Matches in SemOpenAlex for { <https://semopenalex.org/work/W2890623477> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2890623477 abstract "Neural Machine Translation (NMT) in low-resource settings and of morphologically rich languages is made difficult in part by data sparsity of vocabulary words. Several methods have been used to help reduce this sparsity, notably Byte-Pair Encoding (BPE) and a character-based CNN layer (charCNN). However, the charCNN has largely been neglected, possibly because it has only been compared to BPE rather than combined with it. We argue for a reconsideration of the charCNN, based on cross-lingual improvements on low-resource data. We translate from 8 languages into English, using a multi-way parallel collection of TED transcripts. We find that in most cases, using both BPE and a charCNN performs best, while in Hebrew, using a charCNN over words is best." @default.
- W2890623477 created "2018-09-27" @default.
- W2890623477 creator A5054715496 @default.
- W2890623477 creator A5070418792 @default.
- W2890623477 date "2018-09-05" @default.
- W2890623477 modified "2023-09-27" @default.
- W2890623477 title "BPE and CharCNNs for Translation of Morphology: A Cross-Lingual Comparison and Analysis" @default.
- W2890623477 cites W1938755728 @default.
- W2890623477 cites W2084086631 @default.
- W2890623477 cites W2220350356 @default.
- W2890623477 cites W2725082186 @default.
- W2890623477 cites W2767899794 @default.
- W2890623477 cites W2949335953 @default.
- W2890623477 cites W2962784628 @default.
- W2890623477 cites W46679369 @default.
- W2890623477 cites W6908809 @default.
- W2890623477 hasPublicationYear "2018" @default.
- W2890623477 type Work @default.
- W2890623477 sameAs 2890623477 @default.
- W2890623477 citedByCount "6" @default.
- W2890623477 countsByYear W28906234772018 @default.
- W2890623477 countsByYear W28906234772019 @default.
- W2890623477 countsByYear W28906234772020 @default.
- W2890623477 countsByYear W28906234772021 @default.
- W2890623477 crossrefType "posted-content" @default.
- W2890623477 hasAuthorship W2890623477A5054715496 @default.
- W2890623477 hasAuthorship W2890623477A5070418792 @default.
- W2890623477 hasConcept C104317684 @default.
- W2890623477 hasConcept C105580179 @default.
- W2890623477 hasConcept C138885662 @default.
- W2890623477 hasConcept C149364088 @default.
- W2890623477 hasConcept C154945302 @default.
- W2890623477 hasConcept C203005215 @default.
- W2890623477 hasConcept C204321447 @default.
- W2890623477 hasConcept C206345919 @default.
- W2890623477 hasConcept C2524010 @default.
- W2890623477 hasConcept C2777601683 @default.
- W2890623477 hasConcept C2780861071 @default.
- W2890623477 hasConcept C31258907 @default.
- W2890623477 hasConcept C33923547 @default.
- W2890623477 hasConcept C41008148 @default.
- W2890623477 hasConcept C41895202 @default.
- W2890623477 hasConcept C55493867 @default.
- W2890623477 hasConcept C86803240 @default.
- W2890623477 hasConcept C91304198 @default.
- W2890623477 hasConceptScore W2890623477C104317684 @default.
- W2890623477 hasConceptScore W2890623477C105580179 @default.
- W2890623477 hasConceptScore W2890623477C138885662 @default.
- W2890623477 hasConceptScore W2890623477C149364088 @default.
- W2890623477 hasConceptScore W2890623477C154945302 @default.
- W2890623477 hasConceptScore W2890623477C203005215 @default.
- W2890623477 hasConceptScore W2890623477C204321447 @default.
- W2890623477 hasConceptScore W2890623477C206345919 @default.
- W2890623477 hasConceptScore W2890623477C2524010 @default.
- W2890623477 hasConceptScore W2890623477C2777601683 @default.
- W2890623477 hasConceptScore W2890623477C2780861071 @default.
- W2890623477 hasConceptScore W2890623477C31258907 @default.
- W2890623477 hasConceptScore W2890623477C33923547 @default.
- W2890623477 hasConceptScore W2890623477C41008148 @default.
- W2890623477 hasConceptScore W2890623477C41895202 @default.
- W2890623477 hasConceptScore W2890623477C55493867 @default.
- W2890623477 hasConceptScore W2890623477C86803240 @default.
- W2890623477 hasConceptScore W2890623477C91304198 @default.
- W2890623477 hasLocation W28906234771 @default.
- W2890623477 hasOpenAccess W2890623477 @default.
- W2890623477 hasPrimaryLocation W28906234771 @default.
- W2890623477 hasRelatedWork W2752630748 @default.
- W2890623477 hasRelatedWork W2755686320 @default.
- W2890623477 hasRelatedWork W2783181001 @default.
- W2890623477 hasRelatedWork W2798473395 @default.
- W2890623477 hasRelatedWork W2805490244 @default.
- W2890623477 hasRelatedWork W2807931239 @default.
- W2890623477 hasRelatedWork W2885225491 @default.
- W2890623477 hasRelatedWork W2887945126 @default.
- W2890623477 hasRelatedWork W2949862135 @default.
- W2890623477 hasRelatedWork W2962784628 @default.
- W2890623477 hasRelatedWork W2963831310 @default.
- W2890623477 hasRelatedWork W2964296923 @default.
- W2890623477 hasRelatedWork W2971081948 @default.
- W2890623477 hasRelatedWork W2988094279 @default.
- W2890623477 hasRelatedWork W2989094907 @default.
- W2890623477 hasRelatedWork W3015650676 @default.
- W2890623477 hasRelatedWork W3184958573 @default.
- W2890623477 hasRelatedWork W3209166801 @default.
- W2890623477 hasRelatedWork W3118883478 @default.
- W2890623477 hasRelatedWork W3213112269 @default.
- W2890623477 isParatext "false" @default.
- W2890623477 isRetracted "false" @default.
- W2890623477 magId "2890623477" @default.
- W2890623477 workType "article" @default.