Grammatical Error Correction on BEA-2019 (test)

Metric: F0.5 (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	F0.5▼	Extra Data	Paper	Date↕	Code
1	Majority-voting ensemble on best 7 models	81.4	No	Pillars of Grammatical Error Correction: Compreh...	2024-04-23	Code
2	GRECO (voting+ESC)	80.84	No	System Combination via Quality Estimation for Gr...	2023-10-23	Code
3	ESC	79.9	No	-	-	Code
4	RedPenNet	77.6	Yes	RedPenNet for Grammatical Error Correction: Outp...	2023-09-19	Code
5	clang_large_ft2-gector	77.1	No	-	-	Code
6	Unsupervised GEC + cLang8	76.5	No	-	-	Code
7	DeBERTa + RoBERTa + XLNet	76.05	No	Ensembling and Knowledge Distilling of Large Seq...	2022-03-24	Code
8	MoECE	74.07	No	Efficient and Interpretable Grammatical Error Co...	2024-10-30	Code
9	Sequence tagging + token-level transformations + two-stage fine-tuning (+RoBERTa, XLNet)	73.7	No	GECToR -- Grammatical Error Correction: Tag, Not...	2020-05-26	Code
10	BEA Combination	73.2	No	Learning to combine Grammatical Error Corrections	2019-06-10	Code
11	GEC-DI (LM+GED)	73.1	No	Improving Seq2Seq Grammatical Error Correction v...	2023-10-23	Code
12	LM-Critic	72.9	No	LM-Critic: Language Models for Unsupervised Gram...	2021-09-14	Code
13	Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet)	72.4	No	GECToR -- Grammatical Error Correction: Tag, Not...	2020-05-26	Code
14	Transformer + Pre-train with Pseudo Data	70.2	No	An Empirical Study of Incorporating Pseudo Data ...	2019-09-02	Code
15	Transformer + Pre-train with Pseudo Data (+BERT)	69.8	No	Encoder-Decoder Models Can Benefit from Pre-trai...	2020-05-03	Code
16	Transformer	69.5	No	-	-	Code
17	Transformer	69	No	A Neural Grammatical Error Correction System Bui...	2019-07-02	Code
18	VERNet	68.9	No	Neural Quality Estimation with Multiple Hypothes...	2021-05-10	Code
19	Ensemble of models	66.78	No	-	-	-

#1Majority-voting ensemble on best 7 modelsSOTA
81.4
F0.5· 2024-04-23
Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models Code
#2GRECO (voting+ESC)SOTA
80.84
F0.5· 2023-10-23
System Combination via Quality Estimation for Grammatical Error Correction Code
#3ESC
79.9
F0.5
No paperCode
#4RedPenNetSOTA
77.6
F0.5· Extra Data· 2023-09-19
RedPenNet for Grammatical Error Correction: Outputs to Tokens, Attentions to Spans Code
#5clang_large_ft2-gector
77.1
F0.5
No paperCode
#6Unsupervised GEC + cLang8
76.5
F0.5
No paperCode
#7DeBERTa + RoBERTa + XLNetSOTA
76.05
F0.5· 2022-03-24
Ensembling and Knowledge Distilling of Large Sequence Taggers for Grammatical Error Correction Code
#8MoECE
74.07
F0.5· 2024-10-30
Efficient and Interpretable Grammatical Error Correction with Mixture of Experts Code
#9Sequence tagging + token-level transformations + two-stage fine-tuning (+RoBERTa, XLNet)SOTA
73.7
F0.5· 2020-05-26
GECToR -- Grammatical Error Correction: Tag, Not Rewrite Code
#10BEA CombinationSOTA
73.2
F0.5· 2019-06-10
Learning to combine Grammatical Error Corrections Code
#11GEC-DI (LM+GED)
73.1
F0.5· 2023-10-23
Improving Seq2Seq Grammatical Error Correction via Decoding Interventions Code
#12LM-Critic
72.9
F0.5· 2021-09-14
LM-Critic: Language Models for Unsupervised Grammatical Error Correction Code
#13Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet)
72.4
F0.5· 2020-05-26
GECToR -- Grammatical Error Correction: Tag, Not Rewrite Code
#14Transformer + Pre-train with Pseudo Data
70.2
F0.5· 2019-09-02
An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction Code
#15Transformer + Pre-train with Pseudo Data (+BERT)
69.8
F0.5· 2020-05-03
Encoder-Decoder Models Can Benefit from Pre-trained Masked Language Models in Grammatical Error Correction Code
#16Transformer
69.5
F0.5
No paperCode
#17Transformer
69
F0.5· 2019-07-02
A Neural Grammatical Error Correction System Built On Better Pre-training and Sequential Transfer Learning Code
#18VERNet
68.9
F0.5· 2021-05-10
Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction Code
#19Ensemble of models
66.78
F0.5
No paper