SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser

TL;DR

SynGEC introduces a syntax-enhanced GEC model that uses a tailored parser trained on parallel data to incorporate syntactic information, significantly improving grammatical error correction performance.

Contribution

The paper develops a GEC-oriented parser trained on parallel data and integrates dependency syntax into GEC models, addressing parser unreliability on ungrammatical sentences.

Findings

Outperforms strong baselines on English and Chinese GEC datasets

Effectively incorporates syntactic information into GEC models

Achieves competitive results in grammatical error correction

Abstract

This work proposes a syntax-enhanced grammatical error correction (GEC) approach named SynGEC that effectively incorporates dependency syntactic information into the encoder part of GEC models. The key challenge for this idea is that off-the-shelf parsers are unreliable when processing ungrammatical sentences. To confront this challenge, we propose to build a tailored GEC-oriented parser (GOPar) using parallel GEC training data as a pivot. First, we design an extended syntax representation scheme that allows us to represent both grammatical errors and syntax in a unified tree structure. Then, we obtain parse trees of the source incorrect sentences by projecting trees of the target correct sentences. Finally, we train GOPar with such projected trees. For GEC, we employ the graph convolution network to encode source-side syntactic information produced by GOPar, and fuse them with the outputs of the Transformer encoder. Experiments on mainstream English and Chinese GEC datasets show that our proposed SynGEC approach consistently and substantially outperforms strong baselines and achieves competitive performance. Our code and data are all publicly available at https://github.com/HillZhang1999/SynGEC.