$(RSA)^2$: A Rhetorical-Strategy-Aware Rational Speech Act Framework for Figurative Language Understanding
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916790470180864 |
|---|---|
| author | Piano, Cesare Spinoso-Di Austin, David Piantanida, Pablo Cheung, Jackie Chi Kit |
| author_facet | Piano, Cesare Spinoso-Di Austin, David Piantanida, Pablo Cheung, Jackie Chi Kit |
| contents | 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 Rhetorical-Strategy-Aware RSA $(RSA)^2$ framework which models figurative language use by considering a speaker's employed rhetorical strategy. We show that $(RSA)^2$ enables human-compatible interpretations of non-literal utterances without modeling a speaker's motivations for being non-literal. Combined with LLMs, it achieves state-of-the-art performance on the ironic split of PragMega+, a new irony interpretation dataset introduced in this study. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_09301 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | $(RSA)^2$: A Rhetorical-Strategy-Aware Rational Speech Act Framework for Figurative Language Understanding Piano, Cesare Spinoso-Di Austin, David Piantanida, Pablo Cheung, Jackie Chi Kit Computation and Language Artificial Intelligence 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 Rhetorical-Strategy-Aware RSA $(RSA)^2$ framework which models figurative language use by considering a speaker's employed rhetorical strategy. We show that $(RSA)^2$ enables human-compatible interpretations of non-literal utterances without modeling a speaker's motivations for being non-literal. Combined with LLMs, it achieves state-of-the-art performance on the ironic split of PragMega+, a new irony interpretation dataset introduced in this study. |
| title | $(RSA)^2$: A Rhetorical-Strategy-Aware Rational Speech Act Framework for Figurative Language Understanding |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2506.09301 |