Error detection on FCE (test)
74.1F0.5 ScoreGOPar
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| GOParTrain-set=FCE-train2022.10 | 74.1 | — | — | — | |
| Yuan et al.Train-set=FCE-train2022.10 | 72.93 | — | — | — | |
| GOParTrain-set=CLang82022.10 | 71.53 | — | — | — | |
| Rei and Yannakoudakis2017.07 | 64.3 | — | — | — | |
| bi-directional LSTM with auxiliary objectivestraining_data=17.8M tokens (CLC, NUCLE, Lang-8), auxiliary_objective=POS tag prediction2017.07 | 64.1 | — | 78.4 | 37 | |
| Kaneko and KomachiTrain-set=FCE-train2022.10 | 61.65 | — | — | — | |
| Bell et al.Train-set=FCE-train2022.10 | 57.28 | — | — | — | |
| BiLSTM-JOINTOptimization=all auxiliary objectives2018.11 | 52.07 | — | 65.53 | 28.61 | |
| Ann+PAT+MTTraining Data=Manual annotation + Pattern-based AEG + SMT-based AEG2017.07 | 49.11 | — | 60.67 | 28.08 | |
| Rei (2017)Description=best system from Rei (2017)2018.11 | 48.48 | — | 58.88 | 28.92 | |
| Ann+MTTraining Data=Manual annotation + SMT-based AEG2017.07 | 48.37 | — | 58.38 | 28.84 | |
| Ann+PATTraining Data=Manual annotation + Pattern-based AEG2017.07 | 47.81 | — | 62.47 | 24.7 | |
| Ann+FY14Training Data=Manual annotation + Felice and Yuan (2014) artificial data2017.07 | 46.54 | — | 58.77 | 25.55 | |
| BiLSTM-ATTNSupervision=only for sequence labeling2018.11 | 45.07 | — | 60.73 | 22.33 | |
| AnnotationTraining Data=Manual annotation only2017.07 | 44.84 | — | 53.91 | 26.88 | |
| Rei & Yannakoudakis (2016)Training Data=Provided in original publication2017.07 | 41.1 | — | 46.1 | 28.5 |