We study how different frame annotations complement one another when learning continuous lexical semantics. We learn the representations from a tensorized skip-gram model that consistently encodes syntactic-semantic content better, with multiple 10% gains over baselines.
CITATION STYLE
Ferraro, F., Poliak, A., Cotterell, R., & Van Durme, B. (2017). Frame-based continuous lexical semantics through exponential family tensor factorization and semantic proto-roles. In *SEM 2017 - 6th Joint Conference on Lexical and Computational Semantics, Proceedings (pp. 97–103). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/s17-1011
Mendeley helps you to discover research relevant for your work.