Investigating Markers and Drivers of Gender Bias in Machine Translations

Barclay, Peter; Sami, Ashkan

Abstract

Implicit gender bias in Large Language Models (LLMs) is a well-documented problem, and implications of gender introduced into automatic translations can perpetuate real-world biases. However, some LLMs use heuristics or post-processing to mask such bias, making investigation difficult. Here, we examine bias in LLMs via back-translation, using the DeepL translation API to investigate the bias evinced when repeatedly translating a set of 56 Software Engineering tasks used in a previous study. Each statement starts with ‘she’, and is translated first into a ‘genderless’ intermediate language then back into English; we then examine pronoun-choice in the back-translated texts. We expand prior research in the following ways: (1) by comparing results across five intermediate languages, namely Finnish, Indonesian, Estonian, Turkish and Hungarian; (2) by proposing a novel metric for assessing the variation in gender implied in the repeated translations, avoiding the over-interpretation of individual pronouns, apparent in earlier work; (3) by investigating sentence features that drive bias; (4) and by comparing results from three time-lapsed datasets to establish the reproducibility of the approach. We found that some languages display similar patterns of pronoun use, falling into three loose groups, but that patterns vary between groups; this underlines the need to work with multiple languages. We also identify the main verb appearing in a sentence as a likely significant driver of implied gender in the translations. Moreover, we see a good level of replicability in the results, and establish that our variation metric proves robust despite an obvious change in the behaviour of the DeepL translation API during the course of the study. These results show that the back-translation method can provide further insights into bias in language models.

Citation

Barclay, P., & Sami, A. (2024, March). Investigating Markers and Drivers of Gender Bias in Machine Translations. Presented at IEEE International Conference on Software Analysis, Evolution and Reengineering, Rovaniemi, Finland

• Presentation Conference Type: Conference Paper (published)
• Conference Name: IEEE International Conference on Software Analysis, Evolution and Reengineering
• Start Date: Mar 12, 2024
• End Date: Mar 15, 2024
• Acceptance Date: Dec 16, 2023
• Online Publication Date: Jul 16, 2024
• Publication Date: 2024
• Deposit Date: Apr 30, 2024
• Publicly Available Date: Jul 16, 2024
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Pages: 455-464
• Series ISSN: 2640-7574
• Book Title: 2024 IEEE International Conference on Software Analysis, Evolution and Reengineering (SANER)
• ISBN: 9798350330670
• DOI: https://doi.org/10.1109/SANER60148.2024.00054
• Keywords: back-translation, machine translation, large lan- guage model, gender bias
• Public URL: http://researchrepository.napier.ac.uk/Output/3608153
• Publisher URL: https://conf.researchr.org/home/saner-2024
• Additional Information: To be published in conference proceedings by IEEE.; Pre-print version available on arXiv as noted.