IndT5: A Text-to-Text Transformer for 10 Indigenous Languages

TL;DR

IndT5 is the first Transformer language model designed for ten Indigenous languages, utilizing a new dataset, IndCorpus, and applied to machine translation tasks involving Spanish and Indigenous languages.

Contribution

This work introduces IndT5, a novel Transformer model for Indigenous languages, along with IndCorpus dataset, addressing low-resource language modeling and translation.

Findings

IndT5 outperforms baseline models in translation quality.

IndCorpus enables effective training for low-resource languages.

The model is publicly available for further research.

Abstract

Transformer language models have become fundamental components of natural language processing based pipelines. Although several Transformer models have been introduced to serve many languages, there is a shortage of models pre-trained for low-resource and Indigenous languages. In this work, we introduce IndT5, the first Transformer language model for Indigenous languages. To train IndT5, we build IndCorpus--a new dataset for ten Indigenous languages and Spanish. We also present the application of IndT5 to machine translation by investigating different approaches to translate between Spanish and the Indigenous languages as part of our contribution to the AmericasNLP 2021 Shared Task on Open Machine Translation. IndT5 and IndCorpus are publicly available for research