About Me
About Me
Embedded Files
I am a Ph.D. student majoring in Computer Science at Yonsei University, advised by Prof. Sanghyun Park. Previously, I received M.S. (2022-2024) and B.S. (2017-2022) degrees from Gachon University, advised by Prof. Chang Choi.
My previous research topics are adversarial attack, medical image analysis.
My current interest centers around Neural Network compression (Quantization).
Education
Education
Ph.D. in Computer Science, Yonsei Univ. 2024 - Present
M.S. in Computer Engineering, Gachon Univ. 2022 - 2024
B.S. in Computer Engineering, Gachon Univ. 2017 - 2022
Publications
Publications
TrajeQ: Trajectory-Preserving Flow Matching Quantization for Low-bit Video Generation
Inpyo Hong, Youngwan Jo, Hyojeong Lee, Jeongeun Kim, Seungyeon Choi, Sanghyun Park
The Fortieth Anuual Conference on Neural Information Processing Systems (NeurIPS), 2026 - Sydney
(Under-Review) (Project Page)
Learning Canonical Representations for Unified 3D Molecular Modeling
Seungyeon Choi, Hwanhee Kim, Yoonju Kim, Seungyong Lee, Inpyo Hong, Sanghyun Park
The Fortieth Anuual Conference on Neural Information Processing Systems (NeurIPS), 2026 - Sydney
(Under-Review)
The ”Why” Behind the ”What” : Enhancing Compositional Understanding in Referring Segmentation
Jeongeun Kim, Youngwan Jo, HyoJeong Lee, Inpyo Hong, Sanghyun Park
The 19th European Conference on Computer Vision (ECCV), Under Review
(Under Review)
Diversity-Driven Pseudo Anomaly Generation for Self-Supervised Video Anomaly Detection
Kijung Lee, Youngwan Jo, Inpyo Hong, Sanghyun Park
The 19th European Conference on Computer Vision (ECCV), Under Review
(Under Review)
Segment anything model guided dual-mask framework for anatomically faithful medical image translation
Hyojeong Lee, Youngwan Jo, Inpyo Hong, Jeongeun Kim, Sanghyun Park
Scientific Reports, 2026
GranQ: Efficient Channel-wise Quantization via Vectorized Pre-Scaling for Zero-shot QAT
Inpyo Hong, Youngwan Jo, Hyojeong Lee, Sunghyun Ahn, Kijung Lee, Sanghyun Park
The 41th ACM/SIGAPP Symposium On Applied Computing (ACM SAC), 2026 - Greece
SKIM: Semantic Knowledge Infused Modeling for Medical Report Generation
Hyojeong Lee, Youngwan Jo, Inpyo Hong, Sanghyun Park
The 41th ACM/SIGAPP Symposium On Applied Computing (ACM SAC), 2026 - Greece
(Accepted)
Exploring Synergy of Denoising and Distillation: Novel Method for Adversarial Defense
Knowledge distillation vulnerability of DeiT through CNN adversarial attack
Security Verification Software Platform of Data-efficient Image Transformer Based on Fast Gradient Sign Method
Inpyo Hong, Gyuho Choi, Pankoo Kim, Chang Choi
The 38th ACM/SIGAPP Symposium On Applied Computing (ACM SAC), 2023 - Estonia
Data Augmentation Based on Generative Adversarial Networks for Endoscopic Image Classification
Hyuncheol Park*, Inpyo Hong*,Sahadeve Poudel, Chang Choi
(* = Equal Contribution)
IEEE Access, 2023
Projects
Projects
Development of Ultra-Efficient Lightweight AI Models for Collaborative Document Workflows Enabling Digital Sovereignty - Institute for Information & Communication Technology Planning & evaluation (IITP)
I developed a multi-teacher knowledge distillation framework to build ultra-efficient, lightweight AI models for collaborative document workflows.
Data Engineering Lab, 2026.05 ~ Current
Fundamental Research Program for AI-based Future Climate Technology Development - National Research Foundation of Korea (NRF)
I contributed to optimizing and designing a lightweight model for detecting abnormal patterns related to corrosion and degradation in building structures.
Data Engineering Lab, 2025.04 ~ Current
Korean AI-based High-quality Bio Big Data Analysis Platform Project for Intractable Diseases - Ministry of Science and ICT (MSIT)
I was involved in the optimization and design of AI models to detect clinically significant patterns in Korean bio big data for the analysis of rare and intractable diseases.
Intelligent Information Processing Lab, 2023.08 ~ 2024.02
Development of an Adversarial Attack-based Intelligent Security System for Obstacle Detection in Autonomous Vehicles - Ministry of Science and ICT (MSIT)
I was involved in identifying security weaknesses and implementing adversarial defense strategies for intelligent obstacle detection systems in autonomous driving.
Intelligent Information Processing Lab, 2022.12 ~ 2024.02
Development of a High-reliability Multi-modal Biometric Authentication System Based on Knowledge Reasoning and Deep Learning for Robust Security Authentication - Ministry of Science and ICT (MSIT)
I contributed to the analysis of security weaknesses and the implementation of adversarial defense mechanisms in multi-modal biometric authentication systems, utilizing knowledge inference and deep learning to ensure robust and reliable security.
Intelligent Information Processing Lab, 2021.08 ~ 2024.02
Awards & Honors
Awards & Honors
Encouragement Award (2024)
Smart City Platform Idea Competition, Yonsei University, Republic of KoreaBest Paper Award (2023)
Korean Institute of Smart Media 2023 General Academic ConferenceStudent Travel Award (2023)
The 38th ACM/SIGAPP Symposium on Applied ComputingCum Laude (2023)
Gachon University, Republic of KoreaAcademic Scholarship (2017, 2021)
Gachon University, Republic of Korea
Reviewer Services
Reviewer Services
Conference
The 40th Annual AAAI Conference on Artificial Intelligence (AAAI 2026)
The 41th ACM/SIGAPP Symposium On Applied Computing (ACM SAC 2026)
Forty-Third International Conference on Machine Learning (ICML 2026, Gold Reviewer(Top 25%))
The Fortieth Annual Conference on Neural Information Processing Systems (NeurIPS 2026)
Journal
Future Generation Computer Systems (2024~Present)
Knowledge-Based Systems (2025~Present)
Pattern Recognition (2025~Present)
Scientific Reports (2026~Present)
Page updated
Google Sites
Report abuse