princeton-nlp

mabel-bert-base-uncased

princeton-nlp/mabel-bert-base-uncased

mabel-bert-base-uncased is a fill-mask model from princeton-nlp. Use it when you need the model to fill a missing word. It is set up for transformers.

This is the model for MABEL, as described in our paper, "MABEL: Attenuating Gender Bias using Textual Entailment Data". MABEL is trained from an underlying bert-base-uncased backbone, and demonstrates a good bias-perf…