UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment
Joseph Marvin Imperial, Abdullah Barayan, Regina Stodden, Rodrigo Wilkens, Ricardo Mu{\~n}oz S{\'a}nchez, Lingyun Gao, Melissa Torgbi, Dawn Knight
Abstract
We introduce UniversalCEFR, a large-scale multilingual multidimensional dataset of texts annotated according to the CEFR (Common European Framework of Reference) scale in 13 languages. To enable open research in both automated readability and language proficiency assessment, UniversalCEFR comprises 505,807 CEFR-labeled texts curated from educational and learner-oriented resources, standardized into a unified data format to support consistent processing, analysis, and modeling across tasks and languages. To demonstrate its utility, we conduct benchmark experiments using three modelling paradigms: a) linguistic feature-based classification, b) fine-tuning pre-trained LLMs, and c) descriptor-based prompting of instruction-tuned LLMs. Our results further support using linguistic features and fine-tuning pretrained models in multilingual CEFR level assessment. Overall, UniversalCEFR aims to establish best practices in data distribution in language proficiency research by standardising dataset formats and promoting their accessibility to the global research community.
BibTeX
@inproceedings{emnlp2025_universalcefrena,
title = {UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment},
author = {Joseph Marvin Imperial and Abdullah Barayan and Regina Stodden and Rodrigo Wilkens and Ricardo Mu{\~n}oz S{\'a}nchez and Lingyun Gao and Melissa Torgbi and Dawn Knight and Gail Forey and Reka R. Jablonkai and Ekaterina Kochmar and Robert Joshua Reynolds and Eug{\'e}nio Ribeiro and Horacio Saggion and Elena Volodina and Sowmya Vajjala and Thomas Fran{\c{c}}ois and Fernando Alva-Manchego and Harish Tayyar Madabushi},
booktitle = {EMNLP 2025},
year = {2025}
}