Eunjin.Choi;

Hello! Welcome to my homepage.

I’m interested in representation learning, controllable generation, and multimodal learning for sequential and structured data. My experience includes dataset construction, preprocessing, and model development with Transformer and diffusion architectures. I received my Ph.D. from KAIST, where I worked on symbolic music modeling.

Things I like: automating, archiving, organizing, bleeps/bloops, and my husband.

Eunjin Choi

Experience

  • NHN Cloud
    AI/ML Researcher · Jan. 2026 – Present
    Recommendation Systems · Time-Series Modeling

Other experiences here.

Education

  • Korea Advanced Institute of Science and Technology (KAIST)
    Ph.D. in Culture Technology · Mar. 2022 – Aug. 2026
    Advisor: Juhan Nam
  • Korea Advanced Institute of Science and Technology (KAIST)
    M.S. in Culture Technology · Mar. 2020 – Feb. 2022
    Advisor: Juhan Nam
  • Daegu Gyeongbuk Institute of Science and Technology (DGIST)
    B.E. in Convergence · Mar. 2015 – Feb. 2020
    Magna cum laude

Selected Publications

  1. ICASSPD3PIA: A Discrete Denoising Diffusion Model for Piano Accompaniment Generation from Lead Sheet — preview
    D3PIA: A Discrete Denoising Diffusion Model for Piano Accompaniment Generation from Lead Sheet
    Eunjin Choi, Hounsu Kim, Hayeon Bang, Taegyun Kwon, Juhan Nam
    IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 2026
    • Discrete Diffusion
    • Controllable Generation
    • Symbolic Music Generation
    Paper Code Demo
  2. ISMIROn the De-duplication of the Lakh MIDI Dataset — preview
    On the De-duplication of the Lakh MIDI Dataset
    Eunjin Choi, Hyerin Kim, Jiwoo Ryu, Juhan Nam, Dasaem Jeong
    International Society for Music Information Retrieval Conference (ISMIR), 2025
    • Representation Learning
    • Dataset De-duplication
    Paper Code Demo
  3. ISMIRPianoBind: A Multimodal Joint Embedding Model for Pop-piano Music — preview
    PianoBind: A Multimodal Joint Embedding Model for Pop-piano Music
    Hayeon Bang, Eunjin Choi, Seungheon Doh, Juhan Nam
    International Society for Music Information Retrieval Conference (ISMIR), 2025
    • Representation Learning
    • Audio-Text-MIDI
    • Text-to-Music Retrieval
    Paper Code Demo
  4. ISMIRYM2413-MDB : A Multi-Instrumental FM Video Game Music Dataset with Emotion Annotations — preview
    YM2413-MDB : A Multi-Instrumental FM Video Game Music Dataset with Emotion Annotations
    Eunjin Choi, Yoonjin Chung, Seolhee Lee, Jong Ik Jeon, Taegyun Kwon, and Juhan Nam
    International Society for Music Information Retrieval Conference (ISMIR), 2022
    • Dataset Curation
    • Music Emotion Recognition
    • Emotion-Conditioned Generation
    Paper Code Demo

More publications on Google Scholar.