# Gender Bias in Multilingual Neural Machine Translation: The Architecture Matters

arXiv: https://arxiv.org/abs/2012.13176 · HTML: https://mlsanity.com/paper/2012.13176/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Dec 2020.

## Built on by (most cited first, 11 of 11)

- [Societal Biases in Language Generation: Progress and Challenges](https://mlsanity.com/paper/2105.04054/) (2021)
- [Gender Bias in Machine Translation](https://mlsanity.com/paper/2104.06001/) (2021)
- [Theories of "Gender" in NLP Bias Research](https://mlsanity.com/paper/2205.02526/) (2022)
- [Gender Bias Amplification During Speed-Quality Optimization in Neural Machine Translation](https://mlsanity.com/paper/2106.00169/) (2021)
- [Multilingual Holistic Bias: Extending Descriptors and Patterns to Unveil Demographic Biases in Languages at Scale](https://mlsanity.com/paper/2305.13198/) (2023)
- [The power of Prompts: Evaluating and Mitigating Gender Bias in MT with LLMs](https://mlsanity.com/paper/2407.18786/) (2024)
- [How to Split: the Effect of Word Segmentation on Gender Bias in Speech Translation](https://mlsanity.com/paper/2105.13782/) (2021)
- [Synthetic Pre-Training Tasks for Neural Machine Translation](https://mlsanity.com/paper/2212.09864/) (2022)
- [Are We Paying Attention to Her? Investigating Gender Disambiguation and Attention in Machine Translation](https://mlsanity.com/paper/2505.08546/) (2025)
- [Screening Gender Transfer in Neural Machine Translation](https://mlsanity.com/paper/2202.12568/) (2022)
- [Towards Massive Multilingual Holistic Bias](https://mlsanity.com/paper/2407.00486/) (2024)

## Abstract (arXiv)

Multilingual Neural Machine Translation architectures mainly differ in the amount of sharing modules and parameters among languages. In this paper, and from an algorithmic perspective, we explore if the chosen architecture, when trained with the same data, influences the gender bias accuracy. Experiments in four language pairs show that Language-Specific encoders-decoders exhibit less bias than the Shared encoder-decoder architecture. Further interpretability analysis of source embeddings and the attention shows that, in the Language-Specific case, the embeddings encode more gender information, and its attention is more diverted. Both behaviors help in mitigating gender bias.