Personalizing Grammatical Error Correction: Adaptation to Proficiency Level and L1

TL;DR

This paper explores how to personalize neural grammatical error correction systems based on a user's proficiency level and first language, demonstrating significant performance improvements through adaptation.

Contribution

It introduces the first comprehensive study on adapting GEC systems to proficiency and L1 using limited data, covering multiple languages and proficiency levels.

Findings

Adapting to both proficiency and L1 yields the largest performance gains.

Tailoring GEC systems to user characteristics improves correction accuracy.

Study covers five proficiency levels and twelve languages.

Abstract

Grammar error correction (GEC) systems have become ubiquitous in a variety of software applications, and have started to approach human-level performance for some datasets. However, very little is known about how to efficiently personalize these systems to the user's characteristics, such as their proficiency level and first language, or to emerging domains of text. We present the first results on adapting a general-purpose neural GEC system to both the proficiency level and the first language of a writer, using only a few thousand annotated sentences. Our study is the broadest of its kind, covering five proficiency levels and twelve different languages, and comparing three different adaptation scenarios: adapting to the proficiency level only, to the first language only, or to both aspects simultaneously. We show that tailoring to both scenarios achieves the largest performance improvement (3.6 F0.5) relative to a strong baseline.