An AMR Aligner Tuned by Transition-based Parser

TL;DR

This paper introduces a novel AMR aligner and transition system, improving alignment accuracy and parsing performance by tuning with an oracle parser, leading to higher Smatch scores.

Contribution

The work presents a resource-rich aligner and a transition system for AMR parsing, along with a tuned parser that outperforms previous rule-based aligners.

Findings

Aligner achieves higher F1 score than previous rule-based methods.

Parser ensemble reaches 68.4 Smatch F1 score using only words and POS tags.

The approach improves both alignment quality and parsing accuracy.

Abstract

In this paper, we propose a new rich resource enhanced AMR aligner which produces multiple alignments and a new transition system for AMR parsing along with its oracle parser. Our aligner is further tuned by our oracle parser via picking the alignment that leads to the highest-scored achievable AMR graph. Experimental results show that our aligner outperforms the rule-based aligner in previous work by achieving higher alignment F1 score and consistently improving two open-sourced AMR parsers. Based on our aligner and transition system, we develop a transition-based AMR parser that parses a sentence into its AMR graph directly. An ensemble of our parsers with only words and POS tags as input leads to 68.4 Smatch F1 score.