SMT + BiGRU

Reported on 3 benchmarks across 1 task · 1 paper · 3 SOTA

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

Natural Language Processing3 results

Grammatical Error CorrectiononCoNLL-2014 Shared Task
F0.5· 2018-04-16
56.25
best: 72.8 (Ensembles of best 7 models + GRECO + GTP-rerank)
SOTA
Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation arXiv:1804.05945
Grammatical Error CorrectiononJFLEG
GLEU· 2018-04-16
61.5
best: 62.1 (VERNet)
SOTA
Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation arXiv:1804.05945
Grammatical Error CorrectiononCoNLL-2014 Shared Task (10 annotations)
F0.5· 2018-04-16
72.04
best: 85.21 (GRECO (vote+ESC))
SOTA
Near Human-Level Performance in Grammatical Error Correction with Hybrid Machine Translation arXiv:1804.05945