CNN Seq2Seq + Quality Estimation

Reported on 2 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing2 results

Grammatical Error CorrectiononRestricted
F0.5
56.52
Grammatical Error CorrectiononCoNLL-2014 Shared Task
F0.5
56.52
best: 72.8 (Ensembles of best 7 models + GRECO + GTP-rerank)