| Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models | ✓ Link | 72.8 | 83.9 | 47.5 | Ensembles of best 7 models + GRECO + GTP-rerank | 2024-04-23 |
| Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models | ✓ Link | 71.8 | 83.7 | 45.7 | Majority-voting ensemble on best 7 models | 2024-04-23 |
| System Combination via Quality Estimation for Grammatical Error Correction | ✓ Link | 71.12 | 79.6 | 49.86 | GRECO (voting+ESC) | 2023-10-23 |
| Improving Seq2Seq Grammatical Error Correction via Decoding Interventions | ✓ Link | 69.6 | 79.2 | 46.8 | GEC-DI (LM+GED) | 2023-10-23 |
| Unsupervised Grammatical Error Correction Rivaling Supervised Methods | ✓ Link | 69.6 | 75.0 | 53.8 | Unsupervised GEC + cLang8 | 2023-12-06 |
| Frustratingly Easy System Combination for Grammatical Error Correction | ✓ Link | 69.51 | 81.48 | 43.78 | ESC | |
| A Simple Recipe for Multilingual Grammatical Error Correction | ✓ Link | 68.87 | | | T5 | 2021-06-07 |
| Efficient and Interpretable Grammatical Error Correction with Mixture of Experts | ✓ Link | 67.79 | 74.29 | 50.21 | MoECE | 2024-10-30 |
| SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser | ✓ Link | 67.6 | 74.7 | 49.0 | SynGEC | 2022-10-22 |
| GECToR -- Grammatical Error Correction: Tag, Not Rewrite | ✓ Link | 66.5 | 78.2 | 41.5 | Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet) | 2020-05-26 |
| LM-Critic: Language Models for Unsupervised Grammatical Error Correction | ✓ Link | 65.8 | | | LM-Critic | 2021-09-14 |
| GECToR -- Grammatical Error Correction: Tag, Not Rewrite | ✓ Link | 65.3 | 77.5 | 40.1 | Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet) | 2020-05-26 |
| Encoder-Decoder Models Can Benefit from Pre-trained Masked Language Models in Grammatical Error Correction | ✓ Link | 65.2 | | | Transformer + Pre-train with Pseudo Data (+BERT) | 2020-05-03 |
| An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction | ✓ Link | 65.0 | | | Transformer + Pre-train with Pseudo Data | 2019-09-02 |
| Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction | ✓ Link | 63.7 | | | VERNet | 2021-05-10 |
| Stronger Baselines for Grammatical Error Correction Using Pretrained Encoder-Decoder Model | ✓ Link | 63.0 | 69.9 | 45.1 | BART | 2020-05-24 |
| Parallel Iterative Edit Models for Local Sequence Transduction | ✓ Link | 61.2 | | | Sequence Labeling with edits using BERT, Faster inference | 2019-10-07 |
| Improving Grammatical Error Correction via Pre-Training a Copy-Augmented Architecture with Unlabeled Data | ✓ Link | 61.15 | 71.57 | 38.65 | Copy-augmented Model (4 Ensemble +Denoising Autoencoder) | 2019-03-01 |
| Parallel Iterative Edit Models for Local Sequence Transduction | ✓ Link | 59.7 | | | Sequence Labeling with edits using BERT, Faster inference (Single Model) | 2019-10-07 |
| Neural Quality Estimation of Grammatical Error Correction | ✓ Link | 56.52 | | | CNN Seq2Seq + Quality Estimation | 2018-10-01 |
| Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation | | 56.25 | | | SMT + BiGRU | 2018-04-16 |
| Approaching Neural Grammatical Error Correction as a Low-Resource Machine Translation Task | ✓ Link | 55.8 | | | Transformer | 2018-04-16 |
| A Multilayer Convolutional Encoder-Decoder Neural Network for Grammatical Error Correction | ✓ Link | 54.79 | | | CNN Seq2Seq | 2018-01-26 |