We present a model for semantic proto-role labeling (SPRL) using an adapted bidirectional LSTM encoding strategy that we call "Neural-Davidsonian": predicate-argument structure is represented as pairs of hidden states corresponding to predicate and argument head tokens of the input sequence. We demonstrate: (1) state-of-the-art results in SPRL, and (2) that our network naturally shares parameters between attributes, allowing for learning new attribute types with limited added supervision.
Cite
@article{arxiv.1804.07976,
title = {Neural-Davidsonian Semantic Proto-role Labeling},
author = {Rachel Rudinger and Adam Teichert and Ryan Culkin and Sheng Zhang and Benjamin Van Durme},
journal= {arXiv preprint arXiv:1804.07976},
year = {2019}
}
Comments
Accepted to EMNLP 2018; errata corrected in Appendix Tables 7 and 8