Available for collaboration

Yejin Lee

NLP ยท LLM Safety ยท Hate Speech Detection

Researching LLM safety with a focus on building safer, more trustworthy language models. Interested in jailbreaking and robustness of LLMs against adversarial inputs, as well as implicit hate speech detection as a core safety challenge.

ssgyejin@gmail.com GitHub Blog
>_ console.log("something is hiding here...") F12

01 Education

2024 -

Yonsei University

Integrated M.S./Ph.D. Program in Computer Science

Enrolled
2021 - 2024

Sungkyunkwan University

B.S. in Computer Science and Engineering

Graduated

02 Experience

LG Electronics Jan 2024 - May 2024

R&D Software Engineer

Crois Sep 2023 - Dec 2023

AI Research Intern

NAVER Connect Foundation Jan 2022 - Jun 2022

3rd Boostcamp AI Tech ยท NLP Track

Sungkyunkwan Univ & Shinhan Bank May 2021 - Dec 2021

Industry-Academic Collaboration Project

03 Publications

ACL 2026

RV-HATE: Reinforced Multi-Module Voting for Implicit Hate Speech Detection

Y Lee, H Ahn, YS Han

GitHub
EMNLP 2025

AmpleHate: Amplify target-context relationships for implicit hate speech detection

Y Lee, J Hahn, H Ahn, YS Han

GitHub
Under Review

Adaptive Steering and Remasking for Safe Generation in Diffusion Language Models

Y Lee, YS Han

Under Review

KOTOX: A Korean Toxic Dataset for Deobfuscation and Detoxification

Y Lee, SH Kim, H Jin, D Kim, Y Kim, YS Han

Under Review

How Does the Thinking Step Influence Model Safety? An Entropy-based Safety Reminder for LRMs

SH Kim, H Jin, Y Lee, YS Han

04 Awards

๐Ÿ†

3rd K-AI Manufacturing Data Analysis Competition

2023 ยท KAIST, Ministry of SMEs and Startups, Smart Manufacturing Innovation Promotion Group

Grand Prize
๐Ÿฅ‡

1st 9oormthon

2022 ยท kakao, goorm

Grand Prize
โญ

Outstanding Open Source Software Student

2022, 2023 ยท College of Software, Sungkyunkwan Univ

Encouragement Award
๐ŸŽ–๏ธ

2023 AI HACKATHON

Udemy Curation, Wanted, JoCoding

3rd Place
๐Ÿ“„

2023 Spring Research Paper & Project Presentation

College of Software, Sungkyunkwan Univ

Best Presentation
๐Ÿ“š

4th Undergraduate AI x BOOKATHON

2023 ยท Sungkyunkwan Univ, Industry-Academic-Research Innovation Council

Encouragement Award
๐Ÿ’ก

Jeju ICT Innovation Square Idea Competition

2022 ยท Jeju Industry-University Cooperation Foundation

Encouragement Award

05 Patents

US

Implicit Hate Speech Detection Apparatus Based on Target-Context Relation Amplification

19/551,450 ยท ์ถœ์›

KR

๋‹ค์ค‘๋ชจ๋“ˆ ๊ฐ•ํ™”ํ•™์Šต์„ ํ†ตํ•œ ์•”์‹œ์  ํ˜์˜ค ํ‘œํ˜„ ํƒ์ง€ ์žฅ์น˜ ๋ฐ ๋ฐฉ๋ฒ•

10-2025-0200617 ยท ์ถœ์›

KR

๋Œ€์ƒ-๋ฌธ๋งฅ ๊ด€๊ณ„ ์ฆํญ ๊ธฐ๋ฐ˜ ์•”์‹œ์  ํ˜์˜ค ํ‘œํ˜„ ํƒ์ง€ ์žฅ์น˜ ๋ฐ ๋ฐฉ๋ฒ•

10-2025-0194914 ยท ์ถœ์›

KR

The Caring Seat in a Public Transportation

10-2020-0000289 ยท ์ถœ์›