SOTA Grammatical Error Correction benchmarks and papers with code

Benchmarks

Dataset Name	SOTA Method	Metric
CoNLL 2014 (test)	AGGR-RANK*	F0.5 Score72.2	233	1mo ago
BEA shared task 2019 (test)	GRECO voting+ESC	F0.5 Score80.84	139	4mo ago
JFLEG (test)	Lichtarge et al. (2020)	GLEU64.9	60	2mo ago
NLPCC-18 Chinese GEC (test)	Traditional Voting	Precision67.33	49	4mo ago
BEA 2019 (test)	T5-XXL	F0.575.9	47	1mo ago
JFLEG	Transformer + Pre-training + Iterative refinement + Ensemble Decoding (8)	GLEU63.3	47	4mo ago
UNLP 2023 (test)		Precision78.52	40	1mo ago
GEC	TAG	Hard F1 Score36.5	40	4mo ago
CoNLL 2014		F0.565.2	39	4mo ago
MuCGEC (test)	Traditional Voting	Precision76.13	34	4mo ago
CoNLL M2 14	BERT-fuse GED + R2L	Precision (P)72.6	27	4mo ago
RobustGEC TEM-8* (test)	Qwen3-8B + COCOGEC	Precision55.7	20	1mo ago
BEA 2019 (dev)	GRECO voting+ESC	F0.5 Score63.4	19	4mo ago
IceEC 1.0 (test)	ByT5	GLEU87.4	17	4mo ago
FCGEC (test)	Alirector	Precision69.44	17	4mo ago
NLPCC word-level (test)	Traditional Voting	Precision64.51	17	4mo ago
IceEC dyslexic	ByT5 for GEC	F0.5 Score57	16	4mo ago
IceEC (test)	ByT5 for GEC	F0.5 Score54	16	4mo ago
CoNLL-10	Base + FB learning	F0.5 Score76.88	16	4mo ago
CWEB-G (test)	GECToR (deberta-v3-large)	Precision56.1	15	2mo ago
BEA (dev)	AGGR-RANK	Precision (%)71.7	14	4mo ago
RULEC-GEC Russian (test)	gT5 xxl	F0.5 Score51.62	14	4mo ago
BEA 19	Basic GEC	Precision74.7	12	4mo ago
FCE M2 (test)	BERT-fuse GED	Precision65	10	4mo ago
RONACC	gemma-3-12b-it	GLEU87.1	9	2mo ago

Showing 25 of 61 rows