Limitations of Large Language Models

Hi All,

Welcome to our first meet up in Leeds, co-hosted with SciML and AI^2! Here are the talks for this meetup:

Talk 1: Addressing Biases in Large Language Models
In this talk Prof Serge Sharoff will present different types of biases in Large Language Models (LLMs) and various techniques that can be used for addressing them.

About the Speaker: Professor Serge Sharoff
Serge Sharoff is a Professor of Language Technology. Artificial Intelligence and more specifically Large Language Models, such as ChatGPT, have recently made a profound impact on how we interact with the computers by providing the ability to produce new texts in response to prompts. Fundamental research in this area is at the core of his expertise. He has been doing this since his PhD of 1997 from the Moscow State University. Since then he worked at the Russian Research Institute for Artificial Intelligence, University of Bielefeld (Germany) and finally he has been working at the University of Leeds since 2003. His research interests are related to three domains: linguistics (primarily computational linguistics and corpus linguistics), cognitive science and communication studies. The most interesting bit in his recent research is digital curation of representative corpora automatically collected from the Web, i.e., their annotation in terms of genres, domains or linguistic properties.

Talk 2: The main limitations of LLMs: Imagination and Creativity
In this talk Prof Eric will present two main limitations of LLMs, (1) LLM responses are combinations of snippets of training text which match snippets of input text, so they do not truly mimic imagination or creativity. And, (2) the main limitation regarding the usage and applications of LLMs is our own imagination: we need to think creatively about how to best use this new magic!

About the Speaker: Professor Eric Atwell
Eric is a Professor of Artificial Intelligence for Language. In 1984 he joined Leeds University to teach and research in Artificial Intelligence, Data Mining and Text Analytics in the School of Computing, with a research focus on Large Language Models for education, and for understanding religious texts. He also works 2 days a week at LITE Leeds Institute for Teaching Excellence as a LITE Fellow, on research in AI for Decolonizing Reading Lists. He is also a Turing Fellow at the Alan Turing Institute for Data Science and Artificial Intelligence and a member of LIDA Leeds Institute for Data Analytics and LATL Language At Leeds.

Special thanks to the Leeds Institute for Data Analytics (LIDA) for hosting us! Please let me know if you or someone you know wants to give a talk in our meetups next year. We are a community of reinforcement learning and generative AI engineers and enthusiasts who discover, learn and share skills together. Everyone Welcome :)