Leon Fröhling

A critical look at the data used to combat online hate speech.

Researchers analyze political biases in language models using various personas.

A new library to evaluate AI alignment with human viewpoints.