Fine-tuning a neural translation model on a small subset of sentences selected by test-set similarity can outperform fine-tuning on the full in-domain corpus.
Findings of the 2017 conference on ma- chine translation (wmt17)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Transductive Data-Selection Algorithms for Fine-Tuning Neural Machine Translation
Fine-tuning a neural translation model on a small subset of sentences selected by test-set similarity can outperform fine-tuning on the full in-domain corpus.