From virtual assistants and chatbots to AI search engines and recommendation systems, large language models (LLMs) seem to be ubiquitous in our daily interactions with technology.  But how do these systems actually generate responses, answer questions, and retrieve knowledge? Join us for the second session of this foundational mini-series as we explore the ideas behind modern generative AI systems.

In this session, we will build on the foundations of natural language processing (NLP) to explore how large language models learn language patterns and generate human-like responses. We will discuss concepts such as embeddings and transformers and examine how probabilities and context help AI systems predict and generate text. We will also introduce retrieval-augmented generation (RAG), a method that combines language models with external sources of information to improve accuracy, provide context, and retrieve knowledge from documents and databases.

Join us as we explore how modern AI systems generate language and retrieve information. As a foundational series, this session is designed for a beginner-to-intermediate audience. Whether you are an AI enthusiast or want it gone from your browser, you will come out of this session with a conceptual understanding of how AI decides how to answer your questions. Come join us!