For similar language pairs, phrase-based statistical and neural translation produce comparable BLEU scores, with the better system depending on the specific language pair.
OpenKiwi: An Open Source Framework for Quality Estimation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We introduce OpenKiwi, a PyTorch-based open source framework for translation quality estimation. OpenKiwi supports training and testing of word-level and sentence-level quality estimation systems, implementing the winning systems of the WMT 2015-18 quality estimation campaigns. We benchmark OpenKiwi on two datasets from WMT 2018 (English-German SMT and NMT), yielding state-of-the-art performance on the word-level tasks and near state-of-the-art in the sentence-level tasks.
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The TALP-UPC System for the WMT Similar Language Task: Statistical vs Neural Machine Translation
For similar language pairs, phrase-based statistical and neural translation produce comparable BLEU scores, with the better system depending on the specific language pair.