FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3036558623> ?p ?o ?g. }

• W3036558623 abstract "In this paper, we propose a new adversarial augmentation method for Neural Machine Translation (NMT). The main idea is to minimize the vicinal risk over virtual sentences sampled from two vicinity distributions, of which the crucial one is a novel vicinity distribution for adversarial sentences that describes a smooth interpolated embedding space centered around observed training sentence pairs. We then discuss our approach, AdvAug, to train NMT models using the embeddings of virtual sentences in sequence-to-sequence learning. Experiments on Chinese-English, English-French, and English-German translation benchmarks show that AdvAug achieves significant improvements over the Transformer (up to 4.9 BLEU points), and substantially outperforms other data augmentation techniques (e.g. back-translation) without using extra corpora." @default.
• W3036558623 created "2020-06-25" @default.
• W3036558623 creator A5018646392 @default.
• W3036558623 creator A5036643737 @default.
• W3036558623 creator A5047699861 @default.
• W3036558623 creator A5090730336 @default.
• W3036558623 date "2020-06-21" @default.
• W3036558623 modified "2023-10-18" @default.
• W3036558623 title "AdvAug: Robust Adversarial Augmentation for Neural Machine Translation" @default.
• W3036558623 cites W2099471712 @default.
• W3036558623 cites W2151239833 @default.
• W3036558623 cites W2525778437 @default.
• W3036558623 cites W2561274697 @default.
• W3036558623 cites W2811010710 @default.
• W3036558623 cites W2888519496 @default.
• W3036558623 cites W2889326796 @default.
• W3036558623 cites W2922293812 @default.
• W3036558623 cites W2943865428 @default.
• W3036558623 cites W2946817437 @default.
• W3036558623 cites W2950651087 @default.
• W3036558623 cites W2962784628 @default.
• W3036558623 cites W2963216553 @default.
• W3036558623 cites W2963399829 @default.
• W3036558623 cites W2963403868 @default.
• W3036558623 cites W2963661177 @default.
• W3036558623 cites W2963823140 @default.
• W3036558623 cites W2964040467 @default.
• W3036558623 cites W2964048171 @default.
• W3036558623 cites W2964153729 @default.
• W3036558623 cites W2964265128 @default.
• W3036558623 cites W2964308564 @default.
• W3036558623 cites W2970558573 @default.
• W3036558623 cites W2984051011 @default.
• W3036558623 cites W2990812465 @default.
• W3036558623 cites W3038785057 @default.
• W3036558623 cites W3098341425 @default.
• W3036558623 doi "https://doi.org/10.48550/arxiv.2006.11834" @default.
• W3036558623 hasPublicationYear "2020" @default.
• W3036558623 type Work @default.
• W3036558623 sameAs 3036558623 @default.
• W3036558623 citedByCount "5" @default.
• W3036558623 countsByYear W30365586232020 @default.
• W3036558623 countsByYear W30365586232021 @default.
• W3036558623 crossrefType "posted-content" @default.
• W3036558623 hasAuthorship W3036558623A5018646392 @default.
• W3036558623 hasAuthorship W3036558623A5036643737 @default.
• W3036558623 hasAuthorship W3036558623A5047699861 @default.
• W3036558623 hasAuthorship W3036558623A5090730336 @default.
• W3036558623 hasBestOaLocation W30365586231 @default.
• W3036558623 hasConcept C104317684 @default.
• W3036558623 hasConcept C105580179 @default.
• W3036558623 hasConcept C119599485 @default.
• W3036558623 hasConcept C127413603 @default.
• W3036558623 hasConcept C149364088 @default.
• W3036558623 hasConcept C154945302 @default.
• W3036558623 hasConcept C165801399 @default.
• W3036558623 hasConcept C185592680 @default.
• W3036558623 hasConcept C203005215 @default.
• W3036558623 hasConcept C204321447 @default.
• W3036558623 hasConcept C24687705 @default.
• W3036558623 hasConcept C2777530160 @default.
• W3036558623 hasConcept C2778112365 @default.
• W3036558623 hasConcept C28490314 @default.
• W3036558623 hasConcept C37736160 @default.
• W3036558623 hasConcept C41008148 @default.
• W3036558623 hasConcept C41608201 @default.
• W3036558623 hasConcept C54355233 @default.
• W3036558623 hasConcept C55493867 @default.
• W3036558623 hasConcept C66322947 @default.
• W3036558623 hasConcept C86803240 @default.
• W3036558623 hasConceptScore W3036558623C104317684 @default.
• W3036558623 hasConceptScore W3036558623C105580179 @default.
• W3036558623 hasConceptScore W3036558623C119599485 @default.
• W3036558623 hasConceptScore W3036558623C127413603 @default.
• W3036558623 hasConceptScore W3036558623C149364088 @default.
• W3036558623 hasConceptScore W3036558623C154945302 @default.
• W3036558623 hasConceptScore W3036558623C165801399 @default.
• W3036558623 hasConceptScore W3036558623C185592680 @default.
• W3036558623 hasConceptScore W3036558623C203005215 @default.
• W3036558623 hasConceptScore W3036558623C204321447 @default.
• W3036558623 hasConceptScore W3036558623C24687705 @default.
• W3036558623 hasConceptScore W3036558623C2777530160 @default.
• W3036558623 hasConceptScore W3036558623C2778112365 @default.
• W3036558623 hasConceptScore W3036558623C28490314 @default.
• W3036558623 hasConceptScore W3036558623C37736160 @default.
• W3036558623 hasConceptScore W3036558623C41008148 @default.
• W3036558623 hasConceptScore W3036558623C41608201 @default.
• W3036558623 hasConceptScore W3036558623C54355233 @default.
• W3036558623 hasConceptScore W3036558623C55493867 @default.
• W3036558623 hasConceptScore W3036558623C66322947 @default.
• W3036558623 hasConceptScore W3036558623C86803240 @default.
• W3036558623 hasLocation W30365586231 @default.
• W3036558623 hasOpenAccess W3036558623 @default.
• W3036558623 hasPrimaryLocation W30365586231 @default.
• W3036558623 hasRelatedWork W170118632 @default.
• W3036558623 hasRelatedWork W193726211 @default.
• W3036558623 hasRelatedWork W2359582629 @default.
• W3036558623 hasRelatedWork W2507994462 @default.
• W3036558623 hasRelatedWork W2578428189 @default.
• W3036558623 hasRelatedWork W2606032440 @default.

• next