PhD on Testing Large Language Models

Large Language Models (LLMs) and their associated tools, such as ChatGPT are revolutionizing the way society accesses to information. Despite their impressive performance, their outcomes are often subject to hallucinations and bias. Because of this, it is important to develop novel methods to test these models. This thesis has as a goal to develop a tool to automatically test LLMs. The PhD candidate will collaborate both with another national team (from University of Seville) as well as with an international team, including researchers from Norway and possibly China and Japan!