ORQA working paper introduces a framework to test LLMs’ professional knowledge
The paper’s authors say their approach turns authoritative online sources into test questions to benchmark occupation-specific knowledge at scale.
TLDR
ORQA’s authors present the working paper as a way to evaluate large language models’ professional knowledge across occupations. They describe a scalable approach that turns authoritative online sources into test questions, addressing what they call a difficult and costly evaluation challenge. They also shared links to the paper and a dashboard.
Combined views
2.5K
1 Source, first seen 15d ago