# Robust Incremental Neural Semantic Graph Parsing

## TL;DR

This paper introduces a neural transition-based parser for semantic graphs, achieving high accuracy and speed on MRS and AMR benchmarks, and demonstrating the advantages of MRS as a semantic representation.

## Contribution

It presents the first full-coverage neural parser for MRS, jointly predicting graphs with unlexicalized predicates, and outperforms baselines in accuracy and efficiency.

## Key findings

- Achieves 86.69% Smatch score on MRS, surpassing AMR upper-bound.
- More accurate than attention-based baselines on MRS and AMR.
- GPU batch processing significantly speeds up parsing.

## Abstract

Parsing sentences to linguistically-expressive semantic representations is a key goal of Natural Language Processing. Yet statistical parsing has focused almost exclusively on bilexical dependencies or domain-specific logical forms. We propose a neural encoder-decoder transition-based parser which is the first full-coverage semantic graph parser for Minimal Recursion Semantics (MRS). The model architecture uses stack-based embedding features, predicting graphs jointly with unlexicalized predicates and their token alignments. Our parser is more accurate than attention-based baselines on MRS, and on an additional Abstract Meaning Representation (AMR) benchmark, and GPU batch processing makes it an order of magnitude faster than a high-precision grammar-based parser. Further, the 86.69% Smatch score of our MRS parser is higher than the upper-bound on AMR parsing, making MRS an attractive choice as a semantic representation.