09/2026 - Present
Korea Advanced Institute of Science and Technology
M.S. student, Kim Jaechul Graduate School of AI
Research Interests
My research interests lie in inference-time controllability and alignment. I work on methods that steer generative models toward safe, user-intended outputs without expensive retraining. Recently, I have been focusing on safe generation and discrete diffusion, particularly diffusion language models.
Education
02/2019 - 08/2026
Korea Advanced Institute of Science and Technology
B.S. in Industrial and Systems Engineering & Computer Science (Double Major)
08/2025 - 12/2025
University of Portland
Exchange student, Department of Computer Science
03/2017 - 02/2019
Chungnam Science High School
Early graduation
Publications
NeurIPS 2025 Workshop on Generative and Protective AI for Content Creation
Prompt-Based Safety Guidance Is Ineffective for Unlearned Text-to-Image Diffusion Models
Jiwoo Shin, Byeonghu Na, Mina Kang, Wonhyeok Choi, Il-Chul Moon
NeurIPS 2025
Training-Free Safe Text Embedding Guidance for Text-to-Image Diffusion Models
Byeonghu Na, Mina Kang, Jiseok Kwak, Minsang Park, Jiwoo Shin, SeJoon Jun, Gayoung Lee, Jin-Hwa Kim, Il-Chul Moon
Korea Computer Congress, 2022
Nodes-in-Cluster Significance Score (NCSS): Keyword Extraction with Redefined TW-IDF using Stock News Text Data
Junhyeok Jung*, Kwanghyeon Lee*, Jiwoo Shin*, Jieun Han*
* Equal contribution.
Awards
KAIST
IPESK
KAIST ISE
ROKAF, Class 842
KAIST ISE
KAIST ISE
KAIST ISE