UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment

Imperial, Joseph Marvin;Barayan, Abdullah;Stodden, Regina;Souza Wilkens, Rodrigo;Tayyar Madabushi, Harish;et.al.
(2025) 2025 Conference on Empirical Methods in Natural Language Processing — Location: Suzhou, China (4.November.2025)

Files

2025emnlp-main491.pdf
  • Open Access
  • Adobe PDF
  • 1.34 MB
Download

Details

Authors
  • Imperial, Joseph Marvin
    Author
  • Barayan, Abdullah
    Author
  • Stodden, Regina
    Author
  • Souza Wilkens, Rodrigoorcid-logoUCLouvain
    Author
  • Gao, LingyunUCLouvain
    Author
  • François, Thomasorcid-logoUCLouvain
    Author
  • Tayyar Madabushi, Harish
    Author
Show more
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.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Imperial, J. M., Barayan, A., Stodden, R., Souza Wilkens, R., Muñoz Sánchez, R., Gao, L., Torgbi, M., Knight, D., Forey, G., Jablonkai, R. R., Kochmar, E., Reynolds, R. J., Ribeiro, E., Saggion, H., Volodina, E., Vajjala, S., François, T., Alva-Manchego, F., & Tayyar Madabushi, H. (2025). UniversalCEFR: Enabling Open Multilingual Research on Language Proficiency Assessment. In Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng (ed.), Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing (p. p. 9703–9755). Association for Computational Linguistics. https://doi.org/10.18653/v1/2025.emnlp-main.491