A Nested Attention Neural Hybrid Model for Grammatical Error Correction

TL;DR

This paper introduces a nested attention neural hybrid model for grammatical error correction that effectively addresses both global and local errors, outperforming previous models on standard benchmarks.

Contribution

The paper presents a novel hybrid neural model with nested attention layers that improves grammatical error correction by integrating word and character-level information.

Findings

Significantly outperforms previous neural models on CoNLL-14 benchmark

Nested attention mechanism effectively corrects local orthographic errors

Model demonstrates improved correction of both global and local errors

Abstract

Grammatical error correction (GEC) systems strive to correct both global errors in word order and usage, and local errors in spelling and inflection. Further developing upon recent work on neural machine translation, we propose a new hybrid neural model with nested attention layers for GEC. Experiments show that the new model can effectively correct errors of both types by incorporating word and character-level information,and that the model significantly outperforms previous neural models for GEC as measured on the standard CoNLL-14 benchmark dataset. Further analysis also shows that the superiority of the proposed model can be largely attributed to the use of the nested attention mechanism, which has proven particularly effective in correcting local errors that involve small edits in orthography.