Enhancing Grammatical Error Correction Systems with Explanations

Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics Volume 1: Long Papers, pages 7489–7501 July 9-14, 2023 ©2023 Association for Computational Linguistics Enhancing Grammatical Error Correction Systems with Explanations Yuejiao Fei♠♣∗, Leyang Cui♡†, Sen Yang♢, Wai Lam♢, Zhenzhong Lan♣ , Shuming Shi♡ ♠ Zhejiang University ♡ Tencent AI lab ♢ The Chinese University of Hong Kong ♣ School of Engineering, Westlake University {feiyuejiao,lanzhenzhong}@westlake.edu.cn {leyangcui,shumingshi}@tencent.com {syang, wlam}@se.cuhk.edu.hk Abstract Grammatical error correction systems improve written communication by detecting and cor- recting 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 sys- tems with explanations, we introduce EXPECT, a large dataset annotated with evidence words and grammatical error types. We propose sev- eral 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. 1 Introduction Grammatical Error Correction (GEC) systems aim to detect and correct grammatical errors in a given sentence and thus provide useful information for second-language learners. There are two lines of work for building GEC systems. Sequence- to-sequence methods (Rothe et al., 2021; Flachs et al., 2021; Zhang et al., 2022) take an erroneous sentence as input and generate an error-free sen- tence autoregressively. Sequence labeling methods (Omelianchuk et al., 2020; Tarnavskyi et al., 2022a) transform the target into a sequence of text-editing actions and use the sequence labeling scheme to predict those actions. With advances in large pre-trained models (De- vlin et al., 2018; Lewis et al., 2019) and avail- ability of high-quality GEC corpora (Ng et al., 2014; Bryant et al., 2019), academic GEC sys- tems (Omelianchuk et al., 2020; Rothe et al., 2021) have achieved promising results on benchmarks and serve as strong backbones for modern writing *Work was done during the internship at Tencent AI lab. †Corresponding authors. As a result, I enjoy study accounting. As a result, I enjoy studying accounting. GEC systems correct grammatical errors without giving specific reasons Explainable-GEC system corrects grammatical errors with explanation Input Change ‘study’ to ‘studying’, because after ‘enjoy’ it should follow a “gerund”. As a result, I enjoy studying accounting. “Gerund” Error Figure 1: Comparison between explainable GEC and conventional GEC systems. assistance applications (e.g., Google Docs1, Gram- marly2, and Effidit (Shi et al., 2023)3). Although these academic methods provide high-quality writ- ing suggestions, they rarely offer explanations with specific clues for corrections. Providing a grammar- aware explanation and evidence words to support the correction is important in second-language edu- cation scenarios (Ellis et al., 2006), where language learners need to “know why” than merely “know how”. As a commercial system, Grammarly does provide evidence words, but in very limited cases, and the technical details are still a black box for the research community. Though some existing work has attempted to enhance the explainability …

Content truncated to 500 words