Hybrid Oracle: Making Use of Ambiguity in Transition-based Chinese Dependency Parsing

TL;DR

This paper introduces a hybrid oracle for transition-based Chinese dependency parsing that leverages ambiguity to improve training, resulting in better parser performance and generalization.

Contribution

It presents a novel hybrid oracle that utilizes multiple correct transition sequences during training, enhancing neural parser accuracy and robustness.

Findings

Improved parsing accuracy over traditional oracle methods.

Enhanced generalization ability of the parser.

Analysis confirms linguistic validity of the approach.

Abstract

In the training of transition-based dependency parsers, an oracle is used to predict a transition sequence for a sentence and its gold tree. However, the transition system may exhibit ambiguity, that is, there can be multiple correct transition sequences that form the gold tree. We propose to make use of the property in the training of neural dependency parsers, and present the Hybrid Oracle. The new oracle gives all the correct transitions for a parsing state, which are used in the cross entropy loss function to provide better supervisory signal. It is also used to generate different transition sequences for a sentence to better explore the training data and improve the generalization ability of the parser. Evaluations show that the parsers trained using the hybrid oracle outperform the parsers using the traditional oracle in Chinese dependency parsing. We provide analysis from a linguistic view. The code is available at https://github.com/lancopku/nndep .