SOTAVerified

Grammatical Error Correction

Grammatical Error Correction (GEC) is the task of correcting different kinds of errors in text such as spelling, punctuation, grammatical, and word choice errors.

GEC is typically formulated as a sentence correction task. A GEC system takes a potentially erroneous sentence as input and is expected to transform it to its corrected version. See the example given below:

| Input (Erroneous) | Output (Corrected) | | ------------------------- | ---------------------- | |She see Tom is catched by policeman in park at last night. | She saw Tom caught by a policeman in the park last night.|

Papers

Showing 101150 of 415 papers

TitleStatusHype
Grammatical Error Correction: A Survey of the State of the Art0
From Spelling to Grammar: A New Framework for Chinese Grammatical Error Correction0
Revisiting Grammatical Error Correction Evaluation and BeyondCode1
Towards standardizing Korean Grammatical Error Correction: Datasets and AnnotationCode1
Focus Is What You Need For Chinese Grammatical Error Correction0
FCGEC: Fine-Grained Corpus for Chinese Grammatical Error CorrectionCode1
SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented ParserCode1
Text Editing as Imitation GameCode0
Linguistic Rules-Based Corpus Generation for Native Chinese Grammatical Error CorrectionCode1
Position Offset Label Prediction for Grammatical Error Correction0
Grammatical Error Correction: Are We There Yet?0
IMPARA: Impact-Based Metric for GEC Using Parallel DataCode0
Multi-Perspective Document Revision0
CCTC: A Cross-Sentence Chinese Text Correction Dataset for Native SpeakersCode2
Dynamic Negative Example Construction for Grammatical Error Correction using Contrastive Learning0
Judge a Sentence by Its Content to Generate Grammatical Errors0
Gender Bias and Universal Substitution Adversarial Attacks on Grammatical Error Correction Systems for Automated Assessment0
Chinese grammatical error correction based on knowledge distillationCode1
ErAConD: Error Annotated Conversational Dialog Dataset for Grammatical Error CorrectionCode1
Frustratingly Easy System Combination for Grammatical Error CorrectionCode1
On Assessing and Developing Spoken ’Grammatical Error Correction’ Systems0
Mining Error Templates for Grammatical Error CorrectionCode2
Text Generation with Text-Editing Models0
ProQE: Proficiency-wise Quality Estimation dataset for Grammatical Error Correction0
Automatic Classification of Russian Learner Errors0
Improving Grammatical Error Correction for Multiword Expressions0
Enriching Grammatical Error Correction Resources for Modern Greek0
Developing a Spell and Grammar Checker for Icelandic using an Error Corpus0
Semi-automatically Annotated Learner Corpus for Russian0
MTee: Open Machine Translation Platform for Estonian Government0
EdiT5: Semi-Autoregressive Text-Editing with T5 Warm-Start0
Towards Automated Document Revision: Grammatical Error Correction, Fluency Edits, and BeyondCode1
Sequence-to-Action: Grammatical Error Correction with Action Guided Sequence Generation0
Lossless Acceleration for Seq2seq Generation with Aggressive DecodingCode0
Some Grammatical Errors are Frequent, Others are ImportantCode0
Adjusting the Precision-Recall Trade-Off with Align-and-Predict Decoding for Grammatical Error CorrectionCode0
“Is Whole Word Masking Always Better for Chinese BERT?”: Probing on Chinese Grammatical Error Correction0
A New Evaluation Method: Evaluation Data and Metrics for Chinese Grammar Error Correction0
MuCGEC: a Multi-Reference Multi-Source Evaluation Dataset for Chinese Grammatical Error CorrectionCode2
