Yujin Kim

Hi! 👋 I am a Ph.D. student at KAIST AI (Advisors: Se-Young Yun and Hwanjun Song).

My research focuses on enhancing AI's Reasoning Capabilities and accelerating its Inference Efficiency. Specifically, I enhance model's ability by scaling synthetic data without direct human supervision [C2, C3, C5]. Concurrently, I accelerate inference speeds through novel model architectures, adaptive computation, and efficient decoding strategies [C1, C6, W3].

Contact me: yujin [dot] src [at] gmail [dot] com [CV / Scholar / Github / LinkedIn]

Publications 📑

(C: Conference, W: Workshop, P: Preprint, *: Equal Contribution (1st Authors), ^: Equal Advising)

2025

[C6/W4] Mixture-of-Recursions: Learning Dynamic Recursive Depths for Adaptive Token-Level Computation

Sangmin Bae*, Yujin Kim*, Reza Bayat*, Sungnyun Kim, Jiyoun Ha, Tal Schuster, Adam Fisch, Hrayr Harutyunyan, Ziwei Ji, Aaron Courville^, Se-Young Yun^
The Thirty-Ninth Annual Conference on Neural Information Processing Systems (NeurIPS). 2025. San Diego [paper] [code (468+⭐)]
ICML Workshop on Efficient Systems for Foundation Models (ES-FoMo III). 2025

[W3] Guiding Reasoning in Small Language Models with LLM Assistance

Yujin Kim*, Euiin Yi*, Minu Kim, Se-Young Yun^, Taehyeon Kim^
COLM Workshop on Test-time Scaling and Reasoning Models. 2025. Toronoto. [paper] [code]

[C5] Self-training elicits concise reasoning in large language models

Tergel Munkhbat*, Namgyu Ho*, Seo Hyun Kim*, Yongjin Yang, Yujin Kim, Se-Young Yun
Findings of the Association for Computational Linguistics: ACL 2025 (Findings of ACL). 2025. Vienna [paper] [code]

2024

[C4] BAPO: Base-Anchored Preference Optimization for Personalized Alignment in Large Language Models

Gihun Lee, Minchan Jeong, Yujin Kim, Hojung Jung, Jaehoon Oh, Sangmook Kim, Se-Young Yun
Findings of the Association for Computational Linguistics: EMNLP 2024. Miami [paper]

[C3/W2] Carpe Diem: On the Evaluation of World Knowledge in Lifelong Language Models

Yujin Kim, Jaehong Yoon, Seonghyeon Ye, Sangmin Bae, Namgyu Ho, Sung Ju Hwang^, Se-young Yun^.
Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL). 2024. Mexico City [paper] [code]
NeurIPS Workshop on Synthetic Data Generation with Generative AI. 2023. (Oral)

2023

[C2] HARE: Explainable Hate Speech Detection with Step-by-Step Reasoning

Yongjin Yang*, Joonkee Kim*, Yujin Kim*, Namgyu Ho, James Thorne^, Se-young Yun^.
Findings of the Association for Computational Linguistics: EMNLP 2023 (Findings of EMNLP). 2023. Singapore [paper] [code]

[C1] NASH: A Simple Unified Framework of Structured Pruning for Accelerating Encoder-Decoder Language Models

Jongwoo Ko*, Seungjoon Park*, Yujin Kim, Sumyeong Ahn^, Du-Seong Chang, Euijai Ahn, Se-Young Yun^.
Findings of the Association for Computational Linguistics: EMNLP 2023 (Findings of EMNLP). 2023. Singapore [paper] [code]

2022

[P1] How to Fine-tune Models with Few Samples: Update, Data Augmentation, and Test-time Augmentation

Yujin Kim*, Jaehoon Oh*, Sungnyun Kim, Se-Young Yun.
Preprint. 2022. [paper] [code]

[W1] Revisiting the Updates of a Pre-trained Model for Few-shot Learning

Yujin Kim*, Jaehoon Oh*, Sungnyun Kim, Se-Young Yun.
ICML Workshop on Updatable Machine Learning (UpML). 2022. (Oral) [paper]

Experiences 🌏

Research Intern @NAVER Cloud

Bundang-Gu, Gyeonggi, South Korea
Mar. 2024 - Aug. 2024 (6 month)
Developed an inference-efficient adapter network under multiple task-specific serving scenarios.
NVIDIA-HPE TensorRT-LLM Hackathon

Awards & Honors 🏆

SKT AI Fellowship, Grand Prize (2021)

2D-3D keypoint matching model by incorporating graph attention network.
Panel detection model by constructing panel dataset with synthetic data generation.

Education 🧑‍🎓

Korea Advanced Institue of Science and Technology (KAIST), Seoul, Korea, Mar. 2025 - Ongoing

Ph.D in Kim Jaechul Graduate School of Artificial Intelligence (Advisor: Se-Young Yun and Hwanjun Song)

Korea Advanced Institue of Science and Technology (KAIST), Seoul, Korea, June. 2022 - Feb. 2025

M.S. in Kim Jaechul Graduate School of Artificial Intelligence (Advisor: Se-Young Yun)

Thesis: Carpe Diem: On the Evaluation of World Knowledge in Lifelong Language Model

Sogang University, Seoul, Korea, Mar. 2017 - Aug. 2022

B.A. in Economics and B.S in Artificial Intelligence

Page updated

Google Sites

Report abuse