English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Recent research demonstrates that word embeddings, trained on thehuman-generated corpus, have strong gender biases in embedding spaces, andthese biases can result in the discriminative results from the variousdownstream tasks. Whereas the previous methods project word embeddings into alinear subspace for debiasing, we introduce a \textit{Latent Disentanglement}method with a siamese auto-encoder structure with an adapted gradient reversallayer. Our structure enables the separation of the semantic latent informationand gender latent information of given word into the disjoint latentdimensions. Afterwards, we introduce a \textit{Counterfactual Generation} toconvert the gender information of words, so the original and the modifiedembeddings can produce a gender-neutralized word embedding after geometricalignment regularization, without loss of semantic information. From thevarious quantitative and qualitative debiasing experiments, our method shows tobe better than existing debiasing methods in debiasing word embeddings. Inaddition, Our method shows the ability to preserve semantic information duringdebiasing by minimizing the semantic information losses for extrinsic NLPdownstream tasks.

Neutralizing Gender Bias in Word Embedding with Latent Disentanglement and Counterfactual Generation