Work

A Simple and Accurate Syntax-Agnostic Neural Model for Dependency-based Semantic Role Labeling

Year: 2017

Type: conference-paper

Abstract: We introduce a simple and accurate neural model for dependency-based semantic role labeling.Our model predicts predicate-argument dependencies relying on states of a bidirectional LSTM encoder.The semantic role labeler achieves competitive performance on English, even without any kind of syntactic information and only using local inference.However, when automatically predicted partof-speech tags are provided as input, it substantially outperforms all previous local models and approaches the best reported results on the English CoNLL-2009 dataset.We also consider Chinese, Czech and Spanish where our approach also achieves competitive results.Syntactic parsers are unreliable on out-of-domain data, so standard (i.e., syntactically-informed) SRL models are hindered when tested in this setting.Our syntax-agnostic model appears more robust, resulting in the best reported results on standard out-of-domain test sets.

Source: Conference on Computational Natural Language Learning

DOI: 10.18653/v1/k17-1041

Authors: Diego Marcheggiani, Anton Frolov, Ivan Titov

Institutions: University of Amsterdam, Yandex (Russia), University of Glasgow, University of Edinburgh

Language: English

Topic: Natural Language Processing Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funders: National Science Foundation, Amazon Web Services, Nederlandse Organisatie voor Wetenschappelijk Onderzoek, H2020 European Research Council

Awards: Induction of Broad-Coverage Semantic Parsers, Scaling Semantic Parsing to Unrestricted Domains