Dictionary-based Debiasing of Pre-trained Word Embeddings

Masahiro Kaneko, Danushka Bollegala

Code and debiased word embeddings for the paper: "Dictionary-based Debiasing of Pre-trained Word Embeddings" (In EACL 2021). If you use any part of this work, make sure you include the following citation:

@inproceedings{kaneko-bollegala-2021-dict,
    title={Dictionary-based Debiasing of Pre-trained Word Embeddings},
    author={Masahiro Kaneko and Danushka Bollegala},
    booktitle = {Proc. of the 16th European Chapter of the Association for Computational Linguistics (EACL)},
    year={2021}
}

Requirements

• python==3.7.2
• torch==1.6.0
• gensim==3.7.3
• numpy==1.19.1
• nltk==3.4

To debias your word embeddinngs

cd src
python train.py --embedding path/to/your/embeddings --dictionary ../data/dict_wn.json --config config/hyperparameter.json --save-prefix path/to/save/directory --gpu id --save-binary

Output is a debiased binary word embeddings saved in --save-prefix

Our debiased word embeddings

You can directly download our debiased word embeddings.

License

See the LICENSE file.