Abstract
Neural machine translation (NMT) models often suffer from gender biases that harm users and society at large. In this work, we explore how bridging the gap between languages for which parallel data is not available affects gender bias in multilingual NMT, specifically for zero-shot directions. We evaluate translation between grammatical gender languages which requires preserving the inherent gender information from the source in the target language. We study the effect of encouraging language-agnostic hidden representations on models’ ability to preserve gender and compare pivot-based and zero-shot translation regarding the influence of the bridge language (participating in all language pairs during training) on gender preservation. We find that language-agnostic representations mitigate zero-shot models’ masculine bias, and with increased levels of gender inflection in the bridge language, pivoting surpasses zero-shot translation regarding fairer gender preservation for speaker-related gender agreement.- Anthology ID:
- 2023.gitt-1.3
- Volume:
- Proceedings of the First Workshop on Gender-Inclusive Translation Technologies
- Month:
- June
- Year:
- 2023
- Address:
- Tampere, Finland
- Editors:
- Eva Vanmassenhove, Beatrice Savoldi, Luisa Bentivogli, Joke Daems, Janiça Hackenbuchner
- Venue:
- GITT
- SIG:
- Publisher:
- European Association for Machine Translation
- Note:
- Pages:
- 25–35
- Language:
- URL:
- https://aclanthology.org/2023.gitt-1.3
- DOI:
- Cite (ACL):
- Lena Cabrera and Jan Niehues. 2023. Gender Lost In Translation: How Bridging The Gap Between Languages Affects Gender Bias in Zero-Shot Multilingual Translation. In Proceedings of the First Workshop on Gender-Inclusive Translation Technologies, pages 25–35, Tampere, Finland. European Association for Machine Translation.
- Cite (Informal):
- Gender Lost In Translation: How Bridging The Gap Between Languages Affects Gender Bias in Zero-Shot Multilingual Translation (Cabrera & Niehues, GITT 2023)
- PDF:
- https://preview.aclanthology.org/ingest-acl-2023-videos/2023.gitt-1.3.pdf