Finding the values encoded in AI and whether they could cause cultural clash with their users.
A catalog of unintended consequences of technologies to help raise public awareness.
Measuring alignment of NLP datasets and models with people around the world.
Extracting nobility and social hierarchies from DELPHI, an AI which models moral judgements.
Compete with other players in building the best prompts and see who wins!
Dashboard to display statistics of paper submissions and reviews on ACL Rolling Review.
Probing for overall stance of language models through behavioral probing.
A tool for generating name pronunciations and as a result, collecting them in the wild.
Extract anti-social elements from LMs and see if you are fit to be the next "Guardian of AI".
Internal
