Toward Gender-Inclusive Coreference Resolution

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4568–4595 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 4568 Toward Gender-Inclusive Coreference Resolution Yang Trista Cao University of Maryland ycao95@cs.umd.edu Hal Daum´e III University of Maryland Microsoft Research me@hal3.name Abstract Correctly resolving textual mentions of people fundamentally entails making inferences about those people. Such inferences raise the risk of systemic biases in coreference resolution sys- tems, including biases that can harm binary and non-binary trans and cis stakeholders. To better understand such biases, we foreground nuanced conceptualizations of gender from so- ciology and sociolinguistics, and develop two new datasets for interrogating bias in crowd annotations and in existing coreference reso- lution systems. Through these studies, con- ducted on English text, we confirm that with- out acknowledging and building systems that recognize the complexity of gender, we build systems that lead to many potential harms. 1 Introduction Coreference resolution—the task of determining which textual references resolve to the same real- world entity—requires making inferences about those entities. Especially when those entities are people, coreference resolution systems run the risk of making unlicensed inferences, possibly resulting in harms either to individuals or groups of people. Embedded in coreference inferences are varied as- pects of gender, both because gender can show up explicitly (e.g., pronouns in English, morphology in Arabic) and because societal expectations and stereotypes around gender roles may be explicitly or implicitly assumed by speakers or listeners. This can lead to significant biases in coreference resolu- tion systems: cases where systems “systematically and unfairly discriminate against certain individ- uals or groups of individuals in favor of others” (Friedman and Nissenbaum, 1996, p. 332). Gender bias in coreference resolution can mani- fest in many ways; work by Rudinger et al. (2018), Zhao et al. (2018a), and Webster et al. (2018) fo- cused largely on the case of binary gender dis- crimination in trained coreference systems, show- ing that current systems over-rely on social stereo- types when resolving HE and SHE pronouns1 (see §2). Contemporaneously, critical work in Human- Computer Interaction has complicated discussions around gender in other fields, such as computer vision (Keyes, 2018; Hamidi et al., 2018). Building on both lines of work, and inspired by Keyes’s (2018) study of vision-based automatic gender recognition systems, we consider gender bias from a broader conceptual frame than the bi- nary “folk” model. We investigate ways in which folk notions of gender—namely that there are two genders, assigned at birth, immutable, and in per- fect correspondence to gendered linguistic forms— lead to the development of technology that is exclu- sionary and harmful of binary and non-binary trans and cis people.2 Addressing such issues is critical not just to improve the quality of our systems, but more pointedly to minimize the harms caused by our systems by reinforcing existing unjust social hierarchies (Lambert and Packer, 2019). There are several stakeholder groups who may easily face harms when coreference systems is used (Blodgett et al., 2020). Those harms includes several …

Content truncated to 500 words