Jaeseok Byun

Region-Aware Text-to-Image Retrieval with Efficient Region-Informed Sequential Query Embedding

Seokhyeon Jeong*, Jaeseok Byun*, Paul Hongsuck Seo, and Taesup Moon

*: Equal contribution

Multi-modal Knowledge Preserving Adapter for Embedding Backward Compatibility

Jaeseok Byun, Gukyeong Kwon, Han-Kai Hsu, Meher Gitika Karumuri, Zhikang Zhang, Hao Yang, and Davide Modolo

Work done during internship at Amazon AGI

KVgrad: Query-Agnostic KV Cache Eviction via Gradient-based Global Importance Scoring

Jihwan Kwak*, Sunghwan Joo*, Jungyoon Hwang, Jaeseok Byun, and Taesup Moon

AdaptFM: Resource-Adaptive Foundation Model Inference. *: Equal contribution

Efficient and Robust SEM Image Denoising for Wafer Defect Inspection

Hyungwoong Bae, Jaeseok Byun, Yongwoo Lee, and Taesup Moon

MA-CIR: Multi-modal Arithmetic Benchmarks for Composed Image Retrieval

Jaeseok Byun, Youngkyun Jang, Seokhyeon Jeong, Donghyun Kim, and Taesup Moon

RTD: Reducing Task Discrepancy of Text Encoders for Zero-Shot Composed Image Retrieval

Jaeseok Byun*, Seokhyeon Jeong*, Wonjae Kim, Sanghyuk Chun, and Taesup Moon

*: Equal contribution

MAFA: Managing False Negatives for Vision-Language Pre-training

Jaeseok Byun*, Dohoon Kim*, and Taesup Moon

*: Equal contribution

GRIT-VLP: Grouped Mini-batch Sampling for Vision-Language Pre-training

Jaeseok Byun*, Taebaek Hwang*, Jianlong Fu, and Taesup Moon

Work done during internship at Microsoft Research Asia. *: Equal contribution

FBI-Denoiser: Fast Blind Image Denoiser for Source-Dependent Noise

Jaeseok Byun*, Sungmin Cha*, and Taesup Moon

Oral presentation. *: Equal contribution

Learning Blind Pixelwise Affine Image Denoiser with Single Noisy Images

Jaeseok Byun and Taesup Moon

IEEE Signal Processing Letters