Word Alignment-Based Evaluation of Uniform Meaning Representations

TL;DR

This paper introduces a new node-matching algorithm for evaluating uniform meaning representations that leverages word alignments, improving interpretability over existing methods like smatch.

Contribution

The paper presents a novel node-matching algorithm that uses word alignments for better comparison of meaning representations, avoiding NP-hard search issues.

Findings

The new method is more intuitive and interpretable than smatch.

It effectively compares multiple UMRs of the same sentence.

The script implementing the method is freely available.

Abstract

Comparison and evaluation of graph-based representations of sentence meaning is a challenge because competing representations of the same sentence may have different number of nodes, and it is not obvious which nodes should be compared to each other. Existing approaches favor node mapping that maximizes $F_1$ score over node relations and attributes, regardless whether the similarity is intentional or accidental; consequently, the identified mismatches in values of node attributes are not useful for any detailed error analysis. We propose a node-matching algorithm that allows comparison of multiple Uniform Meaning Representations (UMR) of one sentence and that takes advantage of node-word alignments, inherently available in UMR. We compare it with previously used approaches, in particular smatch (the de-facto standard in AMR evaluation), and argue that sensitivity to word alignment makes the comparison of meaning representations more intuitive and interpretable, while avoiding the NP-hard search problem inherent in smatch. A script implementing the method is freely available.