sandrarrey/mBERT_readability_GL

This model is a fine-tuned version of mBERT (bert-base-multilingual-cased) for text readability classification in Galician (ILG/RAG spelling). It was trained on an in-house dataset that includes the Corlega corpus (https://zenodo.org/records/15441342) and the iRead4Skills Dataset 1(https://zenodo.org/records/13127399). It was then evaluated using a set of texts from the Corlega corpus (https://zenodo.org/record/15441342). The model presented here is the best result of the experiments (mBERT GL+ES+PT+FR) conducted in the following paper:

Rodríguez Rey, S., & García, M. (2025). Clasificación automática de textos por niveis de lecturabilidade: Recursos e modelos para o galego. Linguamática, 17(2), 33–56. https://doi.org/10.21814/lm.17.2.488