Dongjin Kang

Hi! I am a M.S. student at LangAGI Lab advised by Jinyoung Yeo. Previously, I received B.S. in Computer Science from Yonsei University in Feb. 2024.

My research goal is to enable digital agents to interact with their environments and master increasingly complex, long-horizon tasks. I am currently focused on: (i) Reinforcement Learning (RL)—especially exploration strategies that expand the reasoning boundaries of foundational models and the design of effective reward signals for policy learning, and (ii) Tackling the Problem of Overthinking in Models—whether by enabling them to recognize their own errors to mitigate overconfidence or by encouraging reasoning within latent spaces. Additionally, I concentrate on interpretability in language models, identifying opportunities for improvement and using those insights to enhance model performance.

Publications

^‡ indicates equal contribution.

Under-review/Preprints

Web-Shepherd: Advancing PRMs for Reinforcing Web Agents

Hyungjoo Chae^‡, Sunghwan Kim^‡, Junhee Cho, Seungone Kim, Seungjun Moon, Gyeom Hwangbo, Dongha Lim, Minjin Kim, Yeonjun Hwang, Minju Gwak, Dongwook Choi, Minseok Kang, Gwanhoon Im, ByeongUng Cho, Hyojun Kim, Jun Hee Han, Taeyoon Kwon, Minju Kim, Beong-woo Kwak, Dongjin Kang, Jinyoung Yeo

Under Review at NeurIPS 2025

ToolHaystack: Stress-Testing Tool-Augmented Language Models in Realistic Long-Term Interactions

Beong-woo Kwak, Minju Kim, Dongha Lim, Hyungjoo Chae, Dongjin Kang, Sunghwan Kim, Dongil Yang, Jinyoung Yeo

Under Review at EMNLP 2025

2025

Rethinking Reward Model Evaluation Through the Lens of Reward Overoptimization

Sunghwan Kim^‡, Dongjin Kang^‡, Taeyoon Kwon, Hyungjoo Chae, Dongha Lee, Jinyoung Yeo

ACL 2025 main (Oral)

One Missing Piece for Open-Source Reasoning Models: A Dataset to Mitigate Cold-Starting Short CoT LLMs in RL

Hyungjoo Chae^‡, Dongjin Kang^‡, Jihyuk Kim, Beong-woo Kwak, Sunghyun Park, Haeju Park, Jinyoung Yeo, Moontae Lee, Kyungjae Lee

ACL 2024 Industry

2024

Evaluating Robustness of Reward Models for Mathematical Reasoning

Sunghwan Kim^‡, Dongjin Kang^‡, Taeyoon Kwon, Hyungjoo Chae, Jungsoo Won, Dongha Lee, Jinyoung Yeo

Arxiv preprint.

Coffee-gym: An environment for evaluating and improving natural language feedback on erroneous code

Hyungjoo Chae^‡, Taeyoon Kwon^‡, Seungjun Moon^‡, Yongho Song, Dongjin Kang, Kai Tzu-iunn Ong, Beong-woo Kwak, Seonghyeon Bae, Seung-won Hwang, Jinyoung Yeo

EMNLP 2024 main

Cactus: Towards Psychological Counseling Conversations using Cognitive Behavioral Theory

Suyeon Lee^‡, Sunghwan Kim^‡, Minju Kim^‡, Dongjin Kang, Dongil Yang, Harim Kim, Minseok Kang, Dayi Jung, Min Hee Kim, Seungbeen Lee, Kyoung-Mee Chung, Youngjae Yu, Dongha Lee, Jinyoung Yeo

EMNLP 2024 findings

🏆 Can Large Language Models be Good Emotional Supporter? Mitigating Preference Bias on Emotional Support Conversation

Dongjin Kang^‡, Sunghwan Kim^‡, Taeyoon Kwon, Seungjun Moon, Hyunsouk Cho, Youngjae Yu, Dongha Lee, Jinyoung Yeo

[Outstanding Paper Award] ACL 2024 main

Coffee: Boost your code llms by fixing bugs with feedback

Seungjun Moon^‡, Yongho Song^‡, Hyungjoo Chae^‡, Taeyoon Kwon, Dongjin Kang, Kai Tzu-iunn Ong, Seung-won Hwang, Jinyoung Yeo

Arxiv Preprint

Large language models are clinical reasoners: Reasoning-aware diagnosis framework with prompt-generated rationales

Taeyoon Kwon^‡, Kai Tzu-iunn Ong^‡, Dongjin Kang, Seungjun Moon, Jeong Ryong Lee, Dosik Hwang, Yongsik Sim, Beomseok Sohn, Dongha Lee, Jinyoung Yeo

AAAI 2024

Dongjin Kang

Recent News

Publications

Under-review/Preprints

2025

2024