I am interested in the application of large language models to government data, specifically publicly facing data which is written at a readability level outside of the range of the public. At the moment I am focusing on Australian federal government bills and explanatory documents.

My 2022 masters thesis evaluated federal explanatory memoranda using both qualitative and quantitative methods to assess possible access issues associated with cognitive load, across readability and document structure.

My current PhD project explores the use of open large language models to make readable versions of the summaries, notes on clauses, and compatibility statements, with a focus on assessing what is gained and lost in the process of generation. I am using python as my language of choice, and my code will be released on my github at the end of the project.