Zero-Resource Multi-Dialectal Arabic Natural Language Understanding

TL;DR

This paper explores zero-resource dialectal Arabic understanding by using self-training on unlabeled dialect data to improve performance on NER, POS tagging, and sarcasm detection, achieving state-of-the-art results.

Contribution

It introduces a self-training approach with unlabeled dialectal Arabic data to enhance zero-shot transfer from standard Arabic, addressing resource scarcity.

Findings

Self-training improves NER F1 by ~10%

Enhances POS tagging accuracy by 2%

Boosts sarcasm detection F1 by 4.5%

Abstract

A reasonable amount of annotated data is required for fine-tuning pre-trained language models (PLM) on downstream tasks. However, obtaining labeled examples for different language varieties can be costly. In this paper, we investigate the zero-shot performance on Dialectal Arabic (DA) when fine-tuning a PLM on modern standard Arabic (MSA) data only -- identifying a significant performance drop when evaluating such models on DA. To remedy such performance drop, we propose self-training with unlabeled DA data and apply it in the context of named entity recognition (NER), part-of-speech (POS) tagging, and sarcasm detection (SRD) on several DA varieties. Our results demonstrate the effectiveness of self-training with unlabeled DA data: improving zero-shot MSA-to-DA transfer by as large as $\sim$10\% F$_1$ (NER), 2\% accuracy (POS tagging), and 4.5\% F$_1$ (SRD). We conduct an ablation experiment and show that the performance boost observed directly results from the unlabeled DA examples used for self-training. Our work opens up opportunities for leveraging the relatively abundant labeled MSA datasets to develop DA models for zero and low-resource dialects. We also report new state-of-the-art performance on all three tasks and open-source our fine-tuned models for the research community.