GATE: A Challenge Set for Gender-Ambiguous Translation Examples

TL;DR

GATE introduces a diverse corpus and evaluation tools for gender-ambiguous translation, addressing biases and improving linguistic coverage in machine translation systems.

Contribution

The paper presents GATE, a new challenging dataset with evaluation tools to enhance gender-ambiguous translation and reduce stereotypical biases.

Findings

GATE enables better evaluation of gender rewriters.

Current systems show limited coverage on ambiguous inputs.

GATE promotes fairer machine translation practices.

Abstract

Although recent years have brought significant progress in improving translation of unambiguously gendered sentences, translation of ambiguously gendered input remains relatively unexplored. When source gender is ambiguous, machine translation models typically default to stereotypical gender roles, perpetuating harmful bias. Recent work has led to the development of "gender rewriters" that generate alternative gender translations on such ambiguous inputs, but such systems are plagued by poor linguistic coverage. To encourage better performance on this task we present and release GATE, a linguistically diverse corpus of gender-ambiguous source sentences along with multiple alternative target language translations. We also provide tools for evaluation and system analysis when using GATE and use them to evaluate our translation rewriter system.