(RSA)^2: A Rhetorical-Strategy-Aware Rational Speech Act Framework for Figurative Language Understanding
Abstract
Figurative language (e.g., irony, hyperbole, understatement) is ubiquitous in human communication, resulting in utterances where the literal and the intended meanings do not match. The Rational Speech Act (RSA) framework, which explicitly models speaker intentions, is the most widespread theory of probabilistic pragmatics, but existing implementations are either unable to account for figurative expressions or require modeling the implicit motivations for using figurative language (e.g., to express joy or annoyance) in a setting-specific way. In this paper, we introduce the \textbf{R}hetorical-\textbf{S}trategy-\textbf{A}ware \textbf{RSA} \rsatwospace framework which models figurative language use by considering a speaker's employed rhetorical strategy. We show that \rsatwospace enables human-compatible interpretations of non-literal utterances without modeling a speaker's motivations for being non-literal.
Keywords: rsa, figurative language
How to Cite:
Spinoso-Di Piano, C., (2025) “(RSA)^2: A Rhetorical-Strategy-Aware Rational Speech Act Framework for Figurative Language Understanding”, Society for Computation in Linguistics 8(1): 39. doi: https://doi.org/10.7275/scil.3171
Downloads:
Download PDF
37 Views
11 Downloads