← Search

Yoohoon Kang

2 accepted papers

2026

MuCo: Multi-turn Contrastive Learning for Multimodal Embedding Model

CVPR 2026

Universal Multimodal embedding models built on Multimodal Large Language Models (MLLMs) have traditionally employed contrastive learning, which aligns representations of query-target pairs across different modalities. Yet, despite its empirical success, they are primarily built on a "single-turn" fo

Cited by 0SourcecodeScholar
2024

Language-only Training of Zero-shot Composed Image Retrieval

CVPR 2024poster

Composed image retrieval (CIR) task takes a composed query of image and text aiming to search relative images for both conditions. Conventional CIR approaches need a training dataset composed of triplets of query image query text and target image which is very expensive to collect. Several recent wo…