MultiBLiMP 1.0: A Massively Multilingual Benchmark of Linguistic Minimal Pairs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jumelet, Jaap, Weissweiler, Leonie, Nivre, Joakim, Bisazza, Arianna
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911633821925376
author Jumelet, Jaap
Weissweiler, Leonie
Nivre, Joakim
Bisazza, Arianna
author_facet Jumelet, Jaap
Weissweiler, Leonie
Nivre, Joakim
Bisazza, Arianna
contents We introduce MultiBLiMP 1.0, a massively multilingual benchmark of linguistic minimal pairs, covering 101 languages and 2 types of subject-verb agreement, containing more than 128,000 minimal pairs. Our minimal pairs are created using a fully automated pipeline, leveraging the large-scale linguistic resources of Universal Dependencies and UniMorph. MultiBLiMP 1.0 evaluates abilities of LLMs at an unprecedented multilingual scale, and highlights the shortcomings of the current state-of-the-art in modelling low-resource languages.
format Preprint
id arxiv_https___arxiv_org_abs_2504_02768
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MultiBLiMP 1.0: A Massively Multilingual Benchmark of Linguistic Minimal Pairs
Jumelet, Jaap
Weissweiler, Leonie
Nivre, Joakim
Bisazza, Arianna
Computation and Language
We introduce MultiBLiMP 1.0, a massively multilingual benchmark of linguistic minimal pairs, covering 101 languages and 2 types of subject-verb agreement, containing more than 128,000 minimal pairs. Our minimal pairs are created using a fully automated pipeline, leveraging the large-scale linguistic resources of Universal Dependencies and UniMorph. MultiBLiMP 1.0 evaluates abilities of LLMs at an unprecedented multilingual scale, and highlights the shortcomings of the current state-of-the-art in modelling low-resource languages.
title MultiBLiMP 1.0: A Massively Multilingual Benchmark of Linguistic Minimal Pairs
topic Computation and Language
url https://arxiv.org/abs/2504.02768