Speakers: Katie Collins
PhD Student, Machine Learning Group
Professor Mirella Lapata
Professor in the School of Informatics, University of Edinburgh
00:00 Human-Centric Benchmarking
25:00 How did you go about recruiting the ‘right humans’ you would include in this benchmarking process?
25:49 How do we benchmark language in the world of LLMs if we always use English as a first language?
26:41 How do you see this method being adapted by authorities or companies in their day to day practice?
28:28 Can you explain a little more about the process ‘rating the ratings’?