On the cross-lingual transferability of multilingual prototypical models across NLU tasks

TL;DR

This paper explores how combining few-shot learning, prototypical neural networks, and multilingual Transformers enhances cross-lingual transfer in natural language understanding, especially for low-resource languages.

Contribution

It introduces a novel approach integrating meta-learning with prototypical networks and multilingual Transformers to improve transferability across languages.

Findings

Significant performance improvements in low-resource language transfer.

Meta-learning enables generalization of latent language representations.

Approach outperforms traditional transfer methods on MultiATIS++ corpus.

Abstract

Supervised deep learning-based approaches have been applied to task-oriented dialog and have proven to be effective for limited domain and language applications when a sufficient number of training examples are available. In practice, these approaches suffer from the drawbacks of domain-driven design and under-resourced languages. Domain and language models are supposed to grow and change as the problem space evolves. On one hand, research on transfer learning has demonstrated the cross-lingual ability of multilingual Transformers-based models to learn semantically rich representations. On the other, in addition to the above approaches, meta-learning have enabled the development of task and language learning algorithms capable of far generalization. Through this context, this article proposes to investigate the cross-lingual transferability of using synergistically few-shot learning with prototypical neural networks and multilingual Transformers-based models. Experiments in natural language understanding tasks on MultiATIS++ corpus shows that our approach substantially improves the observed transfer learning performances between the low and the high resource languages. More generally our approach confirms that the meaningful latent space learned in a given language can be can be generalized to unseen and under-resourced ones using meta-learning.