You Shall Know a Word’s Gender by the Company it Keeps: Comparing the Role of Context in Human Gender Assumptions with MT

Janiça Hackenbuchner, Joke Daems, Arda Tezcan, Aaron Maladry


Abstract
In this paper, we analyse to what extent machine translation (MT) systems and humans base their gender translations and associations on role names and on stereotypicality in the absence of (generic) grammatical gender cues in language. We compare an MT system’s choice of gender for a certain word when translating from a notional gender language, English, into a grammatical gender language, German, with thegender associations of humans. We outline a comparative case study of gender translation and annotation of words in isolation, out-of-context, and words in sentence contexts. The analysis reveals patterns of gender (bias) by MT and gender associations by humans for certain (1) out-of-context words and (2) words in-context. Our findings reveal the impact of context on gender choice and translation and show that word-level analyses fall short in such studies.
Anthology ID:
2024.gitt-1.4
Volume:
Proceedings of the 2nd International Workshop on Gender-Inclusive Translation Technologies
Month:
June
Year:
2024
Address:
Sheffield, United Kingdom
Editors:
Beatrice Savoldi, Janiça Hackenbuchner, Luisa Bentivogli, Joke Daems, Eva Vanmassenhove, Jasmijn Bastings
Venues:
GITT | WS
SIG:
Publisher:
European Association for Machine Translation (EAMT)
Note:
Pages:
31–41
Language:
URL:
https://aclanthology.org/2024.gitt-1.4
DOI:
Bibkey:
Cite (ACL):
Janiça Hackenbuchner, Joke Daems, Arda Tezcan, and Aaron Maladry. 2024. You Shall Know a Word’s Gender by the Company it Keeps: Comparing the Role of Context in Human Gender Assumptions with MT. In Proceedings of the 2nd International Workshop on Gender-Inclusive Translation Technologies, pages 31–41, Sheffield, United Kingdom. European Association for Machine Translation (EAMT).
Cite (Informal):
You Shall Know a Word’s Gender by the Company it Keeps: Comparing the Role of Context in Human Gender Assumptions with MT (Hackenbuchner et al., GITT-WS 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.gitt-1.4.pdf