Pruning for Performance: Efficient Idiom and Metaphor Classification in Low-Resource Konkani Using mBERT

TL;DR

This paper develops an efficient NLP model for metaphor and idiom classification in low-resource Konkani by combining mBERT with pruning techniques, achieving high accuracy with reduced computational complexity.

Contribution

It introduces a hybrid model with attention head pruning for figurative language classification in Konkani, a low-resource language, with a newly annotated dataset.

Findings

Achieved 78% accuracy in metaphor classification.

Achieved 83% accuracy in idiom classification.

Demonstrated effectiveness of pruning for low-resource NLP tasks.

Abstract

In this paper, we address the persistent challenges that figurative language expressions pose for natural language processing (NLP) systems, particularly in low-resource languages such as Konkani. We present a hybrid model that integrates a pre-trained Multilingual BERT (mBERT) with a bidirectional LSTM and a linear classifier. This architecture is fine-tuned on a newly introduced annotated dataset for metaphor classification, developed as part of this work. To improve the model's efficiency, we implement a gradient-based attention head pruning strategy. For metaphor classification, the pruned model achieves an accuracy of 78%. We also applied our pruning approach to expand on an existing idiom classification task, achieving 83% accuracy. These results demonstrate the effectiveness of attention head pruning for building efficient NLP tools in underrepresented languages.