LLM2Fx-Tools: Tool Calling for Music Post-Production
International Conference on Learning Representations (ICLR), 2026
Research Scientist
Sony Group Corporation, Creative AI Team
Seungheon Doh is a Research Scientist at Sony. His research focuses on music and machine learning, including music representation, understanding, recommendation, and generation. Recently, he has been particularly interested in agentic media processing, where language models coordinate tools to understand and transform multimodal media. He received his Ph.D. in Culture Technology from KAIST, where he was advised by Juhan Nam and Keunwoo Choi.
Previously, he held research internships at Sony AI, Adobe Research, Chartmetric, Naver, and ByteDance, and was a visiting scholar at NYU CILVR and MARL. He was also a founding member of talkpl.ai.
Internships at Sony: Feel free to send me a LinkedIn message with a brief note about your research interests, or ask for a coffee chat if we meet at a conference.
International Conference on Learning Representations (ICLR), 2026
Workshop on AI for Music, Neural Information Processing Systems (NeurIPS-AI4Music), 2025
IEEE Transactions on Multimedia (submitted), 2025
International Society for Music Information Retrieval Conference (ISMIR), 2023