Jiwon Jeon

Ph.D Student

Korea Advanced Institute of Science and Technology (KAIST)

Smart Information Systems Research Lab (SISReL)

Office: Room #619, N1, KAIST

About myself

I received a B.S. degree and M.S degree in the Department of Electrical Engineering from KAIST, Daejeon, Korea in 2020 and 2022. I am currently a Ph.D. student working with Prof. YoungChul Sung in Smart Information Systems Research Lab (SISReL). My research interests span Multi-Agent Reinforcement Learning, Deep Learning, and Reinforcement Learning, and I have recently started working on LLM reasoning.

Education

Ph.D Course (02.2022 - Present)

Korea Advanced Institute of Science and Technology (KAIST)
School of Electrical Engineering
- Advisor: Prof. YoungChul Sung

M.S. (02.2020 - 02.2022)

Korea Advanced Institute of Science and Technology (KAIST)
School of Electrical Engineering
- Advisor: Prof. YoungChul Sung

B.S. (02.2015 - 02.2020)

Korea Advanced Institute of Science and Technology (KAIST)
School of Electrical Engineering
Exchange student at Technical University of Denmark (DTU) (08.2018 - 02.2019)

Work Experience

Research Intern (07.2023 - 03.2024)

LG AI Research

Publications (Reinforcement Learning)

MASER: Multi-agent reinforcement learning with subgoals generated from experience replay buffer (ICML 2022, poster)

- Jeewon Jeon, Woojun Kim, Whiyoung Jung, and Youngchul Sung

- STAIRS-Former: Spatio-Temporal Atention with Interleaved Recursive Structure Transformer for Offline Multi-task Multi-agent Reinforcement Learning (ICLR 2026, poster)

Jiwon Jeon*, Myungsik Cho*, Youngchul Sung (* Equal contribution)

Generalized Per-Agent Advantage Estimation for Multi-Agent Policy Optimization (AAMAS 2026, Best Paper Award Nominee)
- Seongmin Kim, Giseung Park, Woojun Kim, Jiwon Jeon, Seungyul Han, Youngchul Sung

MASTARS: Multi-Agent Sequential Trajectory Augmentation with Return-Conditioned Subgoals (Neurips 2026, under review)
- Jiwon Jeon, Myungsik Cho, Woojun Kim, Seongmin Kim, Woohyeon Byeon, Seungyul Han, Youngchul Sung

AGiR: Mitigating Gift Over-Reliance in Mixed-Motive Games (Neurips 2026, under review)
- Woohyeon Byeon, Seongmin Kim, Jiwon Jeon, Woojun Kim, Youngchul Sung

Addressing Exogeneous Variability in Cooperative Multi-Agent Reinforcement Learning (Neurips 2026, under review)
- Seongmin Kim, Woohyeon Byeon, Jiwon Jeon, Seungyul Han, Youngchul Sung

Publications (Large-Language-Models)

Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs? (COLM 2026, under review)
- Jeonghye Kim, Xufang Luo, Minbeom Kim, Sangmook Lee, Dohyung Kim, Jiwon Jeon, Dongsheng Li, Yuqing Yang

Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR (Neurips 2026, under review)
- Jeonghye Kim*, Jiwon Jeon*, Dongsheng Li, Yuqing Yang (* Equal contribution)

Academic Service

Awards & Honors
- Silver Reviewer, ICML 2026