Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models

TL;DR

This paper thoroughly examines contemporary grammatical error correction methods, emphasizing large language models, and achieves new state-of-the-art results through innovative single-model, ensemble, and ranking techniques.

Contribution

It introduces a comprehensive analysis of GEC approaches, highlighting the integration of large language models and establishing new performance benchmarks.

Findings

Achieved state-of-the-art F0.5 scores of 72.8 on CoNLL-2014-test and 81.4 on BEA-test.

Compared efficiency of ensembling and ranking methods in GEC.

Demonstrated effectiveness of large language models as single systems and ensemble components.

Abstract

In this paper, we carry out experimental research on Grammatical Error Correction, delving into the nuances of single-model systems, comparing the efficiency of ensembling and ranking methods, and exploring the application of large language models to GEC as single-model systems, as parts of ensembles, and as ranking methods. We set new state-of-the-art performance with F_0.5 scores of 72.8 on CoNLL-2014-test and 81.4 on BEA-test, respectively. To support further advancements in GEC and ensure the reproducibility of our research, we make our code, trained models, and systems' outputs publicly available.