NusaWrites: Constructing High-Quality Corpora for Underrepresented and Extremely Low-Resource Languages

Samuel Cahyawijaya, Holy Lovenia, Fajri Koto, Dea Adhista, Emmanuel Dave, Sarah Oktavianti, Salsabil Maulana Akbar, Jhonson Lee, Nuur Shadieq, Tjeng Wawan Cenggoro, Hanung Linuwih, Bryan Wilie, Galih Pradipta Muridan, Alham Fikri Aji, Genta Indra Winata, David Moeljadi, Ayu Purwarianti, Pascale Fung

Keywords: underrepresented and extremely low-resource language, corpus collection method, cultural relevance, NLP benchmark

TL;DR: Our work surmises that although online scraping is effective for high-resource languages, it is not ideal for many low-resource languages. Moreover, our empirical analysis concludes the need to extend the coverage of LMs to underrepresented languages

Abstract: Democratizing access to natural language processing (NLP) technology is crucial, especially for underrepresented and extremely low-resource languages. Previous research has focused on developing labeled and unlabeled corpora for these languages through online scraping and document translation. While these methods have proven effective and cost-efficient, we have identified limitations in the resulting corpora, including a lack of lexical diversity and cultural relevance to local communities. To address this gap, we conduct a case study on Indonesian local languages. We compare the effectiveness of online scraping, human translation, and paragraph writing by native speakers in constructing datasets. Our findings demonstrate that datasets generated through paragraph writing by native speakers exhibit superior quality in terms of lexical diversity and cultural content. In addition, we present the NusaWrites benchmark, encompassing 12 underrepresented and extremely low-resource languages spoken by millions of individuals in Indonesia. Our empirical experiment results using existing multilingual large language models conclude the need to extend these models to more underrepresented languages.