Establishing Baselines for Text Classification in Low-Resource Languages
This work addresses the need for standardized evaluation in low-resource language processing, though it is incremental as it builds on existing transformer methods.
The paper tackles the problem of lacking baselines and benchmarks for text classification in low-resource languages by introducing two new datasets for Filipino, pretraining BERT and DistilBERT models, and proposing a degradation test to measure model performance as training data is reduced.
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.