paper-with-me

Grammatical Error Correction 벤치마크

Grammatical Error Correction on CoNLL-2014 Shared Task

23개 결과 · ⬇ CSV · JSON

F0.5

54.79 59.29 63.8 68.3 72.8 2018-01 2026-09 CNN Seq2Seq — 54.79 (2018-01-26) SMT + BiGRU — 56.25 (2018-04-16) Transformer — 55.8 (2018-04-16) CNN Seq2Seq + Quality Estimation — 56.52 (2018-10-01) Copy-augmented Model (4 Ensemble +Denoising Autoencoder) — 61.15 (2019-03-01) Transformer + Pre-train with Pseudo Data — 65.0 (2019-09-02) Sequence Labeling with edits using BERT, Faster inference — 61.2 (2019-10-07) Sequence Labeling with edits using BERT, Faster inference (Single Model) — 59.7 (2019-10-07) Transformer + Pre-train with Pseudo Data (+BERT) — 65.2 (2020-05-03) BART — 63.0 (2020-05-24) Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet) — 66.5 (2020-05-26) Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet) — 65.3 (2020-05-26) VERNet — 63.7 (2021-05-10) T5 — 68.87 (2021-06-07) LM-Critic — 65.8 (2021-09-14) SynGEC — 67.6 (2022-10-22) GRECO (voting+ESC) — 71.12 (2023-10-23) GEC-DI (LM+GED) — 69.6 (2023-10-23) Unsupervised GEC + cLang8 — 69.6 (2023-12-06) Ensembles of best 7 models + GRECO + GTP-rerank — 72.8 (2024-04-23) Majority-voting ensemble on best 7 models — 71.8 (2024-04-23) MoECE — 67.79 (2024-10-30) CNN Seq2Seq — 54.79 (2018-01-26) SMT + BiGRU — 56.25 (2018-04-16) CNN Seq2Seq + Quality Estimation — 56.52 (2018-10-01) Copy-augmented Model (4 Ensemble +Denoising Autoencoder) — 61.15 (2019-03-01) Transformer + Pre-train with Pseudo Data — 65.0 (2019-09-02) Transformer + Pre-train with Pseudo Data (+BERT) — 65.2 (2020-05-03) Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet) — 66.5 (2020-05-26) T5 — 68.87 (2021-06-07) GRECO (voting+ESC) — 71.12 (2023-10-23) Ensembles of best 7 models + GRECO + GTP-rerank — 72.8 (2024-04-23)
RankModel F0.5PrecisionRecall PaperCodeYear
1 Ensembles of best 7 models + GRECO + GTP-rerank 72.883.947.5 Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models grammarly/pillars-of-gec 2024
2 Majority-voting ensemble on best 7 models 71.883.745.7 Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models grammarly/pillars-of-gec 2024
3 GRECO (voting+ESC) 71.1279.649.86 System Combination via Quality Estimation for Grammatical Error Correction nusnlp/greco 2023
4 GEC-DI (LM+GED) 69.679.246.8 Improving Seq2Seq Grammatical Error Correction via Decoding Interventions Jacob-Zhou/gecdi 2023
4 Unsupervised GEC + cLang8 69.675.053.8 Unsupervised Grammatical Error Correction Rivaling Supervised Methods nusnlp/ugec 2023
6 ESC 69.5181.4843.78 Frustratingly Easy System Combination for Grammatical Error Correction nusnlp/esc
7 T5 68.87–– A Simple Recipe for Multilingual Grammatical Error Correction google-research-datasets/clang8 · gotutiyan/gec-t5 2021
8 MoECE 67.7974.2950.21 Efficient and Interpretable Grammatical Error Correction with Mixture of Experts nusnlp/moece 2024
9 SynGEC 67.674.749.0 SynGEC: Syntax-Enhanced Grammatical Error Correction with a Tailored GEC-Oriented Parser hillzhang1999/syngec 2022
10 Sequence tagging + token-level transformations + two-stage fine-tuning (+BERT, RoBERTa, XLNet) 66.578.241.5 GECToR -- Grammatical Error Correction: Tag, Not Rewrite grammarly/gector · psawa/gecko-app · gotutiyan/gector 2020
11 LM-Critic 65.8–– LM-Critic: Language Models for Unsupervised Grammatical Error Correction grammarly/gector · michiyasunaga/LM-Critic · elaynelemos/gector-api 2021
12 Sequence tagging + token-level transformations + two-stage fine-tuning (+XLNet) 65.377.540.1 GECToR -- Grammatical Error Correction: Tag, Not Rewrite grammarly/gector · psawa/gecko-app · gotutiyan/gector 2020
13 Transformer + Pre-train with Pseudo Data (+BERT) 65.2–– Encoder-Decoder Models Can Benefit from Pre-trained Masked Language Models in Grammatical Error Correction kanekomasahiro/bert-gec 2020
14 Transformer + Pre-train with Pseudo Data 65.0–– An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction butsugiri/gec-pseudodata 2019
15 VERNet 63.7–– Neural Quality Estimation with Multiple Hypotheses for Grammatical Error Correction thunlp/VERNet 2021
16 BART 63.069.945.1 Stronger Baselines for Grammatical Error Correction Using Pretrained Encoder-Decoder Model Katsumata420/generic-pretrained-GEC · gotutiyan/GEC-BART 2020
17 Sequence Labeling with edits using BERT, Faster inference 61.2–– Parallel Iterative Edit Models for Local Sequence Transduction awasthiabhijeet/PIE 2019
18 Copy-augmented Model (4 Ensemble +Denoising Autoencoder) 61.1571.5738.65 Improving Grammatical Error Correction via Pre-Training a Copy-Augmented Architecture with Unlabeled Data zhawe01/fairseq-gec · yuantiku/fairseq-gec · youichiro/transformer-copy · +3 2019
19 Sequence Labeling with edits using BERT, Faster inference (Single Model) 59.7–– Parallel Iterative Edit Models for Local Sequence Transduction awasthiabhijeet/PIE 2019
20 CNN Seq2Seq + Quality Estimation 56.52–– Neural Quality Estimation of Grammatical Error Correction nusnlp/neuqe 2018
1–20 / 23 다음 → 페이지당 10 20 50 100