cv

M.S. in Graduate School of Culture Technology

Korea Advanced Institute of Science and Technology (KAIST)

• Advisor: Juhan Nam

B.S. in Computer Science and Engineering

Seoul National University (SNU)

• Double Major in Statistics
• GPA: 3.90/4.3

Music and Audio Computing Lab, KAIST

Research Intern
Advisor: Eunjin Choi, Prof. Juhan Nam

• Developed AImoclips, a comprehensive benchmark for evaluating how well text-to-music (TTM) generation systems convey intended emotions to human listeners. The benchmark offers valuable insights into model-specific emotion rendering characteristics and supports future development of emotionally aligned TTM systems.

Department of Mathematics, UCLA

Research Intern
Advisor: Hyunsik Chae, Prof. Ernest Ryu

• Developed AVSBench (Atomic Visual Skills Benchmark), a benchmark for evaluating visual understanding ability of visual language models(VLMs) by decomposing it into atomic visual skills. Found that VLMs struggle with most of these atomic visual skills that are obvious to humans.

Music and Audio Research Group, Seoul National University

Research Intern
Prof. Kyogu Lee

• Implemented and tested Theme Transformer, a Transformer-based music generation framework conditioned on a small, thematic musical piece that generates symbolic music consistent with the theme.

Human-Computer Interaction Lab, Seoul National University

UROP(Undergraduate Research Opportunities Program) Student
Advisor: Hyeon Jeon, Prof. Jinwook Seo

• Developed UMATO (Uniform Manifold Approximation with Two-phase Optimization), a novel dimension reduction algorithm for data visualization and evaluated its scalability.

International Research Institute for Cyber Security

Research Intern

• Evaluated a novel lattice-based signature scheme submitted to the Korean Post-Quantum Competition

AttentionX, an AI Research & Startup Group

3rd term member

• A member of a research team studying the evaluation of ambiguity resolution ability in large language models (LLMs), cooperating with Language & Knowledge Lab, KAIST
• Leader of a research team studying music editing methods using generative models

Best Poster Award

Korean Society for Music Perception and Cognition (KSMPC)

Title: Can Emotion Concepts be Reflected Well in Text-to-Music Generation Models?

Seoul National University Semiconductor Specialization School Semiconductor Track Program

Korea Institute for Advancement of Technology (KIAT)

Hanseong Nobel Scholarship

Hanseong Sonjaehan Scholarship Foundation

AImoclips: A Benchmark for Evaluating Emotion Conveyance in Text-to-Music Generation

Gyehun Go, Satbyul Han, Ahyeon Choi, Eunjin Choi, Juhan Nam, Jeong Mi Park

HCMIR25: 3rd Workshop on Human-Centric Music Information Research at ISMIR'25

UMATO: Uniform Manifold Approximation with Two-phase Optimization for Scalable and Accurate Data Visualization

Hyeon Jeon, Kwon Ko, Soohyun Lee, Jake Hyun, Taehyun Yang, Gyehun Go, Jaemin Go, Jinwook Seo

IEEE Transactions on Visualization and Computer Graphics (TVCG)

Decomposing Complex Visual Comprehension into Atomic Visual Skills for Vision Language Models

Hyunsik Chae, Seungwoo Yoon, Chloe Yewon Chun, Gyehun Go, Yongin Cho, Gyeongmin Lee, Ernest K. Ryu

The 4th Workshop on Mathematical Reasoning and AI at NeurIPS'24