BiasMeasures

Code for Trustworthy Social Bias Measurement (Bommasani and Liang, 2022). Sorry that this code is currently quite messy, but will be improved in the future. Direct any questions to rishibommasani@gmail.com. To access the data, all information on sourcing is provided in the paper.

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
.gitignore		.gitignore
LICENSE		LICENSE
README.md		README.md
aligned_estimators.py		aligned_estimators.py
automated_text_associations.py		automated_text_associations.py
convergent_validity.py		convergent_validity.py
data_loader.py		data_loader.py
face_validity.py		face_validity.py
interannotator.py		interannotator.py
our_estimator.py		our_estimator.py
our_word_list.py		our_word_list.py
overamplification.py		overamplification.py
predictive_validity.py		predictive_validity.py
preprocess.py		preprocess.py
preprocess_contextualized_representations.py		preprocess_contextualized_representations.py
primitive_operations.py		primitive_operations.py
probe.py		probe.py
sensitivity.py		sensitivity.py
train.py		train.py
unaligned_estimators.py		unaligned_estimators.py
word_list.py		word_list.py

License

rishibommasani/BiasMeasures

Folders and files

Latest commit

History

Repository files navigation

BiasMeasures

About

Resources

License

Stars

Watchers

Forks

Languages