|
Jiwon Jeon I'm a Ph.D. student in the School of Electrical Engineering at KAIST, where I work with Prof. YoungChul Sung in the Smart Information Systems Research Lab (SISReL). My research focuses on LLM post-training and multi-agent reinforcement learning, and my recent interest is in LLM harnessing. I received my B.S. and M.S. degrees in Electrical Engineering from KAIST, and was a research intern at LG AI Research (2023–2024). |
|
ResearchI'm interested in LLM reasoning (self-correction, self-distillation, RLVR) and multi-agent reinforcement learning. Recently, I'm interested in LLM harnessing. (* denotes equal contribution) |
Large Language Models |
|
Rebellious Student: Reversing Teacher Signals for Reasoning Exploration with Self-Distilled RLVR
Jeonghye Kim*, Jiwon Jeon*, Dongsheng Li, Yuqing Yang NeurIPS, 2026 · COLM Workshop, 2026 (Oral Presentation) paper / bibtex@article{kim2026rebellious,
title={Rebellious student: Reversing teacher signals for reasoning exploration with self-distilled rlvr},
author={Kim, Jeonghye and Jeon, Jiwon and Li, Dongsheng and Yang, Yuqing},
journal={arXiv preprint arXiv:2605.10781},
year={2026}
} |
|
Inject a Turn, Not a Token: Learning Where to Redirect LLM Reasoning for Self-Correction
Jiwon Jeon*, Myungsik Cho*, Jeonghye Kim, Youngchul Sung ICLR, 2027 (submitted) · COLM Workshop, 2026 (Spotlight) bibtex@inproceedings{jeon2026inject,
title={Inject a Turn, Not a Token: Learning Where to Redirect LLM Reasoning for Self-Correction},
author={Jeon, Jiwon and Cho, Myungsik and Kim, Jeonghye and Sung, Youngchul},
booktitle={COLM 2026 Workshop on Efficient Reasoning}
} |
|
Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
Jeonghye Kim, Xufang Luo, Minbeom Kim, Sangmook Lee, Dohyung Kim, Jiwon Jeon, Dongsheng Li, Yuqing Yang COLM, 2026 paper / bibtex@article{kim2026does,
title={Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?},
author={Kim, Jeonghye and Luo, Xufang and Kim, Minbeom and Lee, Sangmook and Kim, Dohyung and Jeon, Jiwon and Li, Dongsheng and Yang, Yuqing},
journal={arXiv preprint arXiv:2603.24472},
year={2026}
} |
|
Be My Tutor: On-Policy Co-Distillation for Mutual LLM Improvement via Peer Feedback
Woohyeon Byeon, Jiwon Jeon, Jeonghye Kim, Youngchul Sung NAACL, 2027 (submitted) paper / bibtex@article{byeon2026my,
title={Be My Tutor: On-Policy Co-Distillation for Mutual LLM Improvement via Peer Feedback},
author={Byeon, Woohyeon and Jeon, Jiwon and Kim, Jeonghye and Sung, Youngchul},
journal={arXiv preprint arXiv:2606.14368},
year={2026}
} |
Reinforcement Learning |
|
MASTARS: Multi-Agent Sequential Trajectory Augmentation with Return-Conditioned Subgoals
Jiwon Jeon, Myungsik Cho, Woojun Kim, Seongmin Kim, Woohyeon Byeon, Seungyul Han, Youngchul Sung NeurIPS, 2026 bibtex@inproceedings{jeon2026mastars,
author = {Jeon, Jiwon and Cho, Myungsik and Kim, Woojun and Kim, Seongmin and Byeon, Woohyeon and Han, Seungyul and Sung, Youngchul},
year = {2026},
title = {MASTARS: Multi-Agent Sequential Trajectory Augmentation with Return-Conditioned Subgoals},
booktitle = {Advances in Neural Information Processing Systems (NeurIPS)}
} |
|
AGiR: Mitigating Gift Over-Reliance in Mixed-Motive Games
Woohyeon Byeon, Seongmin Kim, Jiwon Jeon, Woojun Kim, Youngchul Sung NeurIPS, 2026 bibtex@inproceedings{byeon2026agir,
author = {Byeon, Woohyeon and Kim, Seongmin and Jeon, Jiwon and Kim, Woojun and Sung, Youngchul},
year = {2026},
title = {AGiR: Mitigating Gift Over-Reliance in Mixed-Motive Games},
booktitle = {Advances in Neural Information Processing Systems (NeurIPS)}
} |
|
Addressing Exogenous Variability in Cooperative Multi-Agent Reinforcement Learning
Seongmin Kim, Woohyeon Byeon, Jiwon Jeon, Seungyul Han, Youngchul Sung NeurIPS, 2026 (Spotlight) bibtex@inproceedings{kim2026exogenous,
author = {Kim, Seongmin and Byeon, Woohyeon and Jeon, Jiwon and Han, Seungyul and Sung, Youngchul},
year = {2026},
title = {Addressing Exogenous Variability in Cooperative Multi-Agent Reinforcement Learning},
booktitle = {Advances in Neural Information Processing Systems (NeurIPS)}
} |
|
STAIRS-Former: Spatio-Temporal Attention with Interleaved Recursive Structure Transformer for Offline Multi-task Multi-agent Reinforcement Learning
Jiwon Jeon*, Myungsik Cho*, Youngchul Sung ICLR, 2026 paper / bibtex@article{jeon2026stairs,
title={STAIRS-former: Spatio-temporal attention with interleaved recursive structure transformer for offline multi-task multi-agent reinforcement learning},
author={Jeon, Jiwon and Cho, Myungsik and Sung, Youngchul},
journal={arXiv preprint arXiv:2603.11691},
year={2026}
} |
|
Generalized Per-Agent Advantage Estimation for Multi-Agent Policy Optimization
Seongmin Kim, Giseung Park, Woojun Kim, Jiwon Jeon, Seungyul Han, Youngchul Sung AAMAS, 2026 (Best Paper Award Nominee) paper / bibtex@article{kim2026generalized,
title={Generalized Per-Agent Advantage Estimation for Multi-Agent Policy Optimization},
author={Kim, Seongmin and Park, Giseung and Kim, Woojun and Jeon, Jiwon and Han, Seungyul and Sung, Youngchul},
journal={arXiv preprint arXiv:2603.02654},
year={2026}
} |
|
MASER: Multi-Agent Reinforcement Learning with Subgoals Generated from Experience Replay Buffer
Jeewon Jeon, Woojun Kim, Whiyoung Jung, Youngchul Sung ICML, 2022 paper / bibtex@inproceedings{jeon2022maser,
title={Maser: Multi-agent reinforcement learning with subgoals generated from experience replay buffer},
author={Jeon, Jeewon and Kim, Woojun and Jung, Whiyoung and Sung, Youngchul},
booktitle={International conference on machine learning},
pages={10041--10052},
year={2022},
organization={PMLR}
} |
Experience
Research Intern, LG AI Research — Jul. 2023–Mar. 2024 Academic ServiceSilver Reviewer, ICML 2026 |