Making today
better than yesterday.

PhD Student. Responsible AI Researcher. Tennis & Dessert Lover.

ABOUT

I am a PhD candidate at Korea Advanced Institute of Science and Technology (KAIST), advised by Professor Steven Euijong Whang.

My research mainly focuses on Responsible AI, especially addressing challenges related to fairness, privacy, and hallucination to enable the safe integration of AI systems into our society.

I am currently at NYU, working with Julia Stoyanovich on responsible data use for LLMs. I previously worked as a research intern at Microsoft Research Asia with wonderful mentors. Always happy to learn and collaborate!

Email: purplehibird (at) kaist (dot) ac (dot) kr

Curriculum Vitae / Google Scholar / LinkedIn

NEWS

Sep 2026: Now in New York as a visiting scholar in Julia Stoyanovich's lab at NYU. Amazing view of NYC from the office 🌆

Aug 2026: I am going EMNLP 2026 to present a new paper on generalizable re-ranking for time-sensitive retrieval! See you in Budapest🇭🇺

July 2026: ACL 2026 was special because Kahee, the first junior labmate I mentored, presented our paper as first author 🎉 So proud of you!

May 2026: Happy to recieve the 🏅Gold Reviewer Recognition (top 25% reviewers) for ICML 2026.

Apr 2026: I attended ICLR 2026 to present a new paper on automatic LLM benchmarking on time-sensitive QA! See you in Rio, Brazil 🇧🇷

Dec 2025: NeurIPS 2025 was huge—so great to meet everyone in person and catch up in San Diego 🇺🇲!

Apr 2025: I attended ICLR 2025 to present a new paper on generative models that are both fair and private! Happy to meet in Singapore 🇸🇬

Dec 2024: I attended NeurIPS 2024 to present a spotlight paper on automatic database-driven LLM hallucination evaluation! Let's go 🇨🇦!

Nov 2024: I gave a talk at the KAIST-Google exploreCSR workshop as a student speaker 🎥, discussing how to bridge databases and LLMs!

Sep 2024: I started my first bigh-tech journey at Microsoft Research Asia (Beijing, China 🇨🇳) as a Research Intern.

PUBLICATIONS

ERBench: An Entity-Relationship based Automatically Verifiable Hallucination Benchmark for LLMs

(NeurIPS'24 Datasets and Benchmarks Track Spotlight, co-first author)

PFGuard: A Generative Framework with Privacy and Fairness Safeguards

(ICLR'25, first author)

Harnessing Temporal Databases for Systematic Evaluation of Factual Time-Sensitive Question-Answering in Large Language Models

(ICLR'26, first author)

EDUCATION

Korea Advanced Institute of Science and Technology (KAIST)
Ph.D Candidate in Electric Engineering
Mar. 2023 - Present
Korea Advanced Institute of Science and Technology (KAIST)
Master Student in Electric Engineering
Mar. 2021 - Feb. 2023
Korea Advanced Institute of Science and Technology (KAIST)
B.S. in Electrical Engineering
Mar. 2016 - Feb. 2021

Page updated

Google Sites

Report abuse

Making today better than yesterday.