Establishing Baselines for Text Classification in Low-Resource Languages

TL;DR

This paper introduces new datasets, pretrained models, and a degradation benchmark for low-resource language Filipino text classification, facilitating better evaluation and comparison of models in low-data scenarios.

Contribution

It provides the first benchmark datasets, improved pretrained models, and a degradation test method specifically for Filipino text classification in low-resource settings.

Findings

Pretrained models show resilience to data reduction.

New datasets enable standardized evaluation.

Degradation test reveals model robustness.

Abstract

While transformer-based finetuning techniques have proven effective in tasks that involve low-resource, low-data environments, a lack of properly established baselines and benchmark datasets make it hard to compare different approaches that are aimed at tackling the low-resource setting. In this work, we provide three contributions. First, we introduce two previously unreleased datasets as benchmark datasets for text classification and low-resource multilabel text classification for the low-resource language Filipino. Second, we pretrain better BERT and DistilBERT models for use within the Filipino setting. Third, we introduce a simple degradation test that benchmarks a model's resistance to performance degradation as the number of training samples are reduced. We analyze our pretrained model's degradation speeds and look towards the use of this method for comparing models aimed at operating within the low-resource setting. We release all our models and datasets for the research community to use.