Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890964657> ?p ?o ?g. }
- W2890964657 endingPage "7954" @default.
- W2890964657 startingPage "7944" @default.
- W2890964657 abstract "Neural Machine Translation (NMT) has achieved remarkable progress with the quick evolvement of model structures. In this paper, we propose the concept of layer-wise coordination for NMT, which explicitly coordinates the learning of hidden representations of the encoder and decoder together layer by layer, gradually from low level to high level. Specifically, we design a layer-wise attention and mixed attention mechanism, and further share the parameters of each layer between the encoder and decoder to regularize and coordinate the learning. Experiments show that combined with the state-of-the-art Transformer model, layer-wise coordination achieves improvements on three IWSLT and two WMT translation tasks. More specifically, our method achieves 34.43 and 29.01 BLEU score on WMT16 English-Romanian and WMT14 English-German tasks, outperforming the Transformer baseline." @default.
- W2890964657 created "2018-09-27" @default.
- W2890964657 creator A5013117915 @default.
- W2890964657 creator A5018286848 @default.
- W2890964657 creator A5020025718 @default.
- W2890964657 creator A5021772140 @default.
- W2890964657 creator A5068285627 @default.
- W2890964657 creator A5070990160 @default.
- W2890964657 creator A5079572598 @default.
- W2890964657 date "2018-01-01" @default.
- W2890964657 modified "2023-10-14" @default.
- W2890964657 title "Layer-Wise Coordination between Encoder and Decoder for Neural Machine Translation" @default.
- W2890964657 cites W1902237438 @default.
- W2890964657 cites W1916559533 @default.
- W2890964657 cites W2101105183 @default.
- W2890964657 cites W2128892113 @default.
- W2890964657 cites W2130942839 @default.
- W2890964657 cites W2157331557 @default.
- W2890964657 cites W2170738476 @default.
- W2890964657 cites W2267186426 @default.
- W2890964657 cites W2413794162 @default.
- W2890964657 cites W2487501366 @default.
- W2890964657 cites W2525778437 @default.
- W2890964657 cites W2540404261 @default.
- W2890964657 cites W2597655663 @default.
- W2890964657 cites W2612675303 @default.
- W2890964657 cites W2741040846 @default.
- W2890964657 cites W2767206889 @default.
- W2890964657 cites W2785093437 @default.
- W2890964657 cites W2794365787 @default.
- W2890964657 cites W2803569830 @default.
- W2890964657 cites W2804044248 @default.
- W2890964657 cites W2807964741 @default.
- W2890964657 cites W2889606145 @default.
- W2890964657 cites W2949989304 @default.
- W2890964657 cites W2952339051 @default.
- W2890964657 cites W2962784628 @default.
- W2890964657 cites W2962801832 @default.
- W2890964657 cites W2963248296 @default.
- W2890964657 cites W2963403868 @default.
- W2890964657 cites W2963418779 @default.
- W2890964657 cites W2963551569 @default.
- W2890964657 cites W2964265128 @default.
- W2890964657 cites W2964298349 @default.
- W2890964657 cites W2964308564 @default.
- W2890964657 cites W3204130541 @default.
- W2890964657 hasPublicationYear "2018" @default.
- W2890964657 type Work @default.
- W2890964657 sameAs 2890964657 @default.
- W2890964657 citedByCount "83" @default.
- W2890964657 countsByYear W28909646572018 @default.
- W2890964657 countsByYear W28909646572019 @default.
- W2890964657 countsByYear W28909646572020 @default.
- W2890964657 countsByYear W28909646572021 @default.
- W2890964657 crossrefType "proceedings-article" @default.
- W2890964657 hasAuthorship W2890964657A5013117915 @default.
- W2890964657 hasAuthorship W2890964657A5018286848 @default.
- W2890964657 hasAuthorship W2890964657A5020025718 @default.
- W2890964657 hasAuthorship W2890964657A5021772140 @default.
- W2890964657 hasAuthorship W2890964657A5068285627 @default.
- W2890964657 hasAuthorship W2890964657A5070990160 @default.
- W2890964657 hasAuthorship W2890964657A5079572598 @default.
- W2890964657 hasConcept C104317684 @default.
- W2890964657 hasConcept C105580179 @default.
- W2890964657 hasConcept C111919701 @default.
- W2890964657 hasConcept C11413529 @default.
- W2890964657 hasConcept C118505674 @default.
- W2890964657 hasConcept C119599485 @default.
- W2890964657 hasConcept C127413603 @default.
- W2890964657 hasConcept C149364088 @default.
- W2890964657 hasConcept C154945302 @default.
- W2890964657 hasConcept C165801399 @default.
- W2890964657 hasConcept C178790620 @default.
- W2890964657 hasConcept C185592680 @default.
- W2890964657 hasConcept C203005215 @default.
- W2890964657 hasConcept C2779227376 @default.
- W2890964657 hasConcept C28490314 @default.
- W2890964657 hasConcept C41008148 @default.
- W2890964657 hasConcept C55493867 @default.
- W2890964657 hasConcept C57273362 @default.
- W2890964657 hasConcept C622187 @default.
- W2890964657 hasConcept C66322947 @default.
- W2890964657 hasConceptScore W2890964657C104317684 @default.
- W2890964657 hasConceptScore W2890964657C105580179 @default.
- W2890964657 hasConceptScore W2890964657C111919701 @default.
- W2890964657 hasConceptScore W2890964657C11413529 @default.
- W2890964657 hasConceptScore W2890964657C118505674 @default.
- W2890964657 hasConceptScore W2890964657C119599485 @default.
- W2890964657 hasConceptScore W2890964657C127413603 @default.
- W2890964657 hasConceptScore W2890964657C149364088 @default.
- W2890964657 hasConceptScore W2890964657C154945302 @default.
- W2890964657 hasConceptScore W2890964657C165801399 @default.
- W2890964657 hasConceptScore W2890964657C178790620 @default.
- W2890964657 hasConceptScore W2890964657C185592680 @default.
- W2890964657 hasConceptScore W2890964657C203005215 @default.
- W2890964657 hasConceptScore W2890964657C2779227376 @default.
- W2890964657 hasConceptScore W2890964657C28490314 @default.
- W2890964657 hasConceptScore W2890964657C41008148 @default.