Equalizing Gender Bias in Neural Machine Translation with Word Embeddings Techniques

Citations Over TimeTop 10% of 2019 papers

Abstract

Neural machine translation has significantly pushed forward the quality of the field. However, there are remaining big issues with the output translations and one of them is fairness. Neural models are trained on large text corpora which contain biases and stereotypes. As a consequence, models inherit these social biases. Recent methods have shown results in reducing gender bias in other natural language processing tools such as word embeddings. We take advantage of the fact that word embeddings are used in neural machine translation to propose a method to equalize gender biases in neural machine translation using these representations. We evaluate our proposed system on the WMT English-Spanish benchmark task, showing gains up to one BLEU point. As for the gender bias evaluation, we generate a test set of occupations and we show that our proposed system learns to equalize existing biases from the baseline system.

Related Papers

• → Tied Transformers: Neural Machine Translation with Shared Encoder and Decoder(2019)65 cited
• → Neural Machine Translation with the Transformer and Multi-Source Romance Languages for the Biomedical WMT 2018 task(2018)15 cited
• → Searching Better Architectures for Neural Machine Translation(2020)27 cited
• → The Role of Protected Class Word Lists in Bias Identification of Contextualized Word Representations(2019)17 cited
• → Incorporating Pre-trained Model into Neural Machine Translation(2021)2 cited