A Survey on Non-Autoregressive Generation for Neural Machine Translation and BeyondCode1
BLISS: Robust Sequence-to-Sequence Learning via Self-Supervised Input Representation0
Uncertainty Determines the Adequacy of the Mode and the Tractability of Decoding in Sequence-to-Sequence Models0
Ensembling and Knowledge Distilling of Large Sequence Taggers for Grammatical Error CorrectionCode1
Towards Lithuanian grammatical error correctionCode0
Type-Driven Multi-Turn Corrections for Grammatical Error CorrectionCode0
Interpretability for Language Learners Using Example-Based Grammatical Error CorrectionCode1
"Is Whole Word Masking Always Better for Chinese BERT?": Probing on Chinese Grammatical Error Correction0
EdgeFormer: A Parameter-Efficient Transformer for On-Device Seq2seq GenerationCode0
Error Correction in ASR using Sequence-to-Sequence Models0
A Unified Strategy for Multilingual Grammatical Error Correction with Pre-trained Cross-Lingual Language ModelCode0
Show:102550
← PrevPage 3 of 9Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Ensembles of best 7 models + GRECO + GTP-rerankF0.572.8Unverified
2Majority-voting ensemble on best 7 modelsF0.571.8Unverified
3GRECO (voting+ESC)F0.571.12Unverified
4GEC-DI (LM+GED)F0.569.6Unverified
5Unsupervised GEC + cLang8F0.569.6Unverified
6ESCF0.569.51Unverified
7T5F0.568.87Unverified
8MoECEF0.567.79Unverified
9SynGECF0.567.6Unverified
10Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet)F0.566.5Unverified
#ModelMetricClaimedVerifiedStatus
1Majority-voting ensemble on best 7 modelsF0.581.4Unverified
2GRECO (voting+ESC)F0.580.84Unverified
3ESCF0.579.9Unverified
4RedPenNetF0.577.6Unverified
5clang_large_ft2-gectorF0.577.1Unverified
6Unsupervised GEC + cLang8F0.576.5Unverified
7DeBERTa + RoBERTa + XLNetF0.576.05Unverified
8MoECEF0.574.07Unverified
9Sequence tagging + token-level transformations + two-stage fine-tuning (+RoBERTa, XLNet)F0.573.7Unverified
10BEA CombinationF0.573.2Unverified
#ModelMetricClaimedVerifiedStatus
1Llama + 1M BT + goldF0.576.75Unverified
2mT5-based multimodal MoEF0.576.3Unverified
3gT5 xxlF0.575.96Unverified
4TransformerF0.573.71Unverified
5Transformer - synthetic pretrain onlyF0.551.41Unverified
6Multilayer Convolutional Encoder-DecoderF0.543.35Unverified
#ModelMetricClaimedVerifiedStatus
1VERNetGLEU62.1Unverified
2Transformer + Pre-train with Pseudo Data + BERTGLEU62Unverified
3SMT + BiGRUGLEU61.5Unverified
4Copy-augmented Model (4 Ensemble +Denoising Autoencoder)GLEU61Unverified
5TransformerGLEU59.9Unverified
6CNN Seq2SeqGLEU57.47Unverified
#ModelMetricClaimedVerifiedStatus
1Llama + 1M BT + goldF0.574.09Unverified
2mBART-based model with synthetic dataF0.568.17Unverified
3mT5 large + 10M synthF0.568.09Unverified
4RedPenNetF0.567.71Unverified
5ChatGPT (zero-shot)F0.527.4Unverified
#ModelMetricClaimedVerifiedStatus
1GRECO (vote+ESC)F0.585.21Unverified
2SMT + BiGRUF0.572.04Unverified
3CNN Seq2SeqF0.570.14Unverified
#ModelMetricClaimedVerifiedStatus
1CNN Seq2Seq + Quality EstimationF0.556.52Unverified
2TransformerF0.555.8Unverified
3+ BIFI with no criticF0.518.7Unverified
#ModelMetricClaimedVerifiedStatus
1CNN Seq2Seq + Fluency Boost and inferenceGLEU62.37Unverified
2CNN Seq2Seq + Fluency BoostF0.561.34Unverified
3+ BIFI (ours)F0.542.4Unverified
#ModelMetricClaimedVerifiedStatus
1TransformerGLEU59.9Unverified
2CNN Seq2SeqGLEU57.47Unverified
#ModelMetricClaimedVerifiedStatus
1Llama + 1M BT + goldF0.569.97Unverified
#ModelMetricClaimedVerifiedStatus
1STG-Jointexact match34.1Unverified
#ModelMetricClaimedVerifiedStatus
1GEC-DI (LM+GED)F0.548.61Unverified
#ModelMetricClaimedVerifiedStatus
1RedPenNetF0.577.6Unverified