Mutlitask Learning for Cross-Lingual Transfer of Semantic Dependencies

TL;DR

This paper presents a multitask learning approach combined with annotation projection to develop semantic dependency parsers for low-resource languages, demonstrating improved performance over single-task methods.

Contribution

It introduces a novel multitask learning framework that leverages syntactic parsing as an auxiliary task to enhance cross-lingual semantic dependency transfer.

Findings

Best model improves F1 score by 1.8 in-domain

Out-of-domain F1 score improves by 2.5

Syntactic-semantic dependency direction match is crucial

Abstract

We describe a method for developing broad-coverage semantic dependency parsers for languages for which no semantically annotated resource is available. We leverage a multitask learning framework coupled with an annotation projection method. We transfer supervised semantic dependency parse annotations from a rich-resource language to a low-resource language through parallel data, and train a semantic parser on projected data. We make use of supervised syntactic parsing as an auxiliary task in a multitask learning framework, and show that with different multitask learning settings, we consistently improve over the single-task baseline. In the setting in which English is the source, and Czech is the target language, our best multitask model improves the labeled F1 score over the single-task baseline by 1.8 in the in-domain SemEval data (Oepen et al., 2015), as well as 2.5 in the out-of-domain test set. Moreover, we observe that syntactic and semantic dependency direction match is an important factor in improving the results.