Salvato in:
Dettagli Bibliografici
Autori principali: Chang, Yuan, Qu, Jiaming, Li, Zhu
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2604.04732
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910105870532608
author Chang, Yuan
Qu, Jiaming
Li, Zhu
author_facet Chang, Yuan
Qu, Jiaming
Li, Zhu
contents Large language models (LLMs) are often described as multilingual because they can understand and respond in many languages. However, speaking a language is not the same as reasoning within a culture. This distinction motivates a critical question: do LLMs truly conduct culture-aware reasoning? This paper presents a preliminary computational audit of cultural inclusivity in a creative writing task. We empirically examine whether LLMs act as culturally diverse creative partners or merely as cultural translators that leverage a dominant conceptual framework with localized expressions. Using a metaphor generation task spanning five cultural settings and several abstract concepts as a case study, we find that the model exhibits stereotyped metaphor usage for certain settings, as well as Western defaultism. These findings suggest that merely prompting an LLM with a cultural identity does not guarantee culturally grounded reasoning.
format Preprint
id arxiv_https___arxiv_org_abs_2604_04732
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Metaphors We Compute By: A Computational Audit of Cultural Translation vs. Thinking in LLMs
Chang, Yuan
Qu, Jiaming
Li, Zhu
Computation and Language
Artificial Intelligence
Large language models (LLMs) are often described as multilingual because they can understand and respond in many languages. However, speaking a language is not the same as reasoning within a culture. This distinction motivates a critical question: do LLMs truly conduct culture-aware reasoning? This paper presents a preliminary computational audit of cultural inclusivity in a creative writing task. We empirically examine whether LLMs act as culturally diverse creative partners or merely as cultural translators that leverage a dominant conceptual framework with localized expressions. Using a metaphor generation task spanning five cultural settings and several abstract concepts as a case study, we find that the model exhibits stereotyped metaphor usage for certain settings, as well as Western defaultism. These findings suggest that merely prompting an LLM with a cultural identity does not guarantee culturally grounded reasoning.
title Metaphors We Compute By: A Computational Audit of Cultural Translation vs. Thinking in LLMs
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2604.04732