MaiBERT: A Pre-training Corpus and Language Model for Low-Resourced Maithili Language

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yadav, Sumit, Yadav, Raju Kumar, Maskey, Utsav, Kashyap, Gautam Siddharth, Gautam, Ganesh, Naseem, Usman
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910008111792128
author Yadav, Sumit
Yadav, Raju Kumar
Maskey, Utsav
Kashyap, Gautam Siddharth
Gautam, Ganesh
Naseem, Usman
author_facet Yadav, Sumit
Yadav, Raju Kumar
Maskey, Utsav
Kashyap, Gautam Siddharth
Gautam, Ganesh
Naseem, Usman
contents Natural Language Understanding (NLU) for low-resource languages remains a major challenge in NLP due to the scarcity of high-quality data and language-specific models. Maithili, despite being spoken by millions, lacks adequate computational resources, limiting its inclusion in digital and AI-driven applications. To address this gap, we introducemaiBERT, a BERT-based language model pre-trained specifically for Maithili using the Masked Language Modeling (MLM) technique. Our model is trained on a newly constructed Maithili corpus and evaluated through a news classification task. In our experiments, maiBERT achieved an accuracy of 87.02%, outperforming existing regional models like NepBERTa and HindiBERT, with a 0.13% overall accuracy gain and 5-7% improvement across various classes. We have open-sourced maiBERT on Hugging Face enabling further fine-tuning for downstream tasks such as sentiment analysis and Named Entity Recognition (NER).
format Preprint
id arxiv_https___arxiv_org_abs_2509_15048
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MaiBERT: A Pre-training Corpus and Language Model for Low-Resourced Maithili Language
Yadav, Sumit
Yadav, Raju Kumar
Maskey, Utsav
Kashyap, Gautam Siddharth
Gautam, Ganesh
Naseem, Usman
Computation and Language
Natural Language Understanding (NLU) for low-resource languages remains a major challenge in NLP due to the scarcity of high-quality data and language-specific models. Maithili, despite being spoken by millions, lacks adequate computational resources, limiting its inclusion in digital and AI-driven applications. To address this gap, we introducemaiBERT, a BERT-based language model pre-trained specifically for Maithili using the Masked Language Modeling (MLM) technique. Our model is trained on a newly constructed Maithili corpus and evaluated through a news classification task. In our experiments, maiBERT achieved an accuracy of 87.02%, outperforming existing regional models like NepBERTa and HindiBERT, with a 0.13% overall accuracy gain and 5-7% improvement across various classes. We have open-sourced maiBERT on Hugging Face enabling further fine-tuning for downstream tasks such as sentiment analysis and Named Entity Recognition (NER).
title MaiBERT: A Pre-training Corpus and Language Model for Low-Resourced Maithili Language
topic Computation and Language
url https://arxiv.org/abs/2509.15048