Kim Do Hyung

Dohyung Kim

AI Researcher @ Samsung Electronics AI Center

LLM/VLM | RAG | Agent | Computer Vision

Google Scholar GitHub Email

About Me

I am an AI Researcher at Samsung Electronics AI Center, working on large language models and vision-language models. I received my Ph.D. in Electrical and Electronic Engineering from Yonsei University, where I was advised by Prof. Bumsub Ham at the Computer Vision Lab.

My current research focuses on LLM/VLM training, retrieval-augmented generation (RAG), and AI agents. Previously, I worked on efficient deep learning for computer vision, including network quantization, semantic correspondence, and image super-resolution.

Research Interests

Current

LLM / VLM Training

Training and fine-tuning large language models and vision-language models for real-world applications

Retrieval-Augmented Generation

Enhancing LLM outputs with external knowledge retrieval for accurate and grounded responses

AI Agents

Building autonomous AI agents that reason, plan, and interact with tools and environments

Previous

Network Quantization

Compressing deep neural networks through low-bit quantization for efficient deployment

Semantic Correspondence

Establishing dense correspondences between semantically similar images

Super-Resolution

Recovering high-resolution images from low-resolution inputs using deep learning

Publications

(* indicates equal contribution)  |  Google Scholar

Journal

PR 2026

Token-Based Dynamic Bit-Width Assignment for ViT Quantization

D. Kim*, J. Moon*, J. Lee, G. Lee, J. Jeon, B. Ham

Pattern Recognition, vol. 171, no. 3, pp. 112269, Mar. 2026

TPAMI 2022

Learning Semantic Correspondence Exploiting an Object-level Prior

D. Kim*, J. Lee*, W. Lee, J. Ponce, B. Ham

Paper Project
ESSCIRC 2022

SIF-NPU: A 28nm 3.48 TOPS/W 0.25 TOPS/mm2 CNN Accelerator with Spatially Independent Fusion for Real-Time UHD Super-Resolution

S. Lee, K.-B. Lee, S. Joo, H. K. Ahn, J. Lee, D. Kim, B. Ham

Paper

Conference

ECCV 2024

Toward INT4 Fixed-Point Training via Exploring Quantization Error for Gradients

D. Kim, J. Lee, J. Jeon, J. Moon, B. Ham

Paper
CVPR 2024

Instance-Aware Group Quantization for Vision Transformers

J. Moon, D. Kim, J. Cheon, B. Ham

Paper Project
ICCV 2023

Camera-Driven Representation Learning for Unsupervised Domain Adaptive Person Re-identification

G. Lee, S. Lee, D. Kim, Y. Shin, Y. Yoon, B. Ham

Paper Project
ICCV 2021

Distance-aware Quantization

D. Kim, J. Lee, B. Ham

Paper Project
CVPR 2021

Network Quantization with Element-wise Gradient Scaling

J. Lee, D. Kim, B. Ham

Paper Project
ECCV 2020

Learning with Privileged Information for Efficient Image Super-Resolution

D. Kim*, W. Lee*, J. Lee*, B. Ham

Paper Project
CVPR 2019 Oral

SFNet: Learning Object-aware Semantic Correspondence

D. Kim*, J. Lee*, J. Ponce, B. Ham

Paper Project Video

Experience

AI Researcher

Samsung Electronics, AI Center

Jan. 2025 – Present

LLM/VLM RAG, Agent, Training

AI Researcher

Samsung Electronics, SAIT

Sep. 2024 – Dec. 2024

Education

Ph.D. in Electrical and Electronic Engineering

Yonsei University

Mar. 2018 – Aug. 2024

Advisor: Prof. Bumsub Ham

B.S. in Electrical and Electronic Engineering

Yonsei University

Awards

Silver Prize

25th Samsung Human-Tech Paper Award
Signal Processing

Excellence Award

Graduate Student Paper Contest
Yonsei University, 2020