Scholar
Eunjin Choi
Google Scholar ID: 3VWIf88AAAAJ
KAIST
Citations & Impact
Citations
24
H-index
2
i10-index
1
Publications
11
Co-authors
10
list available
Publications
Resume (English only)
Academic Achievements
- D3PIA: A Discrete Denoising Diffusion Model for Piano Accompaniment Generation from Lead Sheet (Submitted)
- On the De-duplication of the Lakh MIDI Dataset
- PianoBind: A Multimodal Joint Embedding Model for Pop-piano Music
- Teaching Chorale Generation Model to Avoid Parallel Motions
- Mel2Word: A Text-based Melody Representation for Symbolic Music Analysis
- Bridging Audio and Symbolic Piano Data through a Web-Based Annotation Interface
- YM2413-MDB
Background
- Ph.D. Candidate at the Music and Audio Computing Lab, KAIST
- Machine learning researcher focusing on representation learning and controllable generative modeling for sequential and structured data
- Previous work primarily focused on symbolic music data, exploring structured representations such as tokenized sequences, matrices, and graphs for music generation and understanding
- Hands-on experience across the full ML pipeline, including dataset construction, preprocessing, and model development, especially with Transformer- and Diffusion-based architectures
- Long-term goal is to enable machines to learn and understand underlying structures and patterns in human-created data
