Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction

About

Grammatical Error Correction (GEC) aims to correct writing errors and help language learners improve their writing skills. However, existing GEC models tend to produce spurious corrections or fail to detect lots of errors. The quality estimation model is necessary to ensure learners get accurate GEC results and avoid misleading from poorly corrected sentences. Well-trained GEC models can generate several high-quality hypotheses through decoding, such as beam search, which provide valuable GEC evidence and can be used to evaluate GEC quality. However, existing models neglect the possible GEC evidence from different hypotheses. This paper presents the Neural Verification Network (VERNet) for GEC quality estimation with multiple hypotheses. VERNet establishes interactions among hypotheses with a reasoning graph and conducts two kinds of attention mechanisms to propagate GEC evidence to verify the quality of generated hypotheses. Our experiments on four GEC datasets show that VERNet achieves state-of-the-art grammatical error detection performance, achieves the best quality estimation results, and significantly improves GEC performance by reranking hypotheses. All data and source codes are available at https://github.com/thunlp/VERNet.

Zhenghao Liu, Xiaoyuan Yi, Maosong Sun, Liner Yang, Tat-Seng Chua• 2021

Related benchmarks

TaskDatasetResultRank
Grammatical Error CorrectionCoNLL 2014 (test)
F0.5 Score62.85
207
Grammatical Error CorrectionBEA shared task 2019 (test)
F0.5 Score69.67
139
Grammatical Error CorrectionJFLEG
GLEU62.07
47
Grammatical Error CorrectionCoNLL 2014
F0.560.35
39
Grammatical Error CorrectionCoNLL M2 14
Precision (P)71.87
27
Grammatical Error CorrectionBEA 2019 (dev)
F0.5 Score54.8
19
Sentence Level Quality EstimationCoNLL 2014 M2 (test)
Precision69.97
12
Sentence Level Quality EstimationFCE (test)
Precision58.77
12
Sentence Level Quality EstimationBEA19 (test)
Precision69.09
12
Sentence Level Quality EstimationJFLEG (test)
GLEU61.61
12
Showing 10 of 17 rows

Other info

Code

Follow for update