A Bayesian account of pronoun and neopronoun acquisition

TL;DR

The paper addresses equity in pronoun and neopronoun use by modeling how individuals acquire referential forms without relying on appearance cues. It introduces a Bayesian framework based on the nested Chinese Restaurant Franchise Process (nCRFP) to learn and adapt pronouns and names for individuals and communities, assigning probabilities to referential forms with priors that encode social norms such as $P^c$ and $P^{c,t}$. Key contributions include a concrete single-speaker model and a community-norm model that accommodate novel forms (neopronouns, emojipronouns) and rapid revision. Such a framework supports equitable NLP by enabling systems to generate and interpret referring expressions that respect self-determined identities.

Abstract

A major challenge to equity among members of queer communities is the use of one's chosen forms of reference, such as personal names or pronouns. Speakers often dismiss their misuses of pronouns as "unintentional", and claim that their errors reflect many decades of fossilized mainstream language use, as well as attitudes or expectations about the relationship between one's appearance and acceptable forms of reference. We argue for explicitly modeling individual differences in pronoun selection and present a probabilistic graphical modeling approach based on the nested Chinese Restaurant Franchise Process (nCRFP) (Ahmed et al., 2013) to account for flexible pronominal reference such as chosen names and neopronouns while moving beyond form-to-meaning mappings and without lexical co-occurrence statistics to learn referring expressions, as in contemporary language models. We show that such a model can account for variability in how quickly pronouns or names are integrated into symbolic knowledge and can empower computational systems to be both flexible and respectful of queer people with diverse gender expression.

Figures (2)

• Figure 1: Single speaker model
• Figure 2: Community model with many speakers.