Getting Gender Right in Neural Machine Translation

By Eva Vanmassenhove, Christian Hardmeier, Andy Way. 2019

217 citations recorded by the source corpus.

Speakers of different languages must attend to and encode strikingly different aspects of the world in order to use their language correctly (Sapir, 1921; Slobin, 1996). One such difference is related to the way gender is expressed in a language. Saying “I am happy” in English, does not encode any additional knowledge of the speaker that uttered the sentence. However, many other languages do have grammatical gender systems and so such knowledge would be encoded. In order to correctly translate such a sentence into, say, French, the inherent gender information needs to be retained/recovered. The same sentence would become either “Je suis heureux”, for a male speaker or “Je suis heureuse” for a female one. Apart from morphological agreement, demographic factors (gender, age, etc.) also influence our use of language in terms of word choices or syntactic constructions (Tannen, 1991; Pennebaker et al., 2003). We integrate gender information into NMT systems. Our contribution is two-fold: (1) the compilation of large datasets with speaker information for 20 language pairs, and (2) a simple set of experiments that incorporate gender information into NMT for multiple language pairs. Our experiments show that adding a gender feature to an NMT system significantly improves the translation quality for some language pairs.

Subfields: Artificial Intelligence, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Language (Technology) is Power: A Critical Survey of “Bias” in NLP (2020, methodology): “…language modeling (Lu et al., 2018; Bordia and Bowman, 2019), coreference resolution (Rudinger et al., 2018; Zhao et al., 2018a), machine translation (Vanmassenhove et al., 2018; Stanovsky et al., 2019), sentiment analysis (Kiritchenko and Mohammad, 2018), and hate speech/toxicity detection…”
Societal Biases in Language Generation: Progress and Challenges (2021, methodology): “The second category is methods that attach a short preﬁx at training time (Vanmassenhove et al., 2018; Basta et al., 2020; Alhafni et al., 2020) or inference time (Moryossef et al., 2019).”
Scalable Cross Lingual Pivots to Model Pronoun Gender for Translation (2020, background): “Most similar to our current work, Habash et al. (2019); Moryossef et al. (2019); Vanmassenhove et al. (2018) improve gender ﬁdelity in translations in of ﬁrst-person singular pronouns in languages with grammatical gender inﬂection in this position.”
Gender Bias in Contextualized Word Embeddings (2019, background): “, 2018), machine translation (Font and Costa-jussà, 2019; Vanmassenhove et al., 2018), and web search (Kay et al.”
Mitigating Gender Bias in Natural Language Processing: Literature Review (2019, methodology): “This happens because training sets are dominated by male-sourced data points, so the models learn skewed statistical relationships and are thus more likely to predict the speaker to be male when the gender of the source is ambiguous (Vanmassenhove et al., 2018).”
A Survey on Bias and Fairness in Natural Language Processing (2022, neutral): “Vanmassenhove et al. (2018) thus give more emphasis on speaker information being present in parallel data.”
Equalizing Gender Bias in Neural Machine Translation with Word Embeddings Techniques (2019, background): “Vanmassenhove et al. (2018) compile a large multilingual dataset on the politics domain that”
Automatic Gender Identification and Reinflection in Arabic (2019, methodology): “This paper sits in the intersection of efforts like data augmentation for morphologically rich languages (Zmigrod et al., 2019) and gender-aware MT (Vanmassenhove et al., 2018; Elaraby et al., 2018).”
Disembodied Machine Learning: On the Illusion of Objectivity in NLP (2021, background): “In MT, research has identiﬁed a range of issues, including stylistic (Hovy, Bianchi, and Fornaciari 2020) and gender bias (Vanmassenhove, Hardmeier, and Way 2018).”
First the Worst: Finding Better Gender Translations During Beam Search (2021, background): “ties (Vanmassenhove et al., 2018), or if translating the same sentence with different entity genders to produce multiple outputs (Moryossef et al.”
Gender bias and stereotypes in Large Language Models (2023, background): “auto-captioning, sentiment analysis, toxicity detection, machine translation, and more [46, 58, 71, 80, 83, 87, 91, 95].”
Token-level Adaptive Training for Neural Machine Translation (2020, unknown): “In NMT, the class imbalance problem might be the underlying cause of, among others, the gender-biased output problem (Vanmassenhove et al., 2019a), the inability of MT system to handle morphologically richer language correctly (Passban et al., 2018), or the exposure bias problem (Ranzato et al.,…”
Gender bias amplification during Speed-Quality optimization in Neural Machine Translation (2021, unknown): “Vanmassenhove et al. (2018), Stafanoviˇcs et al. (2020) and Saunders et al. (2020) propose a data-annotation scheme in which the NMT model is trained to obey gender-speciﬁc tags provided with the source sentence.”
Towards Mitigating Gender Bias in a decoder-based Neural Machine Translation model by Adding Contextual Information (2020, methodology): “The main contribution of this work is using existing NMT contextual methodologies, both context of the previous sentence (Junczys-Dowmunt, 2019) and speaker identiﬁcation (Vanmassenhove et al., 2018), in a prominent and competitive NMT architecture (Fonollosa et al., 2019).”
Decoding and Diversity in Machine Translation (2020, background): “It has been observed that NMT can often produce misgendered outputs, and solutions focusing on the modeling stage have been proposed, however, our study implicates search as a source of gender pronoun bias (Vanmassenhove et al., 2018; Saunders and Byrne, 2020).”