A PCA gender direction in word2vec is causally debiasable

Bolukbasi et al. compute a gender direction in word2vec (Google News, 300d) as the top principal component of ten gendered word-pair difference vectors [bolukbasi-etal-2016-debiasing-word-embeddings] Gender-neutral words are linearly separable from gender-definitional words along this direction [bolukbasi-etal-2016-debiasing-word-embeddings] Hard debiasing -- projecting the gender direction out of neutral words plus an equalize step for definitional pairs -- is applied directly to the embedding matrix without retraining [bolukbasi-etal-2016-debiasing-word-embeddings] Crowd and benchmark evaluation show stereotypical analogies (e.g. homemaker for women) are eliminated while semantic task performance is preserved [bolukbasi-etal-2016-debiasing-word-embeddings] Follow-up work notes single-direction debiasing only partially removes bias, since indirect bias re-emerges in higher-order structure [bolukbasi-etal-2016-debiasing-word-embeddings]