Cross-lingual Approaches for Task-specific Dialogue Act Recognition

TL;DR

This paper presents a transfer learning approach using cross-lingual models to improve dialogue act recognition in low-resource languages and domains, outperforming existing methods.

Contribution

It introduces a novel transfer learning method that combines CNN and self-attention embeddings for cross-lingual dialogue act recognition.

Findings

Best results achieved by combining multiple transferred sources.

Significant improvement over existing cross-lingual approaches.

Validated on two target languages and domains.

Abstract

In this paper we exploit cross-lingual models to enable dialogue act recognition for specific tasks with a small number of annotations. We design a transfer learning approach for dialogue act recognition and validate it on two different target languages and domains. We compute dialogue turn embeddings with both a CNN and multi-head self-attention model and show that the best results are obtained by combining all sources of transferred information. We further demonstrate that the proposed methods significantly outperform related cross-lingual DA recognition approaches.