I’m a third-year PhD student in Electrical and Computer Engineering at UC Santa Barbara, advised by Prof. Peng Li. I earned my MS and BS degrees in Electrical and Computer Engineering from Seoul National University, where I was advised by Prof. Jungwoo Lee during my MS.

I’m passionate about developing machine learning methods for data-limited real-world problems, where observations are scarce, expensive, or difficult to obtain. My current research focuses on using large language models for black-box optimization, particularly through the mathematical lens of Bayesian optimization. I study how language models can incorporate prior knowledge, interpret problem descriptions, and guide exploration–exploitation decisions when only a limited number of function evaluations are available. More broadly, I am interested in building learning systems that can reason and adapt under uncertainty, limited feedback, and complex objective landscapes. Previously, I worked on model collapse in generative models, multimodal time-series learning, and long-tailed recognition, which shaped my broader interest in data-efficient and robust machine learning.

I love tackling real-world challenges where data is messy, feedback is limited, and objectives are hard to evaluate. Always happy to chat about AI, LLMs, black-box optimization, Bayesian optimization, generative models, and making ML work in the wild!

Contacts: youngseok_yoon@ucsb.edu

News:

• Jun - Sep 2025: Summer internship at Samsung Research America in Mountain View, CA.
• Nov 2024: Introduced my work about long-tailed classification: Link
• Jul 2024: Introduced my work about model collapse: Link
• Sep 2023: Started my PhD at UC Santa Barbara!

Academic Services:

• Reviewer for ICLR, CVPR, ECCV 2026.
• Reviewer for ICLR, CVPR, NeurIPS (Top reviewer), WACV 2025.
• Reviewer for SaTML, ECCV 2024.
• Reviewer for NeurIPS 2024 Worshop AIM-FM.