Abstract
In this paper, we show how a multi-class grammatical error detection (GED) system can be used to improve grammatical error correction (GEC) for English. Specifically, we first develop a new state-of-the-art binary detection system based on pre-trained ELECTRA, and then extend it to multi-class detection using different error type tagsets derived from the ERRANT framework. Output from this detection system is used as auxiliary input to fine-tune a novel encoder-decoder GEC model, and we subsequently re-rank the N-best GEC output to find the hypothesis that most agrees with the GED output. Results show that fine-tuning the GEC system using 4-class GED produces the best model, but re-ranking using 55-class GED leads to the best performance overall. This suggests that different multi-class GED systems benefit GEC in different ways. Ultimately, our system outperforms all other previous work that combines GED and GEC, and achieves a new single-model NMT-based state of the art on the BEA-test benchmark.
Highlights
We first experiment with pretrained language models, and show that finethen extend it to multi-class detection using different error type tagsets derived from the ERRANT framework
Output from this detection system is used as auxiliary input to finetune a novel encoder-decoder GEC model, and we subsequently re-rank the N -best GEC output to find the hypothesis that most agrees with the GED output
Given that binary detection is limited in terms of the specific error type information it can provide to downstream tasks we extend our GED system to 4-class, 25-class and 55-class error detection using different error the GEC system using 4-class GED produces the best model, but re-ranking using 55-class GED leads to the best performance overall
Summary
Approaches to GED focused on specific error types, and in particular article and preposition errors, which are among the most frequent in nonnative English learner writing (Han et al, 2004; Tetreault and Chodorow, 2008). Kaneko et al (2020) fine-tuned BERT for binary GED and incorporated their model into an encoder-decoder GEC framework. Our approach of using additional GED input during GEC training differs from theirs in that we only use a small set of training examples with GED information for fine-tuning. Binary GED has been used in postprocessing to re-rank GEC system output (Yannakoudakis et al, 2017; Yuan et al, 2019; Wang et al, 2020) or filter out unnecessary corrections (Kiyono et al, 2019). Sequence- bridge English Write & Improve + LOCNESS tagging approaches have been proposed for GEC, (W&I) corpus released in the Building Educational where systems learn to predict a sequence of edit Applications (BEA) shared task on GEC Its objective to discriminate between plausible and non-plausible word tokens makes it more closely-related to GED
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.