Jointly Learning to Label Sentences and Tokens

Benchmark Model Rank Results
grammatical-error-detection-on-conll-2014-a1BiLSTM-JOINT (trained on FCE)#2F0.5: 22.14
grammatical-error-detection-on-conll-2014-a2BiLSTM-JOINT (trained on FCE)#2F0.5: 29.65
grammatical-error-detection-on-fceBiLSTM-JOINT#2F0.5: 52.07