Subeen Ho

Subeen Ho

M.S./Ph.D. Student, DAIS Lab, Korea University

hosubin02@korea.ac.kr · GitHub · LinkedIn

scroll

About

I am an M.S./Ph.D. student at the Data & Adaptive Intelligence Systems (DAIS) Lab, Korea University, advised by Prof. Susik Yoon.

My research focuses on building structure-aware agentic retrieval systems that can understand and reason over structured information. I am interested in how AI systems can leverage the inherent structure of tables, graphs, and heterogeneous data to navigate information spaces, discover relevant evidence, and connect it to answer complex questions.

Ultimately, I aim to develop retrieval systems that go beyond simply finding relevant information, systems that can understand how information is structured, decide where to look next, and reason across sources to discover how pieces of knowledge fit together.

News

  • 2026.08 PEARL was accepted to the EMNLP 2026 Main Conference
  • 2026.08 Table Retrieval via Query Augmentation and Vertical Partitioning was selected as a Best Presentation Paper at KCC 2026
  • 2025.12 Received the Dean's Award from the College of Informatics, Korea University
  • 2025.11 Won 3rd Prize at INThon, the Korea University Datathon
  • 2025.08 CREAM was accepted to KDD 2026
  • 2024.12 Joined DAIS Lab, Korea University as an undergraduate intern

Publications

PEARL: Front-Loading Relational Chains for Multi-Hop Table Retrieval

Subeen Ho, Hyeongu Kang, SeongKu Kang, Susik Yoon†

CREAM: Continual Retrieval on Dynamic Streaming Corpora with Adaptive Soft Memory

HuiJeong Son, Hyeongu Kang, Sunho Kim, Subeen Ho, SeongKu Kang, Dongha Lee, Susik Yoon†

Education

  • 2026 — PRESENT

    M.S./Ph.D. in Computer Science and Engineering

    Korea University · DAIS Lab

  • 2022 — 2026

    B.S. in Computer Science and Engineering

    Korea University

Awards

Certifications

  • 2023Advanced Data Analytics Semi-Professional (ADsP)Korea Data Agency
  • 2023SQL Developer (SQLD)Korea Data Agency