Seungjoon Lee
sjlee1218 at postech dot ac dot kr
I am a Ph.D. candidate at POSTECH ML Lab, South Korea, fortunate to be advised by Prof. Jungseul Ok. My goal is to democratize AI by developing cost-efficient yet high-performing algorithms that make advanced technologies accessible to a broader community. My research focuses on agentic and embodied AI, with an emphasis on building powerful agents using small, cost-efficient models.
Previously, I completed my B.S.E. in Computer Science and Engineering at POSTECH.
Selected Publications
-
Towards Faithful Agentic XAI: A Verification Method and an Open-World Benchmark for Better Model Faithfulness
arXiv, 2026 -
ProCompNav: Proactive Instance Navigation with Comparative Judgment for Ambiguous User Queries
NeurIPS 2026 (oral) -
Experience-based Knowledge Correction for Robust Planning in Minecraft
ICLR 2026 -
Active Preference-based Learning for Multi-dimensional Personalization
ACL 2025 Main (long, oral)