paper-with-me

홈 › Papers

Investigating Failures of Automatic Translation in the Case of Unambiguous Gender

2021-04-16 · ACL 2022 5 · Adithya Renduchintala, Adina Williams

Transformer based models are the modern work horses for neural machine translation (NMT), reaching state of the art across several benchmarks. Despite their impressive accuracy, we observe a systemic and rudimentary class of errors made by transformer based models with regards to translating from a language that doesn't mark gender on nouns into others that do. We find that even when the surrounding context provides unambiguous evidence of the appropriate grammatical gender marking, no transformer based model we tested was able to accurately gender occupation nouns systematically. We release an evaluation scheme and dataset for measuring the ability of transformer based NMT models to translate gender morphology correctly in unambiguous contexts across syntactically diverse sentences. Our dataset translates from an English source into 20 languages from several different language families. With the availability of this dataset, our hope is that the NMT community can iterate on solutions for this class of especially egregious errors.

📄 PDF Abstract BibTeX arXiv:2104.07838

Code (0)

등록된 구현이 없습니다.

Tasks

Machine TranslationNMTTranslation

Similar Papers 제목 키워드 기반

Gender, names and other mysteries: Towards the ambiguous for gender-inclusive translation

2023-06-07 · Danielle Saunders, Katrina Olsen

The vast majority of work on gender in MT focuses on 'unambiguous' inputs, where gender markers in the source language are expected to be resolved in the output. Conversely, this paper explores the widespread case where …

SentenceTranslation

Investigating Length Issues in Document-level Machine Translation

2024-12-23 · Ziqian Peng, Rachel Bawden, François Yvon

Transformer architectures are increasingly effective at processing and generating very long chunks of texts, opening new perspectives for document-level machine translation (MT). In this work, we challenge the ability of…

Document Level Machine TranslationMachine TranslationSentenceTranslation

Gender Bias in English-to-Greek Machine Translation

2025-06-11 · Eleni Gkovedarou, Joke Daems, Luna De Bruyne

As the demand for inclusive language increases, concern has grown over the susceptibility of machine translation (MT) systems to reinforce gender stereotypes. This study investigates gender bias in two commercial MT syst…

Machine TranslationTranslation

Testing Untestable Neural Machine Translation: An Industrial Case

2018-07-06 · Wujie Zheng, Wenyu Wang, Dian Liu, Changrong Zhang 외

Neural Machine Translation (NMT) has been widely adopted recently due to its advantages compared with the traditional Statistical Machine Translation (SMT). However, an NMT system still often produces translation failure…

Machine TranslationNMTTranslationvalid

Uncertainty Quantification for Evaluating Machine Translation Bias

2025-07-24 · Ieva Raminta Staliūnaitė, Julius Cheng, Andreas Vlachos arxiv

The predictive uncertainty of machine translation (MT) models is typically used as a quality estimation proxy. In this work, we posit that apart from confidently translating when a single correct translation exists, mode…

Machine Translation