
Yunho Kim
AI Convergence · M.S. Student
rladbsgh797210@gmail.com
I am an M.S. student in the Department of AI Convergence at Gwangju Institute of Science and Technology (GIST), affiliated with the Data Science Lab under Prof. Sundong Kim.
My research interests lie in reinforcement learning and sequential decision-making, with a broader curiosity about how AI systems can be designed to reason in ways that reflect human cognition. I am particularly interested in offline RL settings and in understanding what shapes the quality of learned policies.
Research Interests
Offline Reinforcement Learning
I study how agents learn good sequential decision-making policies from fixed, previously collected data, without further interaction with the environment. I'm particularly interested in what shapes the quality of the policies learned this way.
Human-like Reasoning in AI
I'm broadly curious about how AI systems can be designed to reason in ways that reflect human cognition, beyond optimizing a single objective.
Publications
- On the Role of Proposal Support in Diffusion-Based Offline RL for Sequential Decision-Making ICML 2026 Workshop DEMO · 2026 PDF Poster
- Diffusion-Guided Q-Learning for Offline RL: Adaptive Revaluation for Long-Horizon Decision Making CoRL 2025 Workshop LEAP · 2025 PDF Poster
- 거대언어모델의 추론능력 평가를 위한 MC-LARC 데이터셋 한국소프트웨어종합학술대회 (KSC 2023) · 2023 PDF