Yoonjae Jung

I am an Integrated Ph.D. candidate at the Graduate School of Artificial Intelligence, POSTECH, where I am advised by Prof. Sangdon Park in the Machine Learning Lab. I received my B.S. in Computer Science and Engineering from POSTECH.

My research focuses on reinforcement learning for reliable and scalable physical AI systems. I am particularly interested in developing learning algorithms that enable embodied agents to operate safely and robustly under distribution shifts and unforeseen situations in the real world. Within this broader direction, I study generative policies, such as diffusion and flow-based policies, with an emphasis on preserving diverse and multimodal behaviors during reinforcement learning rather than collapsing to a narrow set of actions. I am also interested in designing reinforcement learning algorithms that remain effective and computationally practical as policy models, datasets, and robotic systems scale up.

News

Sep 15, 2025	Received the POSTECHIAN Fellowship from POSTECH.
Sep 01, 2025	Started my Integrated Ph.D. at POSTECH GSAI, Machine Learning Lab.

Selected Publications

NeurIPS 2026

Online Conformal Abstention for Factuality Control Under Adversarial Bandit Feedback

Minjae Lee^*, Yoonjae Jung^*, and Sangdon Park

arXiv preprint arXiv:2506.14067, Jun 2025

*Equal contribution

Abs arXiv

We address reliability concerns in interactive generative systems by proposing ExAUL, an online learning framework that ensures systems answer only when confident, handling partial user feedback in adversarial settings. A conversion lemma links bandit algorithm regret to false discovery rate bounds, and a feedback unlocking strategy achieves O(sqrt(T) ln|H|) regret, translating to O(sqrt(T)) false discovery rate risk control despite incomplete feedback. We validate the framework on question-answering tasks with large language models under varied challenging conditions.

Honors and awards

POSTECHIAN Fellowship ($2,000), POSTECH. September 2025
Second Prize, 2023 TECHATHON: WAVE, POSTECH. February 2023
Encouragement Award, 2021 Undergraduate Group Research Program, POSTECH. February 2022