Matches in SemOpenAlex for { <https://semopenalex.org/work/W3092626778> ?p ?o ?g. }
- W3092626778 abstract "Document-level neural machine translation has yielded attractive improvements. However, majority of existing methods roughly use all context sentences in a fixed scope. They neglect the fact that different source sentences need different sizes of context. To address this problem, we propose an effective approach to select dynamic context so that the document-level translation model can utilize the more useful selected context sentences to produce better translations. Specifically, we introduce a selection module that is independent of the translation module to score each candidate context sentence. Then, we propose two strategies to explicitly select a variable number of context sentences and feed them into the translation module. We train the two modules end-to-end via reinforcement learning. A novel reward is proposed to encourage the selection and utilization of dynamic context sentences. Experiments demonstrate that our approach can select adaptive context sentences for different source sentences, and significantly improves the performance of document-level translation methods." @default.
- W3092626778 created "2020-10-15" @default.
- W3092626778 creator A5001541516 @default.
- W3092626778 creator A5015785439 @default.
- W3092626778 creator A5086778602 @default.
- W3092626778 creator A5089015542 @default.
- W3092626778 date "2020-10-08" @default.
- W3092626778 modified "2023-09-23" @default.
- W3092626778 title "Dynamic Context Selection for Document-level Neural Machine Translation via Reinforcement Learning" @default.
- W3092626778 cites W1902237438 @default.
- W3092626778 cites W1910131649 @default.
- W3092626778 cites W2053154970 @default.
- W3092626778 cites W2101105183 @default.
- W3092626778 cites W2157331557 @default.
- W3092626778 cites W2561274697 @default.
- W3092626778 cites W2608029998 @default.
- W3092626778 cites W2669742347 @default.
- W3092626778 cites W2799051177 @default.
- W3092626778 cites W2808508619 @default.
- W3092626778 cites W2848493808 @default.
- W3092626778 cites W2888442053 @default.
- W3092626778 cites W2891534142 @default.
- W3092626778 cites W2928667520 @default.
- W3092626778 cites W2935811960 @default.
- W3092626778 cites W2952446148 @default.
- W3092626778 cites W2953190730 @default.
- W3092626778 cites W2962700074 @default.
- W3092626778 cites W2962712961 @default.
- W3092626778 cites W2962784628 @default.
- W3092626778 cites W2962802109 @default.
- W3092626778 cites W2962943802 @default.
- W3092626778 cites W2962985882 @default.
- W3092626778 cites W2963084599 @default.
- W3092626778 cites W2963123301 @default.
- W3092626778 cites W2963260202 @default.
- W3092626778 cites W2963341956 @default.
- W3092626778 cites W2963403868 @default.
- W3092626778 cites W2963842551 @default.
- W3092626778 cites W2964093087 @default.
- W3092626778 cites W2964120396 @default.
- W3092626778 cites W2964202354 @default.
- W3092626778 cites W2964289193 @default.
- W3092626778 cites W2964291396 @default.
- W3092626778 cites W2964308564 @default.
- W3092626778 cites W2970529093 @default.
- W3092626778 cites W2971248291 @default.
- W3092626778 cites W2971278086 @default.
- W3092626778 cites W2971347700 @default.
- W3092626778 cites W2977772147 @default.
- W3092626778 cites W2983108239 @default.
- W3092626778 cites W3034700448 @default.
- W3092626778 cites W3035169087 @default.
- W3092626778 cites W3035520602 @default.
- W3092626778 cites W3102507836 @default.
- W3092626778 doi "https://doi.org/10.48550/arxiv.2010.04314" @default.
- W3092626778 hasPublicationYear "2020" @default.
- W3092626778 type Work @default.
- W3092626778 sameAs 3092626778 @default.
- W3092626778 citedByCount "0" @default.
- W3092626778 crossrefType "posted-content" @default.
- W3092626778 hasAuthorship W3092626778A5001541516 @default.
- W3092626778 hasAuthorship W3092626778A5015785439 @default.
- W3092626778 hasAuthorship W3092626778A5086778602 @default.
- W3092626778 hasAuthorship W3092626778A5089015542 @default.
- W3092626778 hasBestOaLocation W30926267781 @default.
- W3092626778 hasConcept C119857082 @default.
- W3092626778 hasConcept C151730666 @default.
- W3092626778 hasConcept C154945302 @default.
- W3092626778 hasConcept C199360897 @default.
- W3092626778 hasConcept C203005215 @default.
- W3092626778 hasConcept C204321447 @default.
- W3092626778 hasConcept C2777530160 @default.
- W3092626778 hasConcept C2778012447 @default.
- W3092626778 hasConcept C2779343474 @default.
- W3092626778 hasConcept C41008148 @default.
- W3092626778 hasConcept C81917197 @default.
- W3092626778 hasConcept C86803240 @default.
- W3092626778 hasConcept C97541855 @default.
- W3092626778 hasConceptScore W3092626778C119857082 @default.
- W3092626778 hasConceptScore W3092626778C151730666 @default.
- W3092626778 hasConceptScore W3092626778C154945302 @default.
- W3092626778 hasConceptScore W3092626778C199360897 @default.
- W3092626778 hasConceptScore W3092626778C203005215 @default.
- W3092626778 hasConceptScore W3092626778C204321447 @default.
- W3092626778 hasConceptScore W3092626778C2777530160 @default.
- W3092626778 hasConceptScore W3092626778C2778012447 @default.
- W3092626778 hasConceptScore W3092626778C2779343474 @default.
- W3092626778 hasConceptScore W3092626778C41008148 @default.
- W3092626778 hasConceptScore W3092626778C81917197 @default.
- W3092626778 hasConceptScore W3092626778C86803240 @default.
- W3092626778 hasConceptScore W3092626778C97541855 @default.
- W3092626778 hasLocation W30926267781 @default.
- W3092626778 hasOpenAccess W3092626778 @default.
- W3092626778 hasPrimaryLocation W30926267781 @default.
- W3092626778 hasRelatedWork W11360949 @default.
- W3092626778 hasRelatedWork W11511616 @default.
- W3092626778 hasRelatedWork W11890898 @default.
- W3092626778 hasRelatedWork W14058886 @default.
- W3092626778 hasRelatedWork W3939803 @default.
- W3092626778 hasRelatedWork W867563 @default.