under-resourced-language

This study introduces MultiBanFakeDetect, a novel multimodal dataset for Bangla fake news detection, combining textual and visual information. It features TextFakeNet for text analysis and MultiFusionFake for integrating multimodal data.

This repository contains code for the paper "End-to-End Speech Recognition of Tamil Language", published in the Intelligent Automation & Soft Computing Journal, 2021

Official repository for lemmatization and POS-tagging of under-resourced languages with LLM - LoResLM@EACL 2026

I keep creating neologisms for Quechua and Aymara languages, now, dealing with the demand instead of the offer how I did the first time

Universal Dependencies (UD) parser for Cuzco/Southern Quechua (an under-resourced language of Peru). Trained on a Quechua treebank using 10-fold cross-validation with Stanford NLP's Stanza.