Scholar
Sunghwan Kim
Google Scholar ID: gjMbiqIAAAAJ
Yonsei University
Citations & Impact
Citations
176
H-index
6
i10-index
5
Publications
11
Co-authors
5
list available
Publications
Resume (English only)
Academic Achievements
- Published 'Web-Shepherd: Advancing PRMs for Reinforcing Web Agents' at NeuIPS 2025 (Spotlight); 'Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization' at ACL 2025 (Oral); 'Web Agents with World Models: Learning and Leveraging Environment Dynamics in Web Navigation' at ICLR 2025; 'Can Large Language Models be Good Emotional Supporter? Mitigating Preference Bias on Emotional Support Conversation' at ACL 2024, which won the Outstanding Paper Award.
Research Experience
- Worked as a research intern at Microsoft Research Asia (MSRA); involved in multiple research projects such as Web-Shepherd, Embodied Agents Meet Personalization, etc.
Education
- Received B.S. in Materials Science & Engineering from Yonsei University in Aug. 2024; currently a first-year M.S. student at Yonsei University's Language and AGI Lab, advised by Jinyoung Yeo.
Background
- Research interests include reinforcement learning to solve long-horizon tasks, developing intelligent systems that learn through interaction with the environment, and analyzing language models to identify limitations and room for improvement. Aims to build human-like intelligent systems that can autonomously learn, reason, and adapt to diverse environments.
