Enhancing Grammatical Error Correction Systems with Explanations

TL;DR

This paper introduces EXPECT, a large annotated dataset and an explainable GEC system that provides evidence words and error types to help language learners understand corrections.

Contribution

The paper presents EXPECT, a novel dataset with explanations for grammatical errors, and demonstrates how explainable GEC systems can aid second-language learners.

Findings

Human evaluation shows explanations help learners understand corrections.

The dataset enables better analysis of error causes and types.

Baseline models demonstrate the feasibility of explainable GEC.

Abstract

Grammatical error correction systems improve written communication by detecting and correcting language mistakes. To help language learners better understand why the GEC system makes a certain correction, the causes of errors (evidence words) and the corresponding error types are two key factors. To enhance GEC systems with explanations, we introduce EXPECT, a large dataset annotated with evidence words and grammatical error types. We propose several baselines and analysis to understand this task. Furthermore, human evaluation verifies our explainable GEC system's explanations can assist second-language learners in determining whether to accept a correction suggestion and in understanding the associated grammar rule.