tanford CME295 Transformers & LLMs is a graduate-level course designed to provide a deep understanding of modern AI architectures, focusing on Transformers and Large Language Models (LLMs). The Autumn 2025 lecture series guides participants through foundational concepts, advanced reasoning, evaluation methods, and practical tuning of LLMs for real-world applications.
Students learn about the mechanics of Transformers, attention mechanisms, and how these architectures enable powerful natural language understanding. The course explores LLM evaluation metrics, reasoning techniques, and agentic capabilities, helping learners understand both the theoretical and applied aspects of AI. Practical insights on LLM fine-tuning equip participants with strategies to customize models for specialized tasks. Lectures also cover current trends, ethical considerations, and emerging research in the field of AI.
Through in-depth lectures and practical exercises, students gain a comprehensive view of LLM capabilities, limitations, and potential applications. By the end of the course, participants are well-equipped to analyze, deploy, and optimize Transformer-based models, making it ideal for AI researchers, engineers, and advanced students seeking to advance their expertise in cutting-edge NLP