GECToR -- Grammatical Error Correction: Tag, Not Rewrite

WS 2020 Kostiantyn OmelianchukVitaliy AtrasevychArtem ChernodubOleksandr Skurzhanskyi

In this paper, we present a simple and efficient GEC sequence tagger using a Transformer encoder. Our system is pre-trained on synthetic data and then fine-tuned in two stages: first on errorful corpora, and second on a combination of errorful and error-free parallel corpora... (read more)

PDF Abstract WS 2020 PDF WS 2020 Abstract
TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Grammatical Error Correction BEA-2019 (test) Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet) F0.5 72.4 # 2
Grammatical Error Correction BEA-2019 (test) Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet) F0.5 73.6 # 1
Grammatical Error Correction CoNLL-2014 Shared Task Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet) F0.5 65.3 # 2
Precision 77.5 # 2
Recall 40.1 # 2
Grammatical Error Correction CoNLL-2014 Shared Task Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet) F0.5 66.5 # 1
Precision 78.2 # 1
Recall 41.5 # 1

Methods used in the Paper