Hello, I am a Ph.D student in Artificial Intelligence at Yonsei University, advised by Prof. Min Song in DeepText LAB.
My research centers on understanding the structural and symbolic dimensions of music, with a particular interest in how machines can analyze, interpret, and visualize musical patterns. I am especially drawn to exploring the latent representations of music and how they can be used to uncover its underlying compositional logic. Recently, I have been focused on the intersection of symbolic music processing and deep learning, using attention-based models to reveal how machines “perceive” music.
I am currently completing my Ph.D. coursework and plan to transition into full-time research or industry work beginning in the second half of 2025. My work is conducted remotely, and I actively lead the MAAP (Music AI Assemble People) initiative, a community of students and professionals exploring the frontiers of music AI. Through this, I mentor a team of five members from diverse backgrounds, fostering collaboration and innovation in the field.
As part of my doctoral research, I have applied for a research scholarship funded by the National Research Foundation of Korea, proposing a study on Multi-Pitch Estimation Models Based on Synthetic Polyphonic Vocal Data. This project reflects my broader goal of developing systems that can dissect and understand complex vocal structures, laying the foundation for more expressive and musically intelligent AI.
🔥 News
- Sep 2025 : Our AIBA paper is accepted to NeurIPS 2025 Workshop (AI for Music) 🎶
- Sep 2025 : Awarded Brian Impact Foundation (Kakao, USD 1.5K Per paper)
- Aug 2025 : Selected for NRF Ph.D. Research Fellowship (RS-2025-25422688, USD 20K)
- Jul 2025 : I will be working at KRAFTON AI in August!
- May 2025 : I'm going to USC summer session in July!
2024 News
- Oct 2024 : Our model has been downloaded over 33,000 times and has ranked 7th among trending models worldwide!
- Sep 2024 : Illustrious technical report is submitted in arXiv and HuggingFace!
- Aug 2024 : 1st Anniversary at my first company, Onoma AI 🎉
- May 2024 : One paper is accepted to IIAI-AAI 2024!
- Jan 2024 : One paper is accepted to CVPR!
🎓 Education
University of Southern California July 2025 - Aug 2025
Summer Session, Artificial Intelligence
Yonsei University, Seoul, Republic of Korea Mar 2023 - Present
Ph.D. Student, Department of Artificial Intelligence
Soongsil University, Seoul, Republic of Korea Mar 2017 - Feb 2023
B.S., Major in Media
✏️ Publications
Jamendo-QA: A Large-Scale Music Question Answering Dataset
Jun Young Koh, Sooyong Kim, Yongwon Choi, Gyuhyeong Choi
Submitt in ICASSP 2026 [arXiv] [PDF] [HuggingFace]
AIBA: Attention-based Instrument Band Alignment for Text-to-Audio Diffusion
Jun Young Koh, Sooyong Kim, Gyuhyeong Choi, Yongwon Choi
NeurIPS 2025 Workshop on AI for Music: Where Creativity Meets Computation [arXiv] [PDF]
Let Triggers Control : Frequency-aware Dropout for Effective Token Control
Jun Young Koh, Hoyeon Moon, Dongha Kim, Seungmin Lee, Sanghyun Park, Min Song
Under Review
Illustrious : an Open Advanced Illustration Model
Sang Hyun Park*, Jun Young Koh*, Junha Lee, Joy Song, Dongha Kim, Hoyeon Moon, Hyunju Lee, Min Song
Technical Report [arXiv] [PDF] [HuggingFace] [CivitAI]
Improving Text Generation on Images with Synthetic Captions
Jun Young Koh*, Sang Hyun Park*, Joy Song*
IIAI AAI 2024 7th International Conference on Interaction Design and Digital Creation / Computing [arXiv] [PDF]
CAT : Contrastive Adapter Training for Personalized Image Generation
Jaewan Park*, Sang Hyun Park*, Jun Young Koh*, Junha Lee*, Min Song
CVPR 2024 Workshop Generative Models for Computer Vision [arXiv] [PDF]
Proposal of 3D Camera-Based Digital Coordinate Recognition Technology
Junyoung Koh, Kanghee Lee
Proceedings of the Korean Society of Computer Information Conference 2022 [PDF]
* Equal Contribution
📚 Academic Services
Conference Reviewer
ISMIR / AAAI / NeurIPS Reviewer Aug 2025 - Present
Leader
Modulabs MAAP(Music AI Assemble People) Leader May 2025 - Present
Journal Reviewer
IEEE Access (SCI) Reviewer Aug 2023 - Present
IT Volunteer Service
World Friends Korea Paraguay IT Volunteer Aug 2018 - Sep 2018
🏢 Work Experience
Krafton Aug 2025 - Current
AI Researcher Internship
Onoma AI Aug 2023 - Feb 2025
AI Team Leader & AI Researcher
🏆 Honors & Awards
CES 2025 Innovation Award - AI category 2025
CES 2024 Innovation Award - AI category 2024
💰 Funding
National Research Foundation of Korea (Ministry of Science and ICT)
Ph.D. Research Fellowship (Project: Multi-Pitch Estimation Model Based on Synthetic Polyphonic Vocal Data)
Grant No. RS-2025-25422688 · USD 20K (KRW 25M) for 1 year (Sep 2025 – Aug 2026)
Brian Impact Foundation (Kakao)
Publication support program, providing research funding of USD 1.5K (KRW 2M) per accepted paper (2025 – Present)
Project
Development of an AI Media Art Therapy Platform incorporating Color, Play, and Music Therapies
Junyoung Koh, Dain Park, Junghun Ha, Kanghee Lee
[code] [Project Site]