31 citations · 59 across the 10 of their papers we have counts for
3 papers · 1 filter
"I'm sorry to hear that": Finding New Biases in Language Models with a Holistic Descriptor Dataset
Eric Michael Smith, Melissa Hall, Melanie Kambadur +2
As language models grow in popularity, it becomes increasingly important to clearly measure all possible markers of demographic identity in order to avoid perpetuating existing soc…
Understanding out-of-distribution accuracies through quantifying difficulty of test samples
Berfin Simsek, Melissa Hall, Levent Sagun
Existing works show that although modern neural networks achieve remarkable generalization performance on the in-distribution (ID) dataset, the accuracy drops significantly on the…
A Systematic Study of Bias Amplification
Melissa Hall, Laurens van der Maaten, Laura Gustafson +2
Recent research suggests that predictions made by machine-learning models can amplify biases present in the training data. When a model amplifies bias, it makes certain predictions…