Youngbin Choi

Portrait
Combined Ph.D. Student

I am a researcher at the Machine Learning Lab (ML Lab), POSTECH, advised by Prof. Dongwoo Kim.

My research studies how the context a large language model (LLM) operates in, namely its interaction history, user feedback, and the harness around it, shapes model behavior. Building on this understanding, I develop methods that make LLMs easier to guide, more efficient to refine, and adaptive to individual users. My work spans human–AI collaboration, personalization of LLMs, and reinforcement learning for LLMs.

Research Interests
  • Human-AI collaboration
  • LLM personalization
  • Reinforcement learning for LLMs
Curriculum Vitae
Email Google Scholar LinkedIn GitHub
Education
  • POSTECH
    Combined Ph.D. Student @ Machine Learning Lab
    Advised by Prof. Dongwoo Kim
    Feb. 2022 - present
  • POSTECH
    B.S. in Computer Science
    Feb. 2018 - Feb. 2022
News
2026
🏆 In-Place Feedback received the Outstanding Paper Award at EMNLP 2026
Oct 27
Three papers accepted to EMNLP 2026 (2 main, 1 findings)
Aug 20
A Paper accepted to ICML 2026
May 01
A Paper accepted to EACL-Findings 2026
Jan 03
2025
Three papers accepted to EMNLP 2025 (1 main, 2 findings)
Aug 21
Selected Publications (view all )
New Model, Old Habits: Behavior Transfer under Model Swapping in Multi-Session LLM Agents

Youngbin Choi*, Sehyun Hwang*, Suhyeon Jun, Seoyeon Lee, Sunghyun Choi, MoonJeong Park, Dongwoo Kim (* equal contribution)

Under review 2026

[Arxiv]

New Model, Old Habits: Behavior Transfer under Model Swapping in Multi-Session LLM Agents

Youngbin Choi*, Sehyun Hwang*, Suhyeon Jun, Seoyeon Lee, Sunghyun Choi, MoonJeong Park, Dongwoo Kim (* equal contribution)

Under review 2026

[Arxiv]

In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration

Youngbin Choi*, Minjong Lee*, Saemi Moon, Seunghyuk Cho, Chaehyun Chung, MoonJeong Park, Dongwoo Kim (* equal contribution)

EMNLP 2026 (🏆 Outstanding Paper Award, Oral)

[Arxiv]

In-Place Feedback: Reliable Refinement for Multi-Turn Expert-LLM Collaboration

Youngbin Choi*, Minjong Lee*, Saemi Moon, Seunghyuk Cho, Chaehyun Chung, MoonJeong Park, Dongwoo Kim (* equal contribution)

EMNLP 2026 (🏆 Outstanding Paper Award, Oral)

[Arxiv]

The Interplay of Harness Design and Post-Training in LLM Agents

Kyungmin Kim*, Youngbin Choi*, Seoyeon Lee, Suhyeon Jun, Dongwoo Kim†, Sangdon Park† (* equal contribution, † co-advising)

EMNLP-Findings 2026

[Arxiv]

The Interplay of Harness Design and Post-Training in LLM Agents

Kyungmin Kim*, Youngbin Choi*, Seoyeon Lee, Suhyeon Jun, Dongwoo Kim†, Sangdon Park† (* equal contribution, † co-advising)

EMNLP-Findings 2026

[Arxiv]

CoPL: Collaborative Preference Learning for Personalizing LLMs

Youngbin Choi, Seunghyuk Cho, Minjong Lee, MoonJeong Park, Yesong Ko, Jungseul Ok, Dongwoo Kim

EMNLP 2025

[Arxiv] [Github]

CoPL: Collaborative Preference Learning for Personalizing LLMs

Youngbin Choi, Seunghyuk Cho, Minjong Lee, MoonJeong Park, Yesong Ko, Jungseul Ok, Dongwoo Kim

EMNLP 2025

[Arxiv] [Github]

All publications