Seungyeon Jwa
I am a M.S./Ph.D. student in Artificial Intelligence at Seoul National University,
advised by Prof. Jonghyun Choi.
I am currently a visiting graduate student at the Minnesota NLP Group,
University of Minnesota, hosted by Prof. Dongyeop Kang (Apr – Oct 2026).
I am broadly interested in evaluating and training language models with systematic, data-centric approaches,
with a focus on automated and reliable evaluation of LLMs, quantifying LLM behaviors, and data synthesis and curation for training.
Publications
Becoming Experienced Judges: Selective Test-Time Learning for Evaluators
OffsetBias: Leveraging Debiased Data for Tuning Evaluators
Junsoo Park*,
Seungyeon Jwa*,
Meiying Ren, Daeyoung Kim, Sanghyuk Choi
Mitigating Hate Speech in Korean Open-domain Chatbot using CTRL
Seungyeon Jwa*,
Young-rok Cha*, Moonsu Han, Donghoon Shin
HCLT 2021
Education
Seoul National University
Sep 2025 – Present
Integrated M.A./Ph.D. in Artificial Intelligence
Seoul National University
Mar 2015 – Feb 2022
B.S. in Computer Science and Engineering · Computational Sciences · Food and Nutrition
Experience
NLP Researcher · Application LM Team (Alignment) · Feb 2024 – Feb 2025
- Designed a debiasing data pipeline to mitigate evaluator bias
- Trained a bias-robust generative judge model for automatic LLM evaluation
NLP Researcher · Dialogue Model Team · Jan 2022 – Jan 2024
- Constructed instruction datasets for persona-aligned dialogue models
- Developed an emotional state simulator for emotion-aware response generation
NLP Research Intern · Dialogue Model Team · Jan – Aug 2021
- Proposed a control-token-based method for mitigating hate speech in dialogue models