Transition-based Semantic Dependency Parsing with Pointer Networks

TL;DR

This paper introduces a transition system using Pointer Networks for semantic dependency parsing, enhanced with BERT embeddings, achieving state-of-the-art results in producing labeled directed acyclic graphs.

Contribution

It extends transition-based parsing with Pointer Networks to handle semantic graphs and incorporates BERT for improved accuracy, surpassing previous models.

Findings

Outperforms existing transition-based models

Matches best accuracy of graph-based parsers on SemEval 2015

Successfully produces labeled directed acyclic graphs

Abstract

Transition-based parsers implemented with Pointer Networks have become the new state of the art in dependency parsing, excelling in producing labelled syntactic trees and outperforming graph-based models in this task. In order to further test the capabilities of these powerful neural networks on a harder NLP problem, we propose a transition system that, thanks to Pointer Networks, can straightforwardly produce labelled directed acyclic graphs and perform semantic dependency parsing. In addition, we enhance our approach with deep contextualized word embeddings extracted from BERT. The resulting system not only outperforms all existing transition-based models, but also matches the best fully-supervised accuracy to date on the SemEval 2015 Task 18 English datasets among previous state-of-the-art graph-based parsers.