I am a second-year Ph.D. candidate in Electrical and Computer Engineering at Seoul National University, advised by Prof. Kyomin Jung. My research explores methods to make large language models and large multi-modal models reason more reliably and efficiently. More broadly, I am interested in scalable methods for post-training and deploying intelligent agents that can learn from interactions with the environment and solve complex tasks efficiently.

News

Aug. 2026 One paper have been accepted to EMNLP 2026!
Aug. 2026 I will be joining LG AI Research as a research intern!
Jul. 2026 One paper has been accepted to COLM 2026!
Jun. 2026 One paper has been accepted to the ICML 2026 AI4Math Workshop as a Spotlight paper!
May. 2026 One paper has been accepted to ICML 2026!
Nov. 2025 One paper has been accepted to AAAI 2026!
Aug. 2025 Two papers have been accepted to EMNLP 2025!
Sep. 2024 One paper has been accepted to the Findings of EMNLP 2024!

Publications

Are We Measuring Strategy or Phrasing? The Gap Between Surface- and Approach-Level Diversity in LLM Math Reasoning
Sangmook Lee, Minbeom Kim, Jeonghye Kim, Dohyung Kim, Sojeong Rhee, Kyomin Jung
EMNLP 2026 & ICML 2026 AI4Math Workshop Spotlight
Why Does Self-Distillation (Sometimes) Degrade the Reasoning Capability of LLMs?
Jeonghye Kim, Xufang Luo, Minbeom Kim, Sangmook Lee, Dohyung Kim, Jiwon Jeon, Dongsheng Li, Yuqing Yang
COLM 2026 [Code]
Beyond Normalization: Rethinking the Partition Function as a Difficulty Scheduler for RLVR
Dohyung Kim, Minbeom Kim, Jeonghye Kim, Sangmook Lee, Sojeong Rhee, Kyomin Jung
ICML 2026 [Code]
Confidence-Guided Stepwise Model Routing for Cost-Efficient Reasoning
Sangmook Lee*, Dohyung Kim*, Hyukhun Koh, Nakyeong Yang, Kyomin Jung
AAAI 2026 [Code]
ReflAct: World-Grounded Decision Making in LLM Agents via Goal-State Reflection
Jeonghye Kim, Sojeong Rhee, Minbeom Kim, Dohyung Kim, Sangmook Lee, Youngchul Sung, Kyomin Jung
EMNLP 2025 [Code]
Conditional [MASK] Discrete Diffusion Language Model
Hyukhun Koh, Minha Jhang, Dohyung Kim, Sangmook Lee, Kyomin Jung
EMNLP 2025 [Code]
Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metric
Hyukhun Koh, Dohyung Kim, Minwoo Lee, Kyomin Jung
Findings of EMNLP 2024 [Code]

Work Experiences

NLP Research Scientist Intern @ LG AI Research
Seoul, South Korea
2026.08 — Present

Education

Ph.D. Candidate in Electrical and Computer Engineering
Seoul National University
2025.03 — Present
B.S. in Electrical and Computer Engineering
Seoul National University, Cum Laude
2019.03 — 2025.02