SemBleu: A Robust Metric for AMR Parsing Evaluation

Linfeng Song, Daniel Gildea


Abstract
Evaluating AMR parsing accuracy involves comparing pairs of AMR graphs. The major evaluation metric, SMATCH (Cai and Knight, 2013), searches for one-to-one mappings between the nodes of two AMRs with a greedy hill-climbing algorithm, which leads to search errors. We propose SEMBLEU, a robust metric that extends BLEU (Papineni et al., 2002) to AMRs. It does not suffer from search errors and considers non-local correspondences in addition to local ones. SEMBLEU is fully content-driven and punishes situations where a system’s output does not preserve most information from the input. Preliminary experiments on both sentence and corpus levels show that SEMBLEU has slightly higher consistency with human judgments than SMATCH. Our code is available at http://github.com/freesunshine0316/sembleu.
Anthology ID:
P19-1446
Volume:
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2019
Address:
Florence, Italy
Editors:
Anna Korhonen, David Traum, Lluís Màrquez
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4547–4552
Language:
URL:
https://aclanthology.org/P19-1446
DOI:
10.18653/v1/P19-1446
Bibkey:
Cite (ACL):
Linfeng Song and Daniel Gildea. 2019. SemBleu: A Robust Metric for AMR Parsing Evaluation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4547–4552, Florence, Italy. Association for Computational Linguistics.
Cite (Informal):
SemBleu: A Robust Metric for AMR Parsing Evaluation (Song & Gildea, ACL 2019)
Copy Citation:
PDF:
https://aclanthology.org/P19-1446.pdf
Code
 freesunshine0316/sembleu +  additional community code
Data
RARE