IndoNLU: Benchmark and Resources for Evaluating Indonesian Natural Language Understanding

TL;DR

This paper introduces IndoNLU, a comprehensive benchmark and resource suite for Indonesian natural language understanding, including diverse datasets, pre-trained models, and evaluation frameworks to accelerate NLP research for Indonesian.

Contribution

It provides the first extensive set of datasets, pre-trained models, and benchmarking tools specifically for Indonesian NLP tasks, filling a critical resource gap.

Findings

Baseline models established for all twelve tasks.

Diverse datasets from multiple domains and styles.

Pre-trained IndoBERT models trained on large Indonesian corpus.

Abstract

Although Indonesian is known to be the fourth most frequently used language over the internet, the research progress on this language in the natural language processing (NLP) is slow-moving due to a lack of available resources. In response, we introduce the first-ever vast resource for the training, evaluating, and benchmarking on Indonesian natural language understanding (IndoNLU) tasks. IndoNLU includes twelve tasks, ranging from single sentence classification to pair-sentences sequence labeling with different levels of complexity. The datasets for the tasks lie in different domains and styles to ensure task diversity. We also provide a set of Indonesian pre-trained models (IndoBERT) trained from a large and clean Indonesian dataset Indo4B collected from publicly available sources such as social media texts, blogs, news, and websites. We release baseline models for all twelve tasks, as well as the framework for benchmark evaluation, and thus it enables everyone to benchmark their system performances.