Two CFG Nahuatl for automatic corpora expansion

TL;DR

This paper introduces two CFG-based methods to generate artificial Nawatl sentences, significantly expanding the corpus to improve embedding learning and semantic similarity evaluation for this low-resource language.

Contribution

The paper presents two novel CFGs for Nawatl, enabling corpus expansion and improved embedding learning in a low-resource language context.

Findings

Expanded corpus improves embedding quality

Artificial data enhances semantic similarity task performance

Economic embeddings outperform some large language models

Abstract

The aim of this article is to introduce two Context-Free Grammars (CFG) for Nawatl Corpora expansion. Nawatl is an Amerindian language (it is a National Language of Mexico) of the $\pi$-language type, i.e. a language with few digital resources. For this reason the corpora available for the learning of Large Language Models (LLMs) are virtually non-existent, posing a significant challenge. The goal is to produce a substantial number of syntactically valid artificial Nawatl sentences and thereby to expand the corpora for the purpose of learning non contextual embeddings. For this objective, we introduce two new Nawatl CFGs and use them in generative mode. Using these grammars, it is possible to expand Nawatl corpus significantly and subsequently to use it to learn embeddings and to evaluate their relevance in a sentences semantic similarity task. The results show an improvement compared to the results obtained using only the original corpus without artificial expansion, and also demonstrate that economic embeddings often perform better than some LLMs.