University of California, San Diego
Ph.D. in Computer Science and Engineering
2025 – Present
Haven Kim
Ph.D. Student, Computer Science and Engineering
I am a Ph.D. student in Computer Science and Engineering at UC San Diego, fortunate to be advised by Professor Julian McAuley. Concurrently, I am a visiting student at Carnegie Mellon University, working with Professor Chenyan Xiong and Dr. Chen Xu . This summer, I worked as an Applied Scientist Intern at Amazon in Seattle.
Education
University of California, San Diego
Korea Advanced Institute of Science and Technology (KAIST)
M.S. in Culture Technology
2021 – 2023
University of California, San Diego
B.A. in Interdisciplinary Computing and the Arts (GPA 3.93/4.00)
2015 – 2021
Industry / Research Experience
Amazon
Applied Scientist Intern — Seattle, WA
Topic: Generative Recommendation
2026.06 – 2026.09
Media Interaction Group, AIST
Research Intern — Tsukuba, Japan (Remote)
Supervisors: Kento Watanabe and Masataka Goto · Topic: Lyric Translation
2023.03 – 2023.04
Publications
For the most up-to-date list, please see my Google Scholar profile.
- “Reddit2Deezer: A Scalable Dataset for Real-World Grounded Conversational Music Recommendation.” 2026. [arXiv] [dataset]
- “Expressiveness Limits of Autoregressive Semantic ID Generation in Generative Recommendation.” 2026. [arXiv]
- “FusID: Modality-Fused Semantic IDs for Generative Music Recommendation.” Proceedings of the 27th International Society for Music Information Retrieval (ISMIR), 2026. [arXiv]
- “A Design Space for Live Music Agents.” Conference on Human Factors in Computing Systems (CHI), 2026.
- “Video-Guided Text-to-Music Generation Using Public Domain Movie Collections.” Proceedings of the 26th International Society for Music Information Retrieval (ISMIR), 2025. [arXiv] [demo]
- “TeaserGen: Generating Teasers for Long Documentaries.” International Conference on Learning Representations (ICLR), 2025. [arXiv]
- “Audio Prompt Adapter: Unleashing Music Editing Abilities for Text-to-Music with Lightweight Finetuning.” Proceedings of the 25th International Society for Music Information Retrieval (ISMIR), 2024. [arXiv]
- “K-pop Lyric Translation: Dataset, Analysis, and Neural Modelling.” The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING), 2024. [arXiv] [demo]
- “A Computational Evaluation Framework for Singable Lyric Translation.” Proceedings of the 24th International Society for Music Information Retrieval (ISMIR), 2023. [arXiv]
Teaching
Teaching Assistant & Guest Lecturer
CSE 153/253: Machine Learning for Music — UC San Diego
Contributed to homework design and lecture slides; delivered guest lectures for Module 5 (Music Recommendation and Other Trending Topics).
Spring 2026
Academic Service
Reviewer
Conference / Workshop
NeurIPS, ISMIR, SIGIR, ICASSP, IJCNN, EMNLP, NLP4MusA Workshop, AI4Music Workshop
Journal
ACM Transactions on Recommender Systems, ACM Computing Surveys
Media Coverage
Txabarriaga, Rocío. “Can Translated Songs Mirror Meaning and Still Be ‘Singable’?” Slator, September 4, 2023.