1 code implementation • 2 Apr 2024 • Ivo Verhoeven, Pushkar Mishra, Rahel Beloch, Helen Yannakoudakis, Ekaterina Shutova
This mismatch can be partially attributed to the limitations of current evaluation setups that neglect the rapid evolution of online content and the underlying social graph.
1 code implementation • 19 Oct 2023 • Abhijith Chintam, Rahel Beloch, Willem Zuidema, Michael Hanna, Oskar van der Wal
Language models (LMs) exhibit and amplify many types of undesirable biases learned from the training data, including gender bias.