Gated lexical shortcut connections added to the transformer yield 0.9 BLEU average gains on five WMT directions while lowering the lexical content stored in hidden states.
Multi-representation Ensembles and Delayed SGD Updates Improve Syntax-based NMT
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
We explore strategies for incorporating target syntax into Neural Machine Translation. We specifically focus on syntax in ensembles containing multiple sentence representations. We formulate beam search over such ensembles using WFSTs, and describe a delayed SGD update training procedure that is especially effective for long representations like linearized syntax. Our approach gives state-of-the-art performance on a difficult Japanese-English task.
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Widening the Representation Bottleneck in Neural Machine Translation with Lexical Shortcuts
Gated lexical shortcut connections added to the transformer yield 0.9 BLEU average gains on five WMT directions while lowering the lexical content stored in hidden states.