Publications

Reinforcement Learning

Large Language Model (LLM) Agents and Reasoning

• J. Jeon, M. Cho, J. Kim, and Y. Sung*, "Inject a turn, not a token: Learning where to redirect LLM reasoning for self-correction," to be presented at COLM 2026 Workshop on Efficient Reasoning, 2026

• Woohyeon Byeon, Jiwon Jeon, Jeonghye Kim, Youngchul Sung*, "Be my tutor: On-policy co-distillation for mutual LLM improvement via peer feedback," preprint, Jun. 2026

• Seohui Bae, Jeonghye Kim, Youngchul Sung and Woohyung Lim*, "Align while search: Belief-guided exploratory inference for test-time world alignment," IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2026

• Jeonghye Kim+, Sojeong Rhee+, Minbeom Kim, Dohyung Kim, Sangmook Lee, Youngchul Sung* and Kyomin Jung*, "ReflAct: World-grounded decision making in LLM agents via goal-state reflection," Conference of Emplical Methods for Natural Language Processing (EMNLP). (+Equal Contribution), Nov. 2025

• Seohui Bae, Jeonghye Kim, Youngchul Sung and Woohyung Lim*, "Align while search: Belief-guided exploratory inference for test-time world alignment," the Exploration in AI Today (EXAIT) Workshop at ICML 2025, Vancouver, July 2025

Deep Reinforcement Learning

Offline Reinforcement Learning and Imitation Learning

• Jongseong Chae, Jongeui Park, Yongjae Shin, Gyeongmin Kim, Seungyul Han, and Youngchul Sung*,  "Flow Actor-Critic for Offline Reinforcement Learning," International Conference on Learning Representations  (ICLR) , Brazil, Apr. 2026

• Yongjae Shin, Jongseong Chae, Jongeui Park, and Youngchul Sung*, "Flow Matching with Injected Noise for Offline-to-Online Reinforcement Learning," International Conference on Learning Representations  (ICLR) , Brazil, Apr. 2026 

• Jeonghye Kim, Yongjae Shin, Whiyoung Jung, Sunghoon Hong, Deunsol Yoon, Youngchul Sung, Kanghoon Lee, and Woohyung Lim*,  "Penalizing infeasible actions and reward scaling in reinforcement learning with offline data," International Conference on Machine Learning (ICML), Vancouver, Canada, Jul. 2025 (Spotlight Paper, Top 2.6%)

• Yongjae Shin, Jeonghye Kim, Whiyoung Jung, Sunghoon Hong, Deunsol Yoon, Youngsoo Jang, Geonhyeong Kim, Jongseong Chae, Youngchul Sung, Kanghoon Lee, and Woohyung Lim*, "Online pre-training for offline-to-online reinforcement learning," International Conference on Machine Learning (ICML), Vancouver, Canada, Jul. 2025

• Jeonghye Kim, Suyoung Lee, Woojun Kim, and Youngchul Sung*, "Adaptive Q-aid for conditional supervised learning in offline reinforcement learning," Conference on Neural Information Processing Systems  (NeurIPS ), 2024

• Jeonghyee Kim, Suyoung Lee, Woojun Kim*, and Youngchul Sung*, “Decision convformer: Local filtering in metaformer is sufficient for decision making,” International Conference on Learning Representations  (ICLR) , May 2024 (Spotlight Paper, Top 5%)

• Sungho Choi, Seungyul Han*, Woojun Kim, Jongseong Chae, Whiyoung Jung, and Youngchul Sung, "Domain adaptive imitation learning with visual observation," Conference on Neural Information Processing Systems  (NeurIPS ), New Orleans, LA, Dec. 2023

• Jongseong Chae, Seungyul Han*, Whiyoung Jung, Myungsik Cho, Sungho Choi, and Youngchul Sung, “Robust imitation learning against variations in environment dynamics,” International Conference on Machine Learning (ICML)  2022 , Baltimore, MD, USA, Jul. 2022

Exploration and Sample Efficiency

• Giseung Park, Whiyoung Jung, Seungyul Han, Sungho Choi and Youngchul Sung*, "Adaptive multi-model fusion learning for sparse-reward reinforcement learning," Neurocomputing, vol. 633, Jun. 2025

• Woojun Kim+, Jeonghye Kim+ and Youngchul Sung*, "LESSON: Learning to integrate exploration strategies for reinforcement learning via an option framework," International Conference on Machine Learning (ICML), Hawaii, HI, Jul. 2023 (+Equal contribution)

• Seungyul Han and Youngchul Sung*, "A max-min entropy framework for reinforcement learning," Conference on Neural Information Processing Systems (NeurIPS) ,  Dec. 2021. 

• Seungyul Han and Youngchul Sung*, "Diversity actor-critic: Sample-aware entropy regularization for sample-efficient exploration," International Conference on Machine Learning (ICML) , Jul. 2021  

• Whiyoung Jung, Giseung Park, and Youngchul Sung*, "Population-Guided Parallel Policy Search for Reinforcement Learning," International Conference on Learning Representations (ICLR) , Vienna, Austria, May 2020

• Seungyul Han and Youngchul Sung*, "Dimension-Wise Importance Sampling Weight Clipping for Sample-Efficient Reinforcement Learning," International Conference on Machine Learning (ICML) , Long Beach, CA, USA, Jun. 2019

Safe Reinforcement Learning

• Woojun Kim+, Yongjae Shin+, Jongeui Park, and Youngchul Sung*, "Sample-efficient and safe deep reinforcement learning via reset deep ensemble agents," Conference on Neural Information Processing Systems  (NeurIPS ), New Orleans, LA, Dec. 2023 (+Equal contribution)

