Fast semantic parsing with well-typedness guarantees

TL;DR

This paper introduces new A* and transition-based parsers for AM dependency parsing that ensure well-typedness and significantly boost speed without sacrificing accuracy.

Contribution

It presents the first well-typedness guaranteeing parsers for AM dependency parsing, achieving up to 1000x faster performance while maintaining or improving accuracy.

Findings

Parsing speed increased by up to 3 orders of magnitude.

Maintained or improved parsing accuracy.

First parsers with well-typedness guarantees for AM dependency parsing.

Abstract

AM dependency parsing is a linguistically principled method for neural semantic parsing with high accuracy across multiple graphbanks. It relies on a type system that models semantic valency but makes existing parsers slow. We describe an A* parser and a transition-based parser for AM dependency parsing which guarantee well-typedness and improve parsing speed by up to 3 orders of magnitude, while maintaining or improving accuracy.