PRISM Lab

The goal of our research is to help people better understand, develop, and use AI through human-centered approaches. Working at the intersection of HCI, responsible AI, and data visualization, we study real-world challenges AI practitioners face and build interactive tools and data analysis interfaces to address them. Like a prism that spreads a single beam of light into a spectrum, we strive to help people see AI systems and the data behind them from multiple angles.

• Sept. 2026We are now at Seoul National University. We are recruiting new lab members through the Graduate School of AI.
• Sept. 2026Dasol Choi presented our bias analysis framework at ECCV 2026.
• July 2026Our work led by Dasol Choi on incorporating organizational policies into AI was presented at ACL 2026.
• June 2026Minji Jung presented our work on analyzing and rethinking LLM leaderboards at FAccT 2026 in Montreal.
• June 2026Two collaborative studies with Google and Harvard on text-to-image safety were presented at FAccT 2026.
• May 2026Minsuk Kahng is serving as an Area Chair for NeurIPS 2026.
• Apr. 2026Minjae Lee presented our work on human-in-the-loop AI development workflows at CHI 2026 in Barcelona.
• Apr. 2026Transformer Explainer, our collaborative work with Georgia Tech, was presented at CHI 2026.
• Apr. 2026We started a new IITP-funded research project in collaboration with Korea Univ., SNU, and KAIST.
• Feb. 2026We began a new research collaboration with MLCommons and Google on culturally grounded AI safety.

Recent and representative work, selected from the full publication list. * indicates corresponding authors.

• 
  Data-Prompt Co-Evolution: Growing Test Sets to Refine LLM Behavior

  Minjae Lee and Minsuk Kahng*

  CHI 2026 DOI arXiv PDF

• 
  Transformer Explainer: Learning LLM Transformers with Interactive Visual Explanation and Experimentation

  Aeree Cho, Grace C. Kim, Alexander Karpekov, Seongmin Lee, Alec Helbling, Benjamin Hoover, Zijie J. Wang, Minsuk Kahng*, and Duen Horng Chau*

  CHI 2026 DOI PDF

• 
  Who Defines "Best"? Towards Interactive, User-Defined Evaluation of LLM Leaderboards

  Minji Jung, Minjae Lee, Yejin Kim, Sarang Choi, and Minsuk Kahng*

  FAccT 2026 DOI arXiv PDF

• 
  When Cars Have Stereotypes: Auditing Demographic Bias in Objects from Text-to-Image Models

  Dasol Choi, Jihwan Lee, Minjae Lee, and Minsuk Kahng*

  ECCV 2026 arXiv PDF

• 
  COMPASS: A Framework for Evaluating Organization-Specific Policy Alignment in LLMs

  Dasol Choi, DongGeon Lee, Brigitta Jesica Kartono, Helena Berndt, Taeyoun Kwon, Joonwon Jang, Haon Park, Hwanjo Yu*, and Minsuk Kahng*

  ACL 2026 arXiv PDF

• 
  SafePath: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment

  Wonje Jeung, Sangyeon Yoon, Minsuk Kahng*, and Albert No*

  NeurIPS 2025 arXiv PDF

• 
  LLM Comparator: Interactive Analysis of Side-by-Side Evaluation of Large Language Models

  Minsuk Kahng, Ian Tenney, Mahima Pushkarna, Michael Xieyang Liu, James Wexler, Emily Reif, Krystal Kallarackal, Minsuk Chang, Michael Terry, and Lucas Dixon

  VIS 2024 DOI PDF Blog Code

  Deployed on Google's LLM evaluation platforms. Featured at Google I/O on Gemini Open Models and in the Responsible AI Toolkit.

• 
  Adversarial Nibbler: An Open Red-Teaming Method for Identifying Diverse Harms in Text-to-Image Generation

  Jessica Quaye, Alicia Parrish, Oana Inel, Charvi Rastogi, Hannah Rose Kirk, Minsuk Kahng, Erin van Liemt, Max Bartolo, Jess Tsang, Justin White, Nathan Clement, Rafael Mosquera, Juan Ciro, Vijay Janapa Reddi, and Lora Aroyo

  FAccT 2024 DOI arXiv PDF Blog

• 
  Gemma 2: Improving Open Language Models at a Practical Size

  Google DeepMind Gemma Team

  arXiv preprint, 2024 arXiv

• 
  VLSlice: Interactive Vision-and-Language Slice Discovery

  Eric Slyman, Minsuk Kahng*, and Stefan Lee*

  ICCV 2023 DOI arXiv PDF Talk Website

• 
  Visualizing Linguistic Diversity of Text Datasets Synthesized by Large Language Models

  Emily Reif, Minsuk Kahng*, and Savvas Petridis

  VIS 2023 (Short) DOI arXiv PDF Code

• 
  DendroMap: Visual Exploration of Large-Scale Image Datasets for Machine Learning with Treemaps

  Donald Bertucci, Md Montaser Hamid, Yashwanthi Anand, Anita Ruangrotsakun, Delyar Tabatabai, Melissa Perez, and Minsuk Kahng*

  VIS 2022 DOI arXiv PDF Demo

Director

Minsuk Kahng 강민석

Assistant Professor, Seoul National University

Website Email Google Scholar LinkedIn

Students

• 
  Minjae Lee 이민재

  M.S. student, 2025–present
• 
  Minji Jung 정민지

  M.S. student, 2025–present

Alumni

• Yejin Kim Undergraduate intern, 2025–26
• Sarang Choi Undergraduate intern, 2025–26
• Dasol Choi Project advising, 2025–26; now Researcher at AIM Intelligence
• Eric Slyman Ph.D., 2021–25; now at Adobe Research
• Montaser Hamid Ph.D. student, 2021–22
• Yashwanthi Anand Ph.D. student, 2021–22
• Delyar Tabatabai M.S., 2020–22; now at Apple
• Anita Ruangrotsakun M.S., 2020–22; now at Momento
• Dayeon Oh M.S., 2020–22; now at LG
• Roli Khanna M.S., 2020–21; now at Expedia
• Donny Bertucci B.S., 2020–22; started as a Ph.D. student at Georgia Tech, now at Axiom Bio
• Mark Ser B.S., 2020–21; now at Microsoft
• Kristina Lee B.S., 2020–21; now at Oracle
• Thuy-Vy Nguyen B.S., 2020–21; now at Oracle

We are actively recruiting graduate students and undergraduate interns, including students starting in Spring 2027 at the Graduate School of AI. We welcome students with different strengths and backgrounds. If you're interested, please email Minsuk Kahng your CV, transcript, and a short statement. For more information, please refer to this webpage written in Korean.