Profile photo of Hyukkyu Kang

Hyukkyu Kang

I am a Ph.D. candidate in the Data Systems Lab at Pohang University of Science and Technology (POSTECH), advised by Professor Wook-Shin Han. My research interest lies in the field of Natural Language Processing (NLP) and Information Retrieval (IR), particularly in developing foundation models with retrieval. I have hands-on experience in pre-training language models and fine-tuning them for downstream tasks such as text retrieval and semantic parsing.

News

Aug. 2025
One paper was accepted to EMNLP 2025 Main Conference (Oral Presentation)
Jan. 2025
Visiting Scholar at Global AI Frontier Lab, New York University

Education

Ph.D. in Computer Science and Engineering (M.S./Ph.D. integrated)

Pohang University of Science and Technology (POSTECH)

B.S. in Computer Science and Engineering

Pohang University of Science and Technology (POSTECH)

Work Experience

Visiting Scholar

Global AI Frontier Lab, New York University

  • Collaborated with Prof. Kyunghyun Cho on "Disentangling Reasoning and Knowledge in Retrieval-Augmented Language Models."

Research Intern

Database Lab, POSTECH

  • Researched Natural Language Interface to Databases (NLIDB).

Research Intern

Center of Netmarble AI Revolution, Netmarble Company

  • Developed AI player for a real-time strategy game using reinforcement learning.

Publications

Hyukkyu Kang, Injung Kim, Wook-Shin Han

TRIAL: Token Relations and Importance Aware Late-interaction for Accurate Text Retrieval

Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP) 2025, Main Conference (Oral Presentation)

Projects

AI Guardians: Development of Robust, Controllable, and Unbiased Trustworthy AI Technology

IITP (Korea government MSIT), Global AI Frontier Lab

  • Developing RAQUEL: Robust Validation of Machine Unlearning through Query Executions over an Aligned Database.

Knowledge Graph Construction for Knowledge Inference

Samsung SDS

  • Developed an end-to-end system to construct knowledge graphs from raw document corpora, structuring domain knowledge for downstream applications.

Development of a Conversational, Self-tuning DBMS

IITP (Korea government MSIT)

  • Built a deep learning–based text-to-SQL semantic parser for natural-language interfaces to DBMS.

Activities

Global Leadership Program (GLP), Computer Vision Lab, POSTECH

  • Researched real-time object detection and tracking.

POSTECH Laboratory for Unix Security (PLUS)

  • Studied Linux system vulnerabilities; participated in CTFs (e.g., DEFCON CTF 2017).

Teaching

Teaching Assistant for CSED 273 Digital System Design, POSTECH
Teaching Assistant for CSED 312 Operating System, POSTECH

Contact