Abstract
Meaning Representation (AMR) parsing aims at converting sentences into AMR representations. These are graphs and not trees because AMR supports reentrancies (nodes with more than one parent). Following previous findings on the importance of reentrancies for AMR, we empirically find and discuss several linguistic phenomena responsible for reentrancies in AMR, some of which have not received attention before. We categorize the types of errors AMR parsers make with respect to reentrancies. Furthermore, we find that correcting these errors provides an increase of up to 5% Smatch in parsing performance and 20% in reentrancy prediction.
Cite
CITATION STYLE
Szubert, I., Damonte, M., Cohen, S. B., & Steedman, M. (2020). The role of reentrancies in abstract meaning representation parsing. In Findings of the Association for Computational Linguistics Findings of ACL: EMNLP 2020 (pp. 2198–2207). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2020.findings-emnlp.199
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
