FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3116179216> ?p ?o ?g. }

• W3116179216 endingPage "13342" @default.
• W3116179216 startingPage "13333" @default.
• W3116179216 abstract "Neural machine translation often adopts the fine-tuning approach to adapt to specific domains. However, nonrestricted fine-tuning can easily degrade on the general domain and over-fit to the target domain. To mitigate the issue, we propose Prune-Tune, a novel domain adaptation method via gradual pruning. It learns tiny domain-specific sub-networks during fine-tuning on new domains. Prune-Tune alleviates the over-fitting and the degradation problem without model modification. Furthermore, Prune-Tune is able to sequentially learn a single network with multiple disjoint domain-specific sub-networks for multiple domains. Empirical experiment results show that Prune-Tune outperforms several strong competitors in the target domain test set without sacrificing the quality on the general domain in both single and multi-domain settings. The source code and data are available at https://github.com/ohlionel/Prune-Tune." @default.
• W3116179216 created "2021-01-05" @default.
• W3116179216 creator A5017907143 @default.
• W3116179216 creator A5043256966 @default.
• W3116179216 creator A5044665993 @default.
• W3116179216 creator A5055734022 @default.
• W3116179216 creator A5075889094 @default.
• W3116179216 date "2021-05-18" @default.
• W3116179216 modified "2023-09-30" @default.
• W3116179216 title "Finding Sparse Structures for Domain Specific Neural Machine Translation" @default.
• W3116179216 cites W1753482797 @default.
• W3116179216 cites W1905522558 @default.
• W3116179216 cites W2101105183 @default.
• W3116179216 cites W2114766824 @default.
• W3116179216 cites W2137387514 @default.
• W3116179216 cites W2250876691 @default.
• W3116179216 cites W2560647685 @default.
• W3116179216 cites W2567571499 @default.
• W3116179216 cites W2626778328 @default.
• W3116179216 cites W2740718109 @default.
• W3116179216 cites W2744813330 @default.
• W3116179216 cites W2759461255 @default.
• W3116179216 cites W2764043458 @default.
• W3116179216 cites W2788388592 @default.
• W3116179216 cites W2791091755 @default.
• W3116179216 cites W2803241009 @default.
• W3116179216 cites W2805394970 @default.
• W3116179216 cites W2886342729 @default.
• W3116179216 cites W2945383715 @default.
• W3116179216 cites W2949888546 @default.
• W3116179216 cites W2949920209 @default.
• W3116179216 cites W2950760213 @default.
• W3116179216 cites W2951569836 @default.
• W3116179216 cites W2954647460 @default.
• W3116179216 cites W2962784628 @default.
• W3116179216 cites W2962796826 @default.
• W3116179216 cites W2962965870 @default.
• W3116179216 cites W2963072899 @default.
• W3116179216 cites W2963216553 @default.
• W3116179216 cites W2963250244 @default.
• W3116179216 cites W2963626527 @default.
• W3116179216 cites W2963674932 @default.
• W3116179216 cites W2963813662 @default.
• W3116179216 cites W2964121744 @default.
• W3116179216 cites W2964308564 @default.
• W3116179216 cites W2970286654 @default.
• W3116179216 cites W2970925270 @default.
• W3116179216 cites W2971234316 @default.
• W3116179216 cites W2980994576 @default.
• W3116179216 cites W2995998574 @default.
• W3116179216 cites W2998103904 @default.
• W3116179216 cites W3034214887 @default.
• W3116179216 cites W3034709122 @default.
• W3116179216 cites W3204406378 @default.
• W3116179216 cites W630532510 @default.
• W3116179216 doi "https://doi.org/10.1609/aaai.v35i15.17574" @default.
• W3116179216 hasPublicationYear "2021" @default.
• W3116179216 type Work @default.
• W3116179216 sameAs 3116179216 @default.
• W3116179216 citedByCount "3" @default.
• W3116179216 countsByYear W31161792162022 @default.
• W3116179216 countsByYear W31161792162023 @default.
• W3116179216 crossrefType "journal-article" @default.
• W3116179216 hasAuthorship W3116179216A5017907143 @default.
• W3116179216 hasAuthorship W3116179216A5043256966 @default.
• W3116179216 hasAuthorship W3116179216A5044665993 @default.
• W3116179216 hasAuthorship W3116179216A5055734022 @default.
• W3116179216 hasAuthorship W3116179216A5075889094 @default.
• W3116179216 hasBestOaLocation W31161792161 @default.
• W3116179216 hasConcept C104317684 @default.
• W3116179216 hasConcept C105580179 @default.
• W3116179216 hasConcept C108010975 @default.
• W3116179216 hasConcept C114614502 @default.
• W3116179216 hasConcept C119857082 @default.
• W3116179216 hasConcept C134306372 @default.
• W3116179216 hasConcept C149364088 @default.
• W3116179216 hasConcept C154945302 @default.
• W3116179216 hasConcept C177264268 @default.
• W3116179216 hasConcept C185592680 @default.
• W3116179216 hasConcept C199360897 @default.
• W3116179216 hasConcept C203005215 @default.
• W3116179216 hasConcept C2776760102 @default.
• W3116179216 hasConcept C33923547 @default.
• W3116179216 hasConcept C36503486 @default.
• W3116179216 hasConcept C41008148 @default.
• W3116179216 hasConcept C45340560 @default.
• W3116179216 hasConcept C50644808 @default.
• W3116179216 hasConcept C55493867 @default.
• W3116179216 hasConcept C6557445 @default.
• W3116179216 hasConcept C86803240 @default.
• W3116179216 hasConceptScore W3116179216C104317684 @default.
• W3116179216 hasConceptScore W3116179216C105580179 @default.
• W3116179216 hasConceptScore W3116179216C108010975 @default.
• W3116179216 hasConceptScore W3116179216C114614502 @default.
• W3116179216 hasConceptScore W3116179216C119857082 @default.
• W3116179216 hasConceptScore W3116179216C134306372 @default.
• W3116179216 hasConceptScore W3116179216C149364088 @default.
• W3116179216 hasConceptScore W3116179216C154945302 @default.

• next