Grace C. Kim

4th-year BS/MS Student at Georgia Tech (HCI + ML)

My research focuses on improving the interpretability and trustworthiness of ML models. I develop interactive visualization tools for visually analyzing model decision-making, and interpretability methods for understanding and controlling model behavior.

Education

Aug. 2025 — May 2026
M.S. in Computer Science
Georgia Institute of Technology, Atlanta, GA
Advisor: Duen Horng (Polo) Chau , Co-advisor: Mark Riedl
Concentration in Machine Learning
Stamps President's Scholarship (full-ride merit scholarship)
Aug. 2022 — May 2025
B.S. in Computer Science
Georgia Institute of Technology, Atlanta, GA
Thesis: Transformer Explainer: Interactive Visual Learning of LLM Transformers
Advisor: Duen Horng (Polo) Chau , Co-advisor: Betsy DiSalvo (Mentor: Annabel Rothschild)
Concentrations in Artificial Intelligence and Theory, Overall GPA: 3.9/4.0, Highest Honors
Stamps President's Scholarship (full-ride merit scholarship)
Summer 2023
Berlin Summer Program
Technische Universität Berlin, Germany
Study abroad program by Georgia Tech College of Computing

Research Experience

Jan. 2024 — Present
Georgia Institute of Technology, Atlanta, GA
Research Assistant (now Graduate Assistant), Polo Club of Data Science
Advisor: Duen Horng (Polo) Chau
Study the intersection of machine learning, human-computer interaction, and data visualization. Published at EMNLP 2025 Main, AAAI 2025 Demo, and IEEE VIS 2024 Poster.
May 2025 — Present
Georgia Institute of Technology, Atlanta, GA
Research Assistant, Entertainment Intelligence and Human-Centered AI Labs
Advisor: Mark Riedl
Study explainable AI and mechanistic interpretability.
May 2024 — Aug. 2024
CISPA Helmholtz Center for Information Security, Saarbrücken, Germany
DAAD RISE Research Intern, SprintML Lab
Advisor: Adam Dziedzic, Franziska Boenisch
Studied memorization in CLIP models. Published at ICLR 2025.
Aug. 2023 — Oct. 2024
Georgia Institute of Technology, Atlanta, GA
Research Assistant, Culture and Technology Lab
Advisor: Betsy DiSalvo , Mentor: Annabel Rothschild
Studied critical data studies and responsible AI. Published at AIES 2024.

Honors and Awards

2022
Stamps President's Scholarship, Georgia Tech
4-year full-ride merit scholarship with $15,000 in extracurricular funding awarded to the top 0.1% of applicants (40 students) to Georgia Tech
2024
Best Poster, IEEE VIS
For "Transformer Explainer: Interactive Learning of Text-Generative Models"
2025
Best Submission, Workshop on Visualization for AI Explainability at IEEE VIS
For Transformer Explainer
2024
DAAD RISE Scholarship, German Federal Foreign Office
Selected as one of 320 scholarship recipients from 2,358 applicants to conduct a summer research internship in Germany
2025
President's Undergraduate Research Awards Travel Award, Georgia Tech
For "Transformer Explainer: Interactive Learning of Text-Generative Models"
2024
President's Undergraduate Research Awards Travel Award, Georgia Tech
For "What's Your Stake in Sustainability of AI?: An Informed Insider's Guide"

Publications

4
Transformer Explainer: Interactive Learning of Text-Generative Models
Aeree Cho*, Grace C. Kim*, Alexander Karpekov*, Alec Helbling, Zijie J. Wang, Seongmin Lee, Benjamin Hoover, Duen Horng (Polo) Chau
IEEE Visualization Conference (VIS Poster). 2024., AAAI Conference on Artificial Intelligence (AAAI Demo). 2025.
Demo PDF Recording Code BibTeX Best Poster *Authors contributed equally
410,000+ users across 180+ countries, 6,000+ GitHub stars
#1 paper of the day on Hugging Face
Taught at top universities (e.g., MIT, Columbia, University of Toronto) and in Georgia Tech CSE 6242 Data and Visual Analytics (250-300 students per semester)
3
Interpretation Meets Safety: A Survey on Interpretation Methods and Tools for Improving LLM Safety
Seongmin Lee, Aeree Cho, Grace C. Kim, ShengYun Peng, Mansi Phute, Duen Horng (Polo) Chau
Conference on Empirical Methods in Natural Language Processing (EMNLP Main). 2025.
PDF BibTeX
2
Captured by Captions: On Memorization and its Mitigation in CLIP Models
Wenhao Wang, Adam Dziedzic, Grace C. Kim, Michael Backes, Franziska Boenisch
International Conference on Learning Representations (ICLR). 2025.
PDF BibTeX
1
What's Your Stake in Sustainability of AI?: An Informed Insider's Guide
Grace C. Kim, Annabel Rothschild, Carl DiSalvo, Betsy DiSalvo
AAAI/ACM Conference on AI, Ethics, and Society (AIES). 2024.
PDF BibTeX

Grants and Funding

2025
Google Gift ($75,000)
Co-PIs: Duen Horng (Polo) Chau
Co-first authored a proposal with one graduate student resulting in a $75,000 unrestricted gift from Google

Conference Travels & Presentations

2025
IEEE Visualization Conference (VIS) 2025
2025
AAAI Conference on Artificial Intelligence (AAAI) 2025
For "Transformer Explainer: Interactive Learning of Text-Generative Models"
2024
AAAI/ACM Conference on AI, Ethics, and Society (AIES) 2024
For "What's Your Stake in Sustainability of AI?: An Informed Insider's Guide"
2024
IEEE Visualization Conference (VIS) 2024
For "Transformer Explainer: Interactive Learning of Text-Generative Models"

Teaching

Spring 2023
Teaching Assistant
Georgia Institute of Technology, Atlanta, GA
Object-Oriented Programming (CS 1331), Instructor: Richard Landry
Led weekly recitation classes for 50 students, held weekly office hours, and created and graded course materials for a course with 800 students enrolled.

Service

May 2025 — Present
Founding Member
Georgia Institute of Technology, Atlanta, GA
Digital Equity & Empowerment Project
Design and lead workshops on AI safety and responsible use for Atlanta-based youths
Sep. 2022 — Apr. 2023
Web Developer & Administrator
Georgia Institute of Technology, Atlanta, GA
Stamps Scholars National Convention
Managed website development and administration for a convention with 700 attendees from 40 schools

References

Dr. Polo Chau, Professor
School of Computational Science and Engineering
Georgia Institute of Technology
Dr. Mark Riedl, Professor
School of Interactive Computing
Georgia Institute of Technology
Dr. Annabel Rothschild, Assistant Professor
Computer Science
Bard College