Dependency and Span, Cross-Style Semantic Role Labeling on PropBank and NomBank

TL;DR

This paper introduces a cross-style semantic role labeling framework that unifies dependency and span formalisms, improving their comparability and leveraging their linguistic connections through joint optimization and syntax-aided learning.

Contribution

It proposes a novel cross-style SRL model and a syntax-aided method that enhance both dependency and span SRL performance and comparability.

Findings

Effective on span SRL benchmarks

Improves dependency SRL results

Unifies dependency and span SRL outputs

Abstract

The latest developments in neural semantic role labeling (SRL) have shown great performance improvements with both the dependency and span formalisms/styles. Although the two styles share many similarities in linguistic meaning and computation, most previous studies focus on a single style. In this paper, we define a new cross-style semantic role label convention and propose a new cross-style joint optimization model designed around the most basic linguistic meaning of a semantic role, providing a solution to make the results of the two styles more comparable and allowing both formalisms of SRL to benefit from their natural connections in both linguistics and computation. Our model learns a general semantic argument structure and is capable of outputting in either style. Additionally, we propose a syntax-aided method to uniformly enhance the learning of both dependency and span representations. Experiments show that the proposed methods are effective on both span and dependency SRL benchmarks.