Evaluating Gender Bias in Machine Translation
This addresses gender bias in machine translation, which is a critical issue for fairness and accuracy in AI applications, though it is incremental as it builds on existing coreference resolution datasets.
The authors tackled the problem of gender bias in machine translation by creating the first challenge set and evaluation protocol, showing that both industrial and academic MT systems are significantly prone to gender-biased translation errors across eight target languages.
We present the first challenge set and evaluation protocol for the analysis of gender bias in machine translation (MT). Our approach uses two recent coreference resolution datasets composed of English sentences which cast participants into non-stereotypical gender roles (e.g., "The doctor asked the nurse to help her in the operation"). We devise an automatic gender bias evaluation method for eight target languages with grammatical gender, based on morphological analysis (e.g., the use of female inflection for the word "doctor"). Our analyses show that four popular industrial MT systems and two recent state-of-the-art academic MT models are significantly prone to gender-biased translation errors for all tested target languages. Our data and code are made publicly available.