Yubin Choi

AssurAI: Experience with Constructing Korean Socio-cultural Datasets to Discover Potential Risks of Generative AI

Chae-Gyun Lim, Seung-Ho Han, Jeongyun Han, Soohyun Cho et autres

The rapid evolution of generative AI necessitates robust safety evaluations. However, current safety datasets are predominantly English-centric, failing to capture specific risks in non-English, socio-cultural contexts such as Korean, and are often limited to the text modality. To address this gap, we …

Don’t Miss the Forest for the Trees: Attentional Vision Calibration for Large Vision Language Models

Sangmin Woo, Donguk Kim, Jae‐Hyuk Jang, Yubin Choi et autres

Large Vision Language Models (LVLMs) demonstrate strong capabilities in visual understanding and description, yet often suffer from hallucinations-attributing incorrect or misleading features to images.We observe that LVLMs disproportionately focus on a small subset of image tokens-termed blind tokenswhich are typically irrelevant to …

ca (code pays fourni par la source)

Music Emotion Recognition Using Hierarchical Contrastive Learning

Yubin Choi, E. Cho Smith, Mia Y. Wang

The Music Emotion Recognition (MER) task has garnered significant attention from both academic and industrial fields due to its versatility in various fields, such as music recommendation systems and psychotherapy. Due to advancements in language model performance, there has been a rise …

kr, us (code pays fourni par la source)

RITUAL: Random Image Transformations as a Universal Anti-hallucination Lever in Large Vision Language Models

Sangmin Woo, Jae‐Hyuk Jang, Donguk Kim, Yubin Choi et autres

Recent advancements in Large Vision Language Models (LVLMs) have revolutionized how machines understand and generate textual responses based on visual inputs, yet they often produce "hallucinatory" outputs that misinterpret visual information, posing challenges in reliability and trustworthiness. We propose RITUAL, a simple …

Don't Miss the Forest for the Trees: Attentional Vision Calibration for Large Vision Language Models

Sangmin Woo, Donguk Kim, Jae‐Hyuk Jang, Yubin Choi et autres

Large Vision Language Models (LVLMs) demonstrate strong capabilities in visual understanding and description, yet often suffer from hallucinations, attributing incorrect or misleading features to images. We observe that LVLMs disproportionately focus on a small subset of image tokens--termed blind tokens--which are typically …

DEVELOPMENT OF A MULTI-DIMENSIONAL ARRAY DATABASE BASED MASSIVE SATELLITE INFORMATION PROCESSING AND ANALYSIS SYSTEM: KIWI-SAT

S. Han, J. Han, Yubin Choi, S. Lee et autres

Abstract. Information and communication technology (ICT) is mainly applied to finance, telecommunications, and public sectors. However, since the early 2010s, there have been efforts to apply ICT to various fields such as aerospace, life science, energy, and automobiles. Recently, artificial intelligence and …

kr (code pays fourni par la source)