Gender Disambiguation in Machine Translation: Diagnostic Evaluation in Decoder-Only Architectures

Chiara Manna, Hosein Mohebbi, Afra Alishahi, Frederic Blain, Eva Vanmassenhove


Abstract
While Large Language Models achieve state-of-the-art results across a wide range of NLP tasks, they remain prone to systematic biases. Among these, gender bias is particularly salient in MT, due to systematic differences across languages in whether and how gender is marked. As a result, translation often requires disambiguating implicit source signals into explicit gender-marked forms. In this context, standard benchmarks may capture broad disparities but fail to reflect the full complexity of gender bias in modern MT. In this paper, we extend recent frameworks on bias evaluation by: (i) introducing a novel measure coined ’Prior Bias’, capturing a model’s default gender assumptions, and (ii) applying the framework to decoder-only MT models. Our results show that, despite their scale and state-of-the-art status, decoder-only models do not generally outperform encoder-decoder architectures on gender-specific metrics; however, post-training (e.g., instruction tuning) not only improves contextual awareness but also reduces the masculine Prior Bias.
Anthology ID:
2026.lrec-1.673
Volume:
Proceedings of the Fifteenth Language Resources and Evaluation Conference
Month:
May
Year:
2026
Address:
Palma de Mallorca, Spain
Editors:
Stelios Piperidis, Núria Bel, Henk van den Heuvel, Nancy Ide, Simon Krek, Antonio Toral
Venue:
LREC
SIG:
Publisher:
ELRA Language Resource Association
Note:
Pages:
8535–8550
Language:
External URL:
https://lrec.elra.info/lrec2026-main-673
DOI:
10.63317/4wphxianzxf6
Bibkey:
Cite (ACL):
Chiara Manna, Hosein Mohebbi, Afra Alishahi, Frederic Blain, and Eva Vanmassenhove. 2026. Gender Disambiguation in Machine Translation: Diagnostic Evaluation in Decoder-Only Architectures. In Proceedings of the Fifteenth Language Resources and Evaluation Conference, pages 8535–8550, Palma de Mallorca, Spain. ELRA Language Resource Association.
Cite (Informal):
Gender Disambiguation in Machine Translation: Diagnostic Evaluation in Decoder-Only Architectures (Manna et al., LREC 2026)
Copy Citation: