Reaction
Can LLMs recognize unstated cultural norms in health queries?
A post introduces CCBench as an assessment of LLM cultural competence, led and presented by Vasudha Varadarajan.
TLDR
A post describes CCBench as assessing whether language models recognize culturally relevant norms that users do not explicitly state in health queries. It credits Vasudha Varadarajan with leading and presenting the project.
Combined views
170
1 Source, first seen ago