Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Six Challenges for Neural Machine Translation

About

We explore six challenges for neural machine translation: domain mismatch, amount of training data, rare words, long sentences, word alignment, and beam search. We show both deficiencies and improvements over the quality of phrase-based statistical machine translation.

Philipp Koehn, Rebecca Knowles• 2017

Related benchmarks

TaskDatasetResultRank
Machine TranslationIWSLT German-to-English '14 (test)
BLEU Score32.77
110
Visual Question AnsweringVQA 2.0 (test)
Accuracy58.93
24
Neural Machine TranslationIWSLT and WMT (test)
PPL7.17
7
Showing 3 of 3 rows

Other info

Follow for update