Chanjun Park

Introducing a method that enhances machine translation by considering component relationships.

New resources enhance assessment of Korean language models.

Study investigates performance trends of Korean language models over eleven months.

A new approach combines two KenLM models for better data filtering.

Revolutionize dataset building with the LP Data Pipeline on regular CPUs.

Discover how AI models tackle coding challenges and their internal mechanics.

Explore the challenges and advancements in Large Language Models' instruction-following abilities.