Approaching Neural Grammatical Error Correction as a Low-Resource Machine Translation Task

Benchmark Model Rank Results
grammatical-error-correction-on-conll-2014-shared-taskTransformer#21F0.5: 55.8
grammatical-error-correction-on-jflegTransformer#4GLEU: 59.9
grammatical-error-correction-on-restrictedTransformer#3F0.5: 55.8