Gastvorlesung Dr. Philipp Wicke (LMU München): »Introduction to Large Language Models (LLMs) and Prompt Engineering« (hybrid)

The field of natural language processing has pushed its boundaries with the introduction of large language models (LLMs), which are now almost synonymous with artificial intelligence. Yet, natural language is both the source and the limitation of these systems. Essentially, a large language model merely predicts the next word based on large amounts of textual data for which its training has achieved an efficient representation. Due to its opaque and autoregressive nature, the LLM is prompted using natural language. Understanding this nature allows us to select the appropriate prompts to yield better results for a variety of tasks.
This lecture has a bipartite structure. In the first half, I will introduce the origins and basics of LLMs. This will allow us to make sense of prompting strategies and prompt engineering, which will be presented in the second half of the lecture.

Short bio:

Dr. Philipp Wicke is a Lecturer and Postdoc Researcher at the Ludwig Maximilian University (LMU). He is working on a diversity of topics related to Artificial Intelligence and Computational Linguistics. His doctoral thesis on Embodied Storytelling using Spatial Metaphor and Gesture provides a contribution to the research of Computational Creativity. In his current role at LMU, Philipp is researching on Natural Language Processing and teaches programming in the B.A. and M.A. Computational Linguistics. His background in Cognitive Science fields his research interest in Computational Creativity, Embodied Cognition and Artificial Intelligence. Philipp Wicke is also the Head of AI Applications of the AI for People Association, which promotes the idea of shaping Artificial Intelligent technology around human and societal needs.