FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950479933> ?p ?o ?g. }

• W2950479933 abstract "Existing neural machine translation systems do not explicitly model what has been translated and what has not during the decoding phase. To address this problem, we propose a novel mechanism that separates the source information into two parts: translated Past contents and untranslated Future contents, which are modeled by two additional recurrent layers. The Past and Future contents are fed to both the attention model and the decoder states, which offers NMT systems the knowledge of translated and untranslated contents. Experimental results show that the proposed approach significantly improves translation performance in Chinese-English, German-English and English-German translation tasks. Specifically, the proposed model outperforms the conventional coverage model in both of the translation quality and the alignment error rate." @default.
• W2950479933 created "2019-06-27" @default.
• W2950479933 creator A5016804474 @default.
• W2950479933 creator A5022574607 @default.
• W2950479933 creator A5024821632 @default.
• W2950479933 creator A5025469065 @default.
• W2950479933 creator A5033423455 @default.
• W2950479933 creator A5045799619 @default.
• W2950479933 creator A5067322430 @default.
• W2950479933 date "2017-11-26" @default.
• W2950479933 modified "2023-09-27" @default.
• W2950479933 title "Modeling Past and Future for Neural Machine Translation" @default.
• W2950479933 cites W1411230545 @default.
• W2950479933 cites W1753482797 @default.
• W2950479933 cites W2101105183 @default.
• W2950479933 cites W2131774270 @default.
• W2950479933 cites W2194690080 @default.
• W2950479933 cites W2409591106 @default.
• W2950479933 cites W2410539690 @default.
• W2950479933 cites W2415583245 @default.
• W2950479933 cites W2470713034 @default.
• W2950479933 cites W2487501366 @default.
• W2950479933 cites W2525778437 @default.
• W2950479933 cites W2535697732 @default.
• W2950479933 cites W2539201987 @default.
• W2950479933 cites W2552838200 @default.
• W2950479933 cites W2580192806 @default.
• W2950479933 cites W2594229957 @default.
• W2950479933 cites W2594990650 @default.
• W2950479933 cites W2743455151 @default.
• W2950479933 cites W2758514087 @default.
• W2950479933 cites W2760713883 @default.
• W2950479933 cites W2783419700 @default.
• W2950479933 cites W2949335953 @default.
• W2950479933 cites W2949643137 @default.
• W2950479933 cites W2949888546 @default.
• W2950479933 cites W2950527759 @default.
• W2950479933 cites W2950577311 @default.
• W2950479933 cites W2951008357 @default.
• W2950479933 cites W2952136670 @default.
• W2950479933 cites W2952479981 @default.
• W2950479933 cites W2962784628 @default.
• W2950479933 cites W2963011474 @default.
• W2950479933 cites W2963347649 @default.
• W2950479933 cites W2963463964 @default.
• W2950479933 cites W2963699608 @default.
• W2950479933 cites W2964121744 @default.
• W2950479933 cites W2964308564 @default.
• W2950479933 doi "https://doi.org/10.48550/arxiv.1711.09502" @default.
• W2950479933 hasPublicationYear "2017" @default.
• W2950479933 type Work @default.
• W2950479933 sameAs 2950479933 @default.
• W2950479933 citedByCount "2" @default.
• W2950479933 countsByYear W29504799332017 @default.
• W2950479933 countsByYear W29504799332019 @default.
• W2950479933 crossrefType "posted-content" @default.
• W2950479933 hasAuthorship W2950479933A5016804474 @default.
• W2950479933 hasAuthorship W2950479933A5022574607 @default.
• W2950479933 hasAuthorship W2950479933A5024821632 @default.
• W2950479933 hasAuthorship W2950479933A5025469065 @default.
• W2950479933 hasAuthorship W2950479933A5033423455 @default.
• W2950479933 hasAuthorship W2950479933A5045799619 @default.
• W2950479933 hasAuthorship W2950479933A5067322430 @default.
• W2950479933 hasBestOaLocation W29504799331 @default.
• W2950479933 hasConcept C104317684 @default.
• W2950479933 hasConcept C105580179 @default.
• W2950479933 hasConcept C111472728 @default.
• W2950479933 hasConcept C11413529 @default.
• W2950479933 hasConcept C119857082 @default.
• W2950479933 hasConcept C138885662 @default.
• W2950479933 hasConcept C149364088 @default.
• W2950479933 hasConcept C154775046 @default.
• W2950479933 hasConcept C154945302 @default.
• W2950479933 hasConcept C185592680 @default.
• W2950479933 hasConcept C203005215 @default.
• W2950479933 hasConcept C204321447 @default.
• W2950479933 hasConcept C2779530757 @default.
• W2950479933 hasConcept C28490314 @default.
• W2950479933 hasConcept C41008148 @default.
• W2950479933 hasConcept C41895202 @default.
• W2950479933 hasConcept C55493867 @default.
• W2950479933 hasConcept C57273362 @default.
• W2950479933 hasConcept C89611455 @default.
• W2950479933 hasConceptScore W2950479933C104317684 @default.
• W2950479933 hasConceptScore W2950479933C105580179 @default.
• W2950479933 hasConceptScore W2950479933C111472728 @default.
• W2950479933 hasConceptScore W2950479933C11413529 @default.
• W2950479933 hasConceptScore W2950479933C119857082 @default.
• W2950479933 hasConceptScore W2950479933C138885662 @default.
• W2950479933 hasConceptScore W2950479933C149364088 @default.
• W2950479933 hasConceptScore W2950479933C154775046 @default.
• W2950479933 hasConceptScore W2950479933C154945302 @default.
• W2950479933 hasConceptScore W2950479933C185592680 @default.
• W2950479933 hasConceptScore W2950479933C203005215 @default.
• W2950479933 hasConceptScore W2950479933C204321447 @default.
• W2950479933 hasConceptScore W2950479933C2779530757 @default.
• W2950479933 hasConceptScore W2950479933C28490314 @default.
• W2950479933 hasConceptScore W2950479933C41008148 @default.
• W2950479933 hasConceptScore W2950479933C41895202 @default.
• W2950479933 hasConceptScore W2950479933C55493867 @default.

• next