Abstract
This paper presents the system used in our submission to the CoNLL 2019 shared task: Cross-Framework Meaning Representation Parsing. Our system is a graph-based parser which combines an extended pointer-generator network that generates nodes and a second-order mean field variational inference module that predicts edges. Our system achieved 1st and 2nd place for the DM and PSD frameworks respectively on the in-framework ranks and achieved 3rd place for the DM framework on the cross-framework ranks.
Cite
CITATION STYLE
Wang, X., Liu, Y., Jia, Z., Jiang, C., & Tu, K. (2020). ShanghaiTech at MRP 2019: Sequence-to-graph transduction with second-order edge inference for cross-framework meaning representation parsing. In CoNLL 2019 - SIGNLL Conference on Computational Natural Language Learning, Proceedings of the Shared Task on Cross-Framework Meaning Representation Parsing at the 2019 Conference on Natural Language Learning (pp. 55–65). Association for Computational Linguistics. https://doi.org/10.18653/v1/K19-2005
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
