<aside> 🗣️

Favorite Quote: “If I have seen further, it is by standing on the shoulders of Giants.” — Isaac Newton (The Correspondence of Isaac Newton: Volume 5, 1709–1713)

</aside>

Research Interests

<aside>

Efficient / Effective Transfer of Pre-trained model

Vision-Language Pre-training

Domain-Adaptive Pre-training of LLM

</aside>

Education

<aside> <img src="/icons/graduate_gray.svg" alt="/icons/graduate_gray.svg" width="40px" /> M.S./Ph.D [2022.03 ~Current] (Advisor: Taesup Moon)

Department of Electrical and Computer Engineering, Seoul National University(SNU)

</aside>

<aside> <img src="/icons/graduate_gray.svg" alt="/icons/graduate_gray.svg" width="40px" />

BS [2015.03~2022.02]

[GPA 4.01/4.3] Department of Mechanical and Aerospace Engineering, Seoul National University

[GPA 3.93/4.3] Department of Computer Science and Engineering, Seoul National University

</aside>

Publications

DoMIX: An Efficient Framework for Exploiting Domain Knowledge in Fine-Tuning

• Dohoon Kim, Donghun Kang, and Taesup Moon
• The 63rd Annual Meeting of the Association for Computational Linguistics (ACL), 2025.

MAFA: Managing False Negatives for Vision-Language Pre-training

• Jaeseok Byun*, Dohoon Kim*, and Taesup Moon (* equal contribution)