Measuring Social Biases in Masked Language Models by Proxy of Prediction Quality

Open in new window