Foundations of Large Language Models. 3 credits, 3 contact hours

Catalog description

Prerequisites: DS 675 or DS 680 or approval of instructor. Although large language models (LLMs) are widely used in daily life (e.g., ChatGPT), many people lack a comprehensive understanding of their foundations. While using AI tools is easy, developing and improving them requires a deeper understanding of how they work. This course provides a foundational introduction to modern LLMs, covering their architectures, training principles, and post-training techniques. Students will learn why current LLMs are so powerful and how they are built, optimized, and aligned for real-world use. The course also offers an opportunity to conduct research and explore emerging directions in LLMs and their interdisciplinary applications.