Recipe for Zero-shot POS Tagging: Is It Useful in Realistic Scenarios?

TL;DR

This research highlights the importance of accurate dataset selection for developing effective zero-shot POS tagging models, particularly, a strong linguistic relationship and high-quality datasets ensure optimal results.

Abstract

POS tagging plays a fundamental role in numerous applications. While POS taggers are highly accurate in well-resourced settings, they lag behind in cases of limited or missing training data. This paper focuses on POS tagging for languages with limited data. We seek to identify the characteristics of datasets that make them favourable for training POS tagging models without using any labelled training data from the target language. This is a zero-shot approach. We compare the accuracies of a multilingual large language model (mBERT) fine-tuned on one or more languages related to the target language. Additionally, we compare these results with models trained directly on the target language itself. We do this for three target low-resource languages. Our research highlights the importance of accurate dataset selection for effective zero-shot POS tagging. Particularly, a strong linguistic relationship and high-quality datasets ensure optimal results. For extremely low-resource languages, zero-shot models prove to be a viable option.

Figures (5)

• Figure 1: Accuracy of fine-tuned models on Afrikaans, represented through learning curves.
• Figure 2: Accuracy of fine-tuned models on Afrikaans, represented through learning curves.
• Figure 3: Accuracy of fine-tuned models on Faroese, represented through learning curves.
• Figure 4: Accuracy of fine-tuned models on Upper Sorbian, represented through learning curves.
• Figure 5: Accuracy of models fine-tuned on the target languages Afrikaans, Faroese, and Upper Sorbian, represented through learning curves accompanied by the peak accuracies of the respective zero-shot models.