Cross-Language Typicality Effects in a Multilingual Large Language Model
eScholarship (California Digital Library) · 18.06.2025
DOI: 10.48448/p7mm-j872 ↗Kurz gefasst
Fachartikel in der Zeitschrift eScholarship (California Digital Library), erschienen am 18.06.2025. Verfasst von Cognitive Science Society 2025 und 6 weiteren Autor:innen.
Abstract (Originalsprache)
The typicality effect is the finding that some members of a category are more central'' and others moreperipheral''. This effect is seminal for understanding the mental representation of concepts. Recently, researchers have looked for typicality effects in the representations learned by machine learning models as evidence of their cognitive alignment. Studies of the typicality effect in Large Language Models (LLMs) have focused on models trained on English corpora and category norms collected from English speakers. Here, we use existing norms to investigate the typicality effect across five languages: English, French, Portuguese, German, and Spanish. We focused on eight categories common across these norms, and asked whether a multilingual LLM, GPT-4o-mini, shows human-like typicality effects across these languages. The results show variation in typicality gradients across languages. Importantly, GPT-4o-mini's typicality judgments show strong alignment with human norms for some languages: English and French. The strong performance for French, in particular, cannot simply be attributed to the representation of that language in the training corpus. We discuss the implications of these findings for future studies exploring alternative model prompting approaches, different languages, and the modeling of new category norms collected using uniform methods.
Schlagwörter
Zitieren
APA
2025, C. S. S., Gupta, S., Haarer, E., Kalnik, M., Mellacheruvu, A., Varma, S., & Vasan, N. (2025). Cross-Language Typicality Effects in a Multilingual Large Language Model. eScholarship (California Digital Library). https://doi.org/10.48448/p7mm-j872
Vancouver
2025 CSS, Gupta S, Haarer E, Kalnik M, Mellacheruvu A, Varma S, et al. Cross-Language Typicality Effects in a Multilingual Large Language Model. eScholarship (California Digital Library). 2025. doi:10.48448/p7mm-j872
BibTeX
@article{2025,
author = {2025, Cognitive Science Society and Gupta, Sneh and Haarer, Ethan and Kalnik, May and Mellacheruvu, Amogh and Varma, Sashank and Vasan, Nikhita},
title = {Cross-Language Typicality Effects in a Multilingual Large Language Model},
journal = {eScholarship (California Digital Library)},
year = {2025},
doi = {10.48448/p7mm-j872}
}
EUVIMED