Dependency Parsing as MRC-based Span-Span Prediction

TL;DR

This paper introduces a novel dependency parsing method that models span-to-span relations using an MRC framework, improving subtree construction accuracy and span retrieval over traditional word-level approaches.

Contribution

It proposes a span-span prediction approach with an MRC-based span linking module, addressing subtree-level edge construction in dependency parsing.

Findings

Effective on PTB, CTB, and UD benchmarks

Higher recall for span proposals due to MRC framework

Improved accuracy in dependency tree construction

Abstract

Higher-order methods for dependency parsing can partially but not fully address the issue that edges in dependency trees should be constructed at the text span/subtree level rather than word level. In this paper, we propose a new method for dependency parsing to address this issue. The proposed method constructs dependency trees by directly modeling span-span (in other words, subtree-subtree) relations. It consists of two modules: the {\it text span proposal module} which proposes candidate text spans, each of which represents a subtree in the dependency tree denoted by (root, start, end); and the {\it span linking module}, which constructs links between proposed spans. We use the machine reading comprehension (MRC) framework as the backbone to formalize the span linking module, where one span is used as a query to extract the text span/subtree it should be linked to. The proposed method has the following merits: (1) it addresses the fundamental problem that edges in a dependency tree should be constructed between subtrees; (2) the MRC framework allows the method to retrieve missing spans in the span proposal stage, which leads to higher recall for eligible spans. Extensive experiments on the PTB, CTB and Universal Dependencies (UD) benchmarks demonstrate the effectiveness of the proposed method. The code is available at \url{https://github.com/ShannonAI/mrc-for-dependency-parsing}