Second-Order Semantic Dependency Parsing with End-to-End Neural Networks

TL;DR

This paper introduces a second-order neural network-based semantic dependency parser that models interactions between edges, utilizing mean field and loopy belief propagation algorithms, achieving state-of-the-art results.

Contribution

It presents a novel second-order parsing method that incorporates edge interactions and can be trained end-to-end using neural network implementations of inference algorithms.

Findings

Achieves state-of-the-art performance on semantic dependency parsing

Effectively models interactions between dependency edges

Unfolds inference algorithms as neural network layers

Abstract

Semantic dependency parsing aims to identify semantic relationships between words in a sentence that form a graph. In this paper, we propose a second-order semantic dependency parser, which takes into consideration not only individual dependency edges but also interactions between pairs of edges. We show that second-order parsing can be approximated using mean field (MF) variational inference or loopy belief propagation (LBP). We can unfold both algorithms as recurrent layers of a neural network and therefore can train the parser in an end-to-end manner. Our experiments show that our approach achieves state-of-the-art performance.