Screening Gender Transfer in Neural Machine Translation

Guillaume Wisniewski (1) , Lichao Zhu (1) , Nicolas Ballier (2) , François Yvon (3)

1 LLF - Laboratoire de Linguistique Formelle
2 CLILLAC-ARP (URP_3967) - Centre de Linguistique Inter-langues, de Lexicologie, de Linguistique Anglaise et de Corpus
3 LISN - Laboratoire Interdisciplinaire des Sciences du Numérique

This paper aims at identifying the information flow in state-of-the-art machine translation systems, taking as example the transfer of gender when translating from French into English. Using a controlled set of examples, we experiment several ways to investigate how gender information circulates in a encoder-decoder architecture considering both probing techniques as well as interventions on the internal representations used in the MT system. Our results show that gender information can be found in all token representations built by the encoder and the decoder and lead us to conclude that there are multiple pathways for gender transfer.

Keywords: Gender bias; Neural Machine Translation; Probing; Transformer architectures

Domains: Computer Science [cs]; Linguistics

Guillaume Wisniewski, Lichao Zhu, Nicolas Ballier, François Yvon. Screening Gender Transfer in Neural Machine Translation. BlackBoxNLP 2021, Nov 2021, Punta Cana, Dominican Republic. ⟨hal-03993451⟩