Jiseon Kim

About

Hi, I am Jiseon Kim, a Ph.D. candidate in User & Information Lab (U&I Lab), KAIST, advised by Alice Oh. My research interests lie in natural language processing (NLP) and computational social science (CSS), with a focus on 1) AI alignment with human and societal values and 2) AI for social good.
In particular, I work on the following topics:

• LLM-Human Alignment & Evaluation: I explore LLM alignment with human values and society, examining their behaviors and limitations (e.g., moral decision making, cultural bias, social reasoning).
• AI for Science & Social Impact: I develop AI frameworks to process large-scale, expertise-driven data, particularly in political science (e.g., legislative processes, lobbying), to uncover hidden dynamics, enhance transparency, and assess societal impact.

Keywords: AI Alignment, LLM Evaluation, AI for Policy & Governance, AI for Social Good, NLP, Computational Social Science

Education

• Korea Advanced Institute of Science and Technology, Ph.D., School of Computing, 2020 - Present
• Korea Advanced Institute of Science and Technology, M.S., School of Computing, 2017 - 2019
• Sookmyung Women's University, B.S., Computer Science, 2013 - 2017

Experience

• Visiting Researcher @ MIT: Summer 2019,2022,2024
• Research Intern @ NAVER AI Lab: 03/2023 - 06/2023
• Researcher @ KAIST: 03/2019 - 02/2020

Publication

Perceptions to Beliefs: Exploring Precursory Inferences for Theory of Mind in Large Language Models
EMNLP 2024
Chani Jung, Dongkwan Kim, Jiho Jin, Jiseon Kim, Yeon Seonwoo, Yejin Choi, Alice Oh, Hyunwoo Kim
PDF

HyperCLOVA X Technical Report
Technical Report 2024
Kang Min Yoo, Jaegeun Han, Sookyo In, Heewon Jeon, ... Jiseon Kim...(additional authors)
PDF  Webpage

KoBBQ: Korean Bias Benchmark for Question Answering
TACL 2024, Present at ACL 2024
Jiho Jin*, Jiseon Kim*, Nayeon Lee*, Haneul Yoo*, Alice Oh, Hwaran Lee [*]equal contribution
PDF  Webpage

Learning Bill Similarity with Annotated and Augmented Corpora of Bills
EMNLP 2021
Jiseon Kim, Elden Griggs, In Song Kim, Alice Oh
PDF   Code

Efficient Contrastive Learning via Novel Data Augmentation and Curriculum Learning
EMNLP 2021
Seonghyeon Ye, Jiseon Kim, Alice Oh
PDF   Code

Dimensional emotion detection from categorical emotion
EMNLP 2021
Sungjoon Park, Jiseon Kim, Seonghyeon Ye, Jaeyeol Jeon, Hee Young Park, Alice Oh
PDF   Code

Denoising recurrent neural networks for classifying crash-related events
IEEE transactions on intelligent transportation systems 2019
Sungjoon Park, Yeon Seonwoo, Jiseon Kim, Jooyeon Kim, Alice Oh
PDF  Code

Workshop

Exploring Persona-dependent LLM Alignment for the Moral Machine Experiment
BiAlign @ ICLR 2025
Jiseon Kim*, Jea Kwon*, Luiz Felipe Vecchietti*, Alice Oh, Meeyoung Cha[*]equal contribution

Understanding Lobbying Strategies in Legislative Process: Bill Position Dataset and Lobbying Analysis
WiML@NeurIPS 2024
Jiseon Kim, Dongkwan Kim, Joohye Jeong, In Song Kim, Alice Oh
Poster

KoBBQ: Korean Bias Benchmark for Question Answering
C3NLP, Co-located with ACL 2024
Jiho Jin*, Jiseon Kim*, Nayeon Lee*, Haneul Yoo*, Alice Oh, Hwaran Lee [*]equal contribution
PDF   Webpage

Preprint

Uncovering Factor Level Preferences to Improve Human-Model Alignment
Arxiv 2024
Juhyun Oh*, Eunsu Kim*, Jiseon Kim, Wenda Xu, Inha Cha, William Yang Wang, Alice Oh [*]equal contribution
PDF

Invited Talk

Uncovering the Hidden Politics of Lawmaking: How Bills and Lobbying Shape U.S. Policy
ExploreCSR@Google March 21, 2025 (KAIST)
Presented on AI for Political Science to understand the legislative process, supported by Google and hosted by KAIST School of Computing

LLMs and the Political-Cultural Lens in Social Science
MPI-SP@Germany Feb 25, 2025 (Max Planck Institute)
Invited talk at Max Planck Institute for Security and Privacy, hosted by Prof. Meeyoung Cha (Data Science for Humanity)

Things I Wish I Had Known Earlier in Grad School
MLAI@Yonsei Jan 2, 2025 (Yonsei University)
Invited talk on networking, self-promotion, and collaboration in academia, hosted by Prof. Kyungwoo Song at the Machine Learning and Artificial Intelligence (MLAI) Lab

Academic Service

• Reviewer
  Feb ACL Rolling Review (ARR) 2025
  Workshop on Bidirectional Human-AI Alignment @ ICLR 2025
  Feb/Apr/June ACL Rolling Review (ARR) 2024
• Volunteer
  FAccT 2022
  COLING 2022
• Undergraduate Research Program @ KAIST
  Spring 2024 (Received an Encouragement Award)
• Individual Research Mentoring @ KAIST
  Spring 2024, Fall 2024
  Spring 2022, Fall 2023
  Spring 2021, Fall 2021
  Spring 2020, Fall 2020

Award & Funding

• 2024 KAIST Graduate Student Outstanding Paper Award
  KAIST, Oct 2024
  Awarded for "KoBBQ: Korean Bias Benchmark for Question Answering"
• MISTI Global Seed Funds
  MIT, Dec 2019
  MIT’s Global Seed Funds facilitate international collaborations for addressing global challenges

Teaching

• TA of Machine Learning for NLP (Fall 2021)
• TA of Machine Learning for NLP (Spring 2021)
• TA of Advanced Data Mining (Fall 2020)
• Head TA of Artificial Intelligence and Machine Learning (Spring 2020)
• TA of Data Structure (Fall 2018)
• TA of Data Structure (Spring 2018)
• TA of Data Structure (Fall 2017)

Reference

Contact

Jiseon Kim
jiseon_kim (at) kaist.ac.kr
Ph.D. Candidate
KAIST