4 papers
Towards Clinically Faithful Medical Image Captioning via Enhanced Vision-Language Alignment
Yunseo Lee, Hyun Jun Kim, Heeseung Shin +1
Medical image captioning is a technique that accelerates early-stage diagnostic workflows and enhances the interpretability of medical diagnostic AI systems. However, unlike genera…
Clinically Structured Surrogate Rewards for Post-SFT Medical Image Captioning
Hyun Jun Kim, Heeseung Shin, Changwon Lim
Medical image captioning requires translating heterogeneous visual evidence into concise clinical descriptions, where errors in findings, assertion states, or anatomical relations…
Continuity-Driven Representation Learning for Industrial Defect Detection
Minjong Kim, Hyun Jun Kim, Jeongrae Kim +2
Industrial defect detection differs from natural-image object detection because inspection images are captured under controlled conditions and contain large normal-dominant regions…
AISTAT lab system for DCASE2025 Task6: Language-based audio retrieval
Hyun Jun Kim, Hyeong Yong Choi, Changwon Lim
This report presents the AISTAT team's submission to the language-based audio retrieval task in DCASE 2025 Task 6. Our proposed system employs dual encoder architecture, where audi…