I am a Postdoctoral Researcher in the Department of AI Convergence at Gwangju Institute of Science and Technology (GIST), Korea. I received my Ph.D. in Computer Science from KAIST under the supervision of Prof. Jae-Gil Lee and Prof. Kyu-Young Whang. My doctoral research focused on spatio-temporal prediction and transfer learning, and my current work centers on advancing human-like reasoning and abstraction in artificial intelligence.
My research aims to develop abstraction and reasoning capabilities toward general and adaptive intelligence, with a particular focus on ARC-AGI, compositional generalization, program synthesis, and reinforcement learning with trajectory modeling. I have published in leading international conferences, including KDD and ICLR, as well as in Q1 SCI-E journals such as ACM Transactions on Intelligent Systems and Technology.
I currently serve as an Area Chair for KDD Datasets and Benchmarks Track and as a Program Committee member for AAAI and KDD. I received the NRF Postdoctoral Fellowship (2024–2026), and was honored with the Excellence Postdoctoral Researcher Award at GIST and selected as a top 10% Outstanding Reviewer at KDD 2025.
Education
- 2009.02 - 2013.08 B.S. in Computer Science, KAIST
- 2013.09 - 2016.08 M.S. in School of Computing, KAIST
- 2017.03 - 2023.02 Ph.D in School of Computing, KAIST
Selected Publications
- ARCTraj: A Dataset and Benchmark of Human Reasoning Trajectories for Abstract Problem Solving , KDD 2026
- TRACED: Transition-aware Regret Approximation with Co-learnability for Environment Design , ICLR 2026
- Reasoning Abilities of Large Language Models: In-Depth Analysis on the Abstraction and Reasoning Corpus , ACM TIST, 2025
- Solution Augmentation for ARC-AGI Problems Using GFlowNet: A Probabilistic Exploration Approach , TMLR, 2025
- Addressing and Visualizing Misalignments in Human Task-Solving Trajectories , KDD 2025