Jihwan Oh

Ph.D. Student · Industrial and Systems Engineering · KAIST

Jihwan Oh

KAIST

Daejeon, South Korea

jh.oh@kaist.ac.kr

I am a Ph.D. student at KAIST, advised by Prof. Hwanjun Song.

My current research interest is AI for Research: using AI agents to support and automate research workflows, including peer review and scientific writing.

Previously, I worked on summarization and alignment, with a focus on faithful summarization, fine-grained feedback, and preference optimization and data.

All publications Curriculum vitae

selected publications

  1. EMNLP 2026
    Show Me How You Cite: Citation Fidelity, Usage, and Topology in the Evidence Behavior of AI Scientific Writing
    Jihwan Oh*, Gyeonghun Sun*, Jisoo Kim, Yuho Lee, Taewon Yun, Jisu Shin, and Hwanjun Song†
    In Empirical Methods in Natural Language Processing (EMNLP), Main, 2026
    Oral presentation
    Paper
  2. COLM 2026
    Quality of Rejections Matters: Preference Data Construction for Faithful Summarization
    Jihwan Oh, Taewon Yun, Minkyu Che, Sundong Kim, and Hwanjun Song†
    In Conference on Language Modeling (COLM), 2026
    Paper
  3. COLING 2025
    Learning to Verify Summary Facts with Fine-Grained LLM Feedback
    Jihwan Oh, Jeonghwan Choi, Nicole Hee-Yeon Kim, Taewon Yun, and Hwanjun Song†
    In International Conference on Computational Linguistics (COLING), 2025
    Oral presentation
    Paper