Language technologies are becoming more pervasive in our everyday lives, and they are also being applied in critical domains involving health, justice, and education. Given the importance of these applications and how they may affect our quality of life, it has become crucial to assess the errors they may make. In characterizing patterns of error, it has been found that systems obtained by machine-learning(ML) techniques from large quantities of text, such as large language models (LLMs), reproduce and amplify stereotypes.4 When deployed in actual applications, amplification of stereotypes can result in discriminatory behavior considered harmful in many jurisdictions. This kind of behavior is known as social bias, in that errors are distributed unevenly across social groups.

Word embeddings are the vector representation of words which act as an input (features) to other downstream tasks and ML models. There are several popular methods for learning word embeddings; among them, the Continous-Bag-of-Words and Glove models are the two most popular techniques. These embeddings act as an input to several NLP applications, i.e. sentiment analysis, document clustering, question answering, paraphrase detection, etc. Large organizations like Google and Facebook have trained these models on large web-scale corpora and made them readily available. Word embeddings encode the words such that words with similar meanings lie close to each other in the embedding space.

Most word embeddings used are glaringly sexist, let us look at some ways to de-bias such embeddings. Note - This article provides a review and the arguments made by Bolukbasi et al. in the paper "Man is to Computer Programmer as Woman is to Homemaker? All graphical drawings are made using draw.io. Word Embeddings are the core of NLP applications, and often, they end up being biased towards a gender due to the inherent stereotype present in the large text corpora they are trained on. Such models, when deployed to production can result in further widening of gender inequality and can have far fetched consequences on our society as a whole. To get a gist of what I'm talking about, here is a snippet from Bolukbasi et al., 2016 "Man is to Computer Programmer as Woman is to Homemaker?

Most word embeddings used are glaringly sexist, let us look at some ways to de-bias such embeddings. Note - This article provides a review and the arguments made by Bolukbasi et al. in the paper "Man is to Computer Programmer as Woman is to Homemaker? All graphical drawings are made using draw.io. Word Embeddings are the core of NLP applications, and often, they end up being biased towards a gender due to the inherent stereotype present in the large text corpora they are trained on. Such models, when deployed to production can result in further widening of gender inequality and can have far fetched consequences on our society as a whole. To get a gist of what I'm talking about, here is a snippet from Bolukbasi et al., 2016 "Man is to Computer Programmer as Woman is to Homemaker?

Recent researches demonstrate that word embeddings, trained on the human-generated corpus, have strong gender biases in embedding spaces, and these biases can result in the prejudiced results from the downstream tasks, i.e. sentiment analysis. Whereas the previous debiasing models project word embeddings into a linear subspace, we introduce a Latent Disentangling model with a siamese auto-encoder structure and a gradient reversal layer. Our siamese auto-encoder utilizes gender word pairs to disentangle semantics and gender information of given word, and the associated gradient reversal layer provides the negative gradient to distinguish the semantics from the gender. Afterwards, we introduce a Counterfactual Generation model to modify the gender information of words, so the original and the modified embeddings can produce a gender-neutralized word embedding after geometric alignment without loss of semantic information. Experimental results quantitatively and qualitatively indicate that the introduced method is better in debiasing word embeddings, and in minimizing the semantic information losses for NLP downstream tasks.

Word embedding models have become a fundamental component in a wide range of Natural Language Processing (NLP) applications. However, embeddings trained on human-generated corpora have been demonstrated to inherit strong gender stereotypes that reflect social constructs. To address this concern, in this paper, we propose a novel training procedure for learning gender-neutral word embeddings. Our approach aims to preserve gender information in certain dimensions of word vectors while compelling other dimensions to be free of gender influence. Based on the proposed method, we generate a Gender-Neutral variant of GloVe (GN-GloVe). Quantitative and qualitative experiments demonstrate that GN-GloVe successfully isolates gender information without sacrificing the functionality of the embedding model.