Transformer + Pre-train with Pseudo Data

Reported on 2 benchmarks across 1 task · 1 paper · 1 SOTA

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 CorrectiononCoNLL-2014 Shared Task
F0.5· 2019-09-02
65
best: 72.8 (Ensembles of best 7 models + GRECO + GTP-rerank)
SOTA
An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction arXiv:1909.00502
Grammatical Error CorrectiononBEA-2019 (test)
F0.5· 2019-09-02
70.2
best: 81.4 (Majority-voting ensemble on best 7 models)
An Empirical Study of Incorporating Pseudo Data into Grammatical Error Correction arXiv:1909.00502