Introduction to LLMs - An overview ST-LE-2026-18
by
Room 2.2070/2.2090, 2nd floor + Zoom
hybrid
Background
LLMs have transformed AI by enabling machines to understand and generate human-like language, driving new capabilities across multiple fields. They are now essential tools for information processing, automation, and decision-making in science and everyday applications, making their understanding important not only for researchers but for anyone interacting with AI-driven technologies, as they become more common and part of our daily lives.
Learning contents
This course provides an introduction to large language models (LLMs), combining intuition on key theoretical concepts with an overview of their main capabilities. We begin by examining how these models’ function, covering fundamental ideas such as tokenization, embeddings, and the transformer architecture, and introducing essential terminology used across the field. Further, the lecture will showcase the usage of LLMs across use cases, presenting the range of applications that LLMs can address. Learning about these individual language model capabilities should pave the basic introduction to problem-solving with LLMs.
Learning objectives
By the end of the session, participants will have a general understanding of both the internal principles and the application landscape of LLMs, enabling them to better assess their relevance and potential use within research and data analysis contexts.
Prior knowledge
Programing is not a requirement but a basic notion would be helpful.
Technical requirements
None
Data Train / U Bremen Research Alliance