matching for Long-Term AIS Trajectory Imputation
ICML Workshop on AI for Science, 2026
Under review at IEEE Transactions on Intelligent Transportation Systems (T-ITS)
-
Beyond Tool Selection: Diagnosing Downstream Failures in MCP Tool-Using LLM Agents
ICML Workshop on Failure Modes in Agentic AI (FAGEN), 2026
Education
Gwangju Institute of Science and Technology
B.S. in Computer Science · Minor in AI Convergence
Advisor: Euiseok Hwang
Outstanding Undergraduate Thesis Award
Daegu Science High School
Physics and Chemistry
Research Experience
GIST Intelligent Information System Lab · Research Intern
2023 Winter EECS Department Outstanding Bachelor Intern
- Campus net power load prediction using renewable energy and power consumption data
- Estimation of household characteristics using smart meter data
CJ OliveNetworks AI Research Lab · Research Intern
- Deep learning-based modeling for palm oil price prediction
- Cosmetic demand forecasting using time-series prediction
Honors & Awards
Outstanding Undergraduate Thesis Award · GIST
A Study on Efficient Feature Selection for Smart Meter Data Analysis
Grand Prize (1st) · 2024 Hello World Hackathon
Platform supporting active seniors in staying socially engaged
Chairman Award (2nd) · INNOPOLIS Foundation AI4GOOD Hackathon
iPhone Shortcut shopping assistant for people with visual impairments
1st Prize · CJ OliveNetworks CTO Cup Paper Competition
Time-series forecasting for palm oil prices · NeurIPS 2023 participation opportunity
1st Prize · CJ OliveNetworks CTO Cup Kaggle Competition
Store sales forecasting · Internship opportunity at CJ OliveNetworks
AWS Head of APJ STJ Prize (2nd) · GDSC × GIST Hackathon
Book-based community iOS app for neighbors
Other Experience
Teaching Assistant
Data Engineering @ GIST
Data Visualization and Me @ Seoul National University
SOPT · iOS Developer
Developed the KAERA iOS application with Team HARA
Republic of Korea Army
Completed compulsory military service
GIST HOUSE · Club President
Led events and community initiatives for dormitory residents