Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100509098> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3100509098 abstract "Active learning is an efficient approach for mitigating data dependency when training neural machine translation (NMT) models. In this paper, we explore new training frameworks by incorporating active learning into various techniques such as transfer learning and iterative back-translation (IBT) under a limited human translation budget. We design a word frequency based acquisition function and combine it with a strong uncertainty based method. The combined method steadily outperforms all other acquisition functions in various scenarios. As far as we know, we are the first to do a large-scale study on actively training Transformer for NMT. Specifically, with a human translation budget of only 20% of the original parallel corpus, we manage to surpass Transformer trained on the entire parallel corpus in three language pairs." @default.
- W3100509098 created "2020-11-23" @default.
- W3100509098 creator A5008993224 @default.
- W3100509098 creator A5010735074 @default.
- W3100509098 creator A5014929343 @default.
- W3100509098 creator A5047580482 @default.
- W3100509098 date "2020-01-01" @default.
- W3100509098 modified "2023-09-25" @default.
- W3100509098 title "Active Learning Approaches to Enhancing Neural Machine Translation" @default.
- W3100509098 cites W1902237438 @default.
- W3100509098 cites W2087388117 @default.
- W3100509098 cites W2101105183 @default.
- W3100509098 cites W2171671120 @default.
- W3100509098 cites W2546938941 @default.
- W3100509098 cites W2785787385 @default.
- W3100509098 cites W2890095519 @default.
- W3100509098 cites W2898846200 @default.
- W3100509098 cites W2899181341 @default.
- W3100509098 cites W2908336025 @default.
- W3100509098 cites W2923622379 @default.
- W3100509098 cites W2950485982 @default.
- W3100509098 cites W2950810106 @default.
- W3100509098 cites W2951065878 @default.
- W3100509098 cites W2952474700 @default.
- W3100509098 cites W2952614664 @default.
- W3100509098 cites W2962784628 @default.
- W3100509098 cites W2962824887 @default.
- W3100509098 cites W2962890089 @default.
- W3100509098 cites W2963088995 @default.
- W3100509098 cites W2963216553 @default.
- W3100509098 cites W2963366389 @default.
- W3100509098 cites W2963403868 @default.
- W3100509098 cites W2963423556 @default.
- W3100509098 cites W2963505445 @default.
- W3100509098 cites W2963602293 @default.
- W3100509098 cites W2963633674 @default.
- W3100509098 cites W2963742748 @default.
- W3100509098 cites W2963949210 @default.
- W3100509098 cites W2964121744 @default.
- W3100509098 cites W2964265128 @default.
- W3100509098 cites W2964282813 @default.
- W3100509098 cites W2970043232 @default.
- W3100509098 cites W2970520743 @default.
- W3100509098 cites W2988121211 @default.
- W3100509098 cites W3034201598 @default.
- W3100509098 cites W3034938700 @default.
- W3100509098 cites W3084906606 @default.
- W3100509098 cites W3101889167 @default.
- W3100509098 cites W3103169714 @default.
- W3100509098 doi "https://doi.org/10.18653/v1/2020.findings-emnlp.162" @default.
- W3100509098 hasPublicationYear "2020" @default.
- W3100509098 type Work @default.
- W3100509098 sameAs 3100509098 @default.
- W3100509098 citedByCount "3" @default.
- W3100509098 countsByYear W31005090982021 @default.
- W3100509098 countsByYear W31005090982022 @default.
- W3100509098 crossrefType "proceedings-article" @default.
- W3100509098 hasAuthorship W3100509098A5008993224 @default.
- W3100509098 hasAuthorship W3100509098A5010735074 @default.
- W3100509098 hasAuthorship W3100509098A5014929343 @default.
- W3100509098 hasAuthorship W3100509098A5047580482 @default.
- W3100509098 hasBestOaLocation W31005090981 @default.
- W3100509098 hasConcept C104317684 @default.
- W3100509098 hasConcept C105580179 @default.
- W3100509098 hasConcept C119857082 @default.
- W3100509098 hasConcept C149364088 @default.
- W3100509098 hasConcept C154945302 @default.
- W3100509098 hasConcept C185592680 @default.
- W3100509098 hasConcept C203005215 @default.
- W3100509098 hasConcept C41008148 @default.
- W3100509098 hasConcept C55493867 @default.
- W3100509098 hasConceptScore W3100509098C104317684 @default.
- W3100509098 hasConceptScore W3100509098C105580179 @default.
- W3100509098 hasConceptScore W3100509098C119857082 @default.
- W3100509098 hasConceptScore W3100509098C149364088 @default.
- W3100509098 hasConceptScore W3100509098C154945302 @default.
- W3100509098 hasConceptScore W3100509098C185592680 @default.
- W3100509098 hasConceptScore W3100509098C203005215 @default.
- W3100509098 hasConceptScore W3100509098C41008148 @default.
- W3100509098 hasConceptScore W3100509098C55493867 @default.
- W3100509098 hasLocation W31005090981 @default.
- W3100509098 hasOpenAccess W3100509098 @default.
- W3100509098 hasPrimaryLocation W31005090981 @default.
- W3100509098 hasRelatedWork W1592339875 @default.
- W3100509098 hasRelatedWork W1974878518 @default.
- W3100509098 hasRelatedWork W2016948484 @default.
- W3100509098 hasRelatedWork W2313858059 @default.
- W3100509098 hasRelatedWork W2392593357 @default.
- W3100509098 hasRelatedWork W2435130738 @default.
- W3100509098 hasRelatedWork W2509777158 @default.
- W3100509098 hasRelatedWork W2961085424 @default.
- W3100509098 hasRelatedWork W4306674287 @default.
- W3100509098 hasRelatedWork W97426361 @default.
- W3100509098 isParatext "false" @default.
- W3100509098 isRetracted "false" @default.
- W3100509098 magId "3100509098" @default.
- W3100509098 workType "article" @default.