I am a Researcher working on Multimodal Large Language Models (MLLMs), with a particular focus on visual token efficiency and efficient deep learning architectures for practical computing platforms.

With over a decade of research experience—including a Ph.D. from Yonsei University (2018) under the supervision of Prof. Jong-Seok Lee at the MCML Group—my technical background spans perceptual image and video quality processing, AI-based camera ISP, generative vision models, and multimodal architectures. By bridging deep expertise in perceptual visual processing with efficient multimodal AI, I strive to advance high-fidelity, resource-conscious intelligent systems.

Research Interests

• Multimodal Large Language Models (MLLMs)
• Visual Token Efficiency
• Efficient Deep Learning Architectures
• AI-based Camera ISP & Perceptual Image/Video Processing
• Perceptual Quality Assessment (IQA / VQA)
• Generative Vision Models