Hyochan Chong

I am a research engineer at Samsung Research, where I work on efficient LLMs and agentic systems.

I received my Bachelor's degree in Mechanical Engineering and Artificial Intelligence (double major) from Sungkyunkwan University. Previously, I was a research intern at SKKU IRIS Lab with Jong Hwan Ko, and a software engineer intern at Samsung Research, where I worked on model compression.

I am interested in building systems that are faster, cheaper, and deployable under real hardware constraints. Recently, I have been focusing on low-bit LLM quantization, LLM inference, and multi-agent systems.

Email  /  CV  /  Scholar  /  Twitter  /  Github

Accurate initialization and efficient reconstruction enables 1-bit and sub-1-bit post-training quantization of large language models.

Quantization-aware training that directly addresses residual error from quantization leads to more accurate 2-bit large language models.