• W. Jung, M. Cho, J. Park and Y. Sung*, "Quantile constrained reinforcement learning: A reinforcement learning framework constraining outage probability," Conference on Neural Information Processing Systems (NeurIPS),  Dec. 2022

Partially-Observable MDPs

• G. Kim, J. Kim, S. Lee, J. Baek, H. Moon, S. Shin and Y. Sung*, "Robust reinforcement learning under dimension-wise state information drop," IEEE Access, 2024

• Giseung Park, Sungho Choi and Youngchul Sung*, "Blockwise sequential model learning for partially observable reinforcement learning," the 36rd AAAI Conference on Artificial Intelligence (AAAI ), Feb.  2022 (Oral Presentation, Top 4.6%)

Multi-Task RL, Meta RL, and Multi-Objective RL

•  G. Park+, H. Nam+, W. Byeon, A. Leshem and Y. Sung*, “Constrained multi-objective reinforcement learning with max-min criterion,” International Conference on Machine Learning (ICML),  Seoul, Korea, Jul. 2026  (+ Equal contribution)

• Woohyeon Byeon, Giseung Park, Jongseong Chae, Amir Leshem and Youngchul Sung, "Multi-objective reinforcement learning with max-min criterion: A game-theoretic approach," Conference on Neural Information Processing Systems  (NeurIPS ) , Dec. 2025

• Myungsik Cho, Jongeui Park, Jeonghye Kim and Youngchul Sung*, "ARS: Adaptive reward scaling for multi-task reinforcement learning," International Conference on Machine Learning (ICML), Jul. 2025 (Solved the MT10 task of the Metaworld Environment.)

• Giseung Park and Youngchul Sung*, "Reward dimension reduction for scalable multi-objective reinforcement learning," International Conference on Learning Representations  (ICLR) , Singapore, Apr. 2025 

• Myungsik Cho, Jongeui Park, Suyoung Lee and Youngchul Sung*, "Hard task first: Multi-task reinforcement learning through task scheduling," International Conference on Machine Learning (ICML) 2024

• Giseung Park, Woohyeon Byeon, Seongmin Kim, Elad Habakuk, Amir Leshem and Youngchul Sung*, "The max-min formulation of multi-objective reinforcement learning: From theory to a model-free algorithm," International Conference on Machine Learning (ICML) 2024

• Suyoung Lee, Myungsik Cho, and Youngchul Sung*, "Parameterizing Non-Parametric Meta-Reinforcement Learning Tasks via Subtask Decomposition, Conference on Neural Information Processing Systems  (NeurIPS ), New Orleans, LA, Dec. 2023

• Myungsik Cho, Whiyoung Jung, and Youngchul Sung*, "Multi-task reinforcement learning with task representation method," ICLR Workshop on Generalizable Policy Learning in Physical World (GPL),  Apr. 2022

Multi-Agent Reinforcement Learning

• Seongmin Kim, Giseung Park, Woojun Kim, Jeewon Jeon, Seungyul Han*, and Youngchul Sung, "Generalized Per-Agent Advantage Estimation for Multi-Agent Policy Optimization," International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2026 (AAMAS '26 Best Paper Award Nominee)

• Jeewon Jeon, Myungsik Cho, Youngchul Sung, "STAIRS-Former: Spatio-Temporal Attention with Interleaved Recursive Structure TransFormer for Offline Mulit-task Multi-agent Reinforcement Learning," International Conference on Learning Representations  (ICLR) , Brazil, Apr. 2026 

• Woojun Kim and Youngchul Sung*, "An adaptive entropy regularization framework for multi-agent reinforcement learning," International Conference on Machine Learning (ICML), Hawaii, HI, Jul. 2023

• Woojun Kim, Whiyoung Jung, Myungsik Cho, and Youngchul Sung*, “A variational approach to mutual information-based coordination for multi-agent reinforcement learning,” International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2023

• Woojun Kim and Youngchul Sung, “Parameter sharing with network pruning for scalable multi-agent deep reinforcement learning,” International Conference on Autonomous Agents and Multiagent Systems (AAMAS), May 2023

•  Jeewon Jeon, Woojun Kim*, Whiyoung Jung and Youngchul Sung, “MASER: Multi-Agent reinforcementlearning with Subgoals generated from Experience Replay buffer,” International Conference on Machine Learning (ICML)  2022 , Baltimore, MD, USA, Jul. 2022

• Woonjun Kim, Jongeui Park, and Youngchul Sung*, "Communication in multi-agent reinforcement learning: Intention sharing," International Conference on Learning Representations (ICLR) 2021, May 2021

• Woojun Kim, Myungsik Cho, and Youngchul Sung*, "Message-dropout: An efficient training method for multi-agent deep reinforcement learning," the 33rd AAAI Conference on Artificial Intelligence (AAAI ) 2019, Honolulu, HW, USA, Jan. 2019 

Applications

• G. Park+, H. Nam+, W. Byeon, A. Leshem and Y. Sung*, "Fairness-aware resource allocation in edge computing via multi-objective reinforcement learning," submitted to IEEE Trans. Network and Service Management, Mar. 2026

• Sohee Bae, Seungyul Han, and Youngchul Sung*, "A reinforcement learning formulation of the Lyapunov optimization: Application to edge computing systems with queue stability," available at arXiv, Dec. 2020

Click Here for Past Works on Communications