FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2983669550> ?p ?o ?g. }

• W2983669550 abstract "Neural Machine Translation (NMT) optimized by Maximum Likelihood Estimation (MLE) lacks the guarantee of translation adequacy. To alleviate this problem, we propose an NMT approach that heightens the adequacy in machine translation by transferring the semantic knowledge learned from bilingual sentence alignment. Specifically, we first design a discriminator that learns to estimate sentence aligning score over translation candidates, and then the learned semantic knowledge is transfered to the NMT model under an adversarial learning framework. We also propose a gated self-attention based encoder for sentence embedding. Furthermore, an N-pair training loss is introduced in our framework to aid the discriminator in better capturing lexical evidence in translation candidates. Experimental results show that our proposed method outperforms baseline NMT models on Chinese-to-English and English-to-German translation tasks. Further analysis also indicates the detailed semantic knowledge transfered from the discriminator to the NMT model." @default.
• W2983669550 created "2019-11-22" @default.
• W2983669550 creator A5005421447 @default.
• W2983669550 creator A5058216405 @default.
• W2983669550 creator A5073550269 @default.
• W2983669550 creator A5075237838 @default.
• W2983669550 creator A5087631670 @default.
• W2983669550 date "2019-01-01" @default.
• W2983669550 modified "2023-10-17" @default.
• W2983669550 title "Improving Neural Machine Translation by Achieving Knowledge Transfer with Sentence Alignment Learning" @default.
• W2983669550 cites W1753482797 @default.
• W2983669550 cites W2101105183 @default.
• W2983669550 cites W2116343275 @default.
• W2983669550 cites W2124807415 @default.
• W2983669550 cites W2130942839 @default.
• W2983669550 cites W2157331557 @default.
• W2983669550 cites W2183341477 @default.
• W2983669550 cites W2525778437 @default.
• W2983669550 cites W2546938941 @default.
• W2983669550 cites W2547875792 @default.
• W2983669550 cites W2555897561 @default.
• W2983669550 cites W2565378226 @default.
• W2983669550 cites W2576713500 @default.
• W2983669550 cites W2601324753 @default.
• W2983669550 cites W2758884106 @default.
• W2983669550 cites W2937971956 @default.
• W2983669550 cites W2962784628 @default.
• W2983669550 cites W2962926939 @default.
• W2983669550 cites W2963163972 @default.
• W2983669550 cites W2963260202 @default.
• W2983669550 cites W2963373786 @default.
• W2983669550 cites W2963403868 @default.
• W2983669550 cites W2963463964 @default.
• W2983669550 cites W2963506925 @default.
• W2983669550 cites W2963551569 @default.
• W2983669550 cites W2963643760 @default.
• W2983669550 cites W2963774520 @default.
• W2983669550 cites W2963970792 @default.
• W2983669550 cites W2964199361 @default.
• W2983669550 cites W2964201867 @default.
• W2983669550 cites W2964265128 @default.
• W2983669550 cites W2964268978 @default.
• W2983669550 cites W2964308564 @default.
• W2983669550 cites W3148140980 @default.
• W2983669550 cites W71795751 @default.
• W2983669550 doi "https://doi.org/10.18653/v1/k19-1025" @default.
• W2983669550 hasPublicationYear "2019" @default.
• W2983669550 type Work @default.
• W2983669550 sameAs 2983669550 @default.
• W2983669550 citedByCount "2" @default.
• W2983669550 countsByYear W29836695502021 @default.
• W2983669550 countsByYear W29836695502022 @default.
• W2983669550 crossrefType "proceedings-article" @default.
• W2983669550 hasAuthorship W2983669550A5005421447 @default.
• W2983669550 hasAuthorship W2983669550A5058216405 @default.
• W2983669550 hasAuthorship W2983669550A5073550269 @default.
• W2983669550 hasAuthorship W2983669550A5075237838 @default.
• W2983669550 hasAuthorship W2983669550A5087631670 @default.
• W2983669550 hasBestOaLocation W29836695501 @default.
• W2983669550 hasConcept C104317684 @default.
• W2983669550 hasConcept C105580179 @default.
• W2983669550 hasConcept C111919701 @default.
• W2983669550 hasConcept C118505674 @default.
• W2983669550 hasConcept C119857082 @default.
• W2983669550 hasConcept C135784402 @default.
• W2983669550 hasConcept C148526163 @default.
• W2983669550 hasConcept C149364088 @default.
• W2983669550 hasConcept C154945302 @default.
• W2983669550 hasConcept C185592680 @default.
• W2983669550 hasConcept C203005215 @default.
• W2983669550 hasConcept C204321447 @default.
• W2983669550 hasConcept C24687705 @default.
• W2983669550 hasConcept C2777530160 @default.
• W2983669550 hasConcept C2779803651 @default.
• W2983669550 hasConcept C28490314 @default.
• W2983669550 hasConcept C41008148 @default.
• W2983669550 hasConcept C41608201 @default.
• W2983669550 hasConcept C55493867 @default.
• W2983669550 hasConcept C76155785 @default.
• W2983669550 hasConcept C94915269 @default.
• W2983669550 hasConceptScore W2983669550C104317684 @default.
• W2983669550 hasConceptScore W2983669550C105580179 @default.
• W2983669550 hasConceptScore W2983669550C111919701 @default.
• W2983669550 hasConceptScore W2983669550C118505674 @default.
• W2983669550 hasConceptScore W2983669550C119857082 @default.
• W2983669550 hasConceptScore W2983669550C135784402 @default.
• W2983669550 hasConceptScore W2983669550C148526163 @default.
• W2983669550 hasConceptScore W2983669550C149364088 @default.
• W2983669550 hasConceptScore W2983669550C154945302 @default.
• W2983669550 hasConceptScore W2983669550C185592680 @default.
• W2983669550 hasConceptScore W2983669550C203005215 @default.
• W2983669550 hasConceptScore W2983669550C204321447 @default.
• W2983669550 hasConceptScore W2983669550C24687705 @default.
• W2983669550 hasConceptScore W2983669550C2777530160 @default.
• W2983669550 hasConceptScore W2983669550C2779803651 @default.
• W2983669550 hasConceptScore W2983669550C28490314 @default.
• W2983669550 hasConceptScore W2983669550C41008148 @default.
• W2983669550 hasConceptScore W2983669550C41608201 @default.
• W2983669550 hasConceptScore W2983669550C55493867 @default.
• W2983669550 hasConceptScore W2983669550C76155785 @default.

• next