Learning Disentangled Semantic Representations for Zero-Shot Cross-Lingual Transfer in Multilingual Machine Reading Comprehension

TL;DR

This paper introduces a novel framework with a Siamese Semantic Disentanglement Model to improve zero-shot cross-lingual transfer in multilingual machine reading comprehension by separating semantic and syntactic information.

Contribution

It proposes a new semantic disentanglement approach with tailored losses to enhance zero-shot transfer in multilingual MRC tasks.

Findings

Outperforms mBERT and XLM-100 based models on XQuAD, MLQA, and TyDi QA datasets.

Effectively disentangles semantics from syntax in multilingual representations.

Improves answer span accuracy in low-resource languages.

Abstract

Multilingual pre-trained models are able to zero-shot transfer knowledge from rich-resource to low-resource languages in machine reading comprehension (MRC). However, inherent linguistic discrepancies in different languages could make answer spans predicted by zero-shot transfer violate syntactic constraints of the target language. In this paper, we propose a novel multilingual MRC framework equipped with a Siamese Semantic Disentanglement Model (SSDM) to disassociate semantics from syntax in representations learned by multilingual pre-trained models. To explicitly transfer only semantic knowledge to the target language, we propose two groups of losses tailored for semantic and syntactic encoding and disentanglement. Experimental results on three multilingual MRC datasets (i.e., XQuAD, MLQA, and TyDi QA) demonstrate the effectiveness of our proposed approach over models based on mBERT and XLM-100. Code is available at:https://github.com/wulinjuan/SSDM_MRC.