FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2803481662> ?p ?o ?g. }

• W2803481662 abstract "Neural attention (NA) has become a key component of sequence-to-sequence models that yield state-of-the-art performance in as hard tasks as abstractive document summarization (ADS) and video captioning (VC). NA mechanisms perform inference of context vectors; these constitute weighted sums of deterministic input sequence encodings, adaptively sourced over long temporal horizons. Inspired from recent work in the field of amortized variational inference (AVI), in this work we consider treating the context vectors generated by soft-attention (SA) models as latent variables, with approximate finite mixture model posteriors inferred via AVI. We posit that this formulation may yield stronger generalization capacity, in line with the outcomes of existing applications of AVI to deep networks. To illustrate our method, we implement it and experimentally evaluate it considering challenging ADS, VC, and MT benchmarks. This way, we exhibit its improved effectiveness over state-of-the-art alternatives." @default.
• W2803481662 created "2018-06-01" @default.
• W2803481662 creator A5026099748 @default.
• W2803481662 creator A5040025256 @default.
• W2803481662 creator A5040795718 @default.
• W2803481662 creator A5082936206 @default.
• W2803481662 date "2018-09-27" @default.
• W2803481662 modified "2023-09-27" @default.
• W2803481662 title "Amortized Context Vector Inference for Sequence-to-Sequence Networks" @default.
• W2803481662 cites W1514535095 @default.
• W2803481662 cites W1586939924 @default.
• W2803481662 cites W1591801644 @default.
• W2803481662 cites W1793121960 @default.
• W2803481662 cites W1902237438 @default.
• W2803481662 cites W1956340063 @default.
• W2803481662 cites W1959608418 @default.
• W2803481662 cites W1987052429 @default.
• W2803481662 cites W2030656756 @default.
• W2803481662 cites W2064675550 @default.
• W2803481662 cites W2097117768 @default.
• W2803481662 cites W2101105183 @default.
• W2803481662 cites W2119717200 @default.
• W2803481662 cites W2122410182 @default.
• W2803481662 cites W2125838338 @default.
• W2803481662 cites W2130942839 @default.
• W2803481662 cites W2133459682 @default.
• W2803481662 cites W2154652894 @default.
• W2803481662 cites W2155893237 @default.
• W2803481662 cites W2172085063 @default.
• W2803481662 cites W2547875792 @default.
• W2803481662 cites W2567525733 @default.
• W2803481662 cites W2574535369 @default.
• W2803481662 cites W2586050494 @default.
• W2803481662 cites W2606974598 @default.
• W2803481662 cites W2859444450 @default.
• W2803481662 cites W2905927205 @default.
• W2803481662 cites W2962784628 @default.
• W2803481662 cites W2962965405 @default.
• W2803481662 cites W2963090522 @default.
• W2803481662 cites W2963135265 @default.
• W2803481662 cites W2963260202 @default.
• W2803481662 cites W2963403868 @default.
• W2803481662 cites W2963650605 @default.
• W2803481662 cites W2963929190 @default.
• W2803481662 cites W2964121744 @default.
• W2803481662 cites W2964308564 @default.
• W2803481662 hasPublicationYear "2018" @default.
• W2803481662 type Work @default.
• W2803481662 sameAs 2803481662 @default.
• W2803481662 citedByCount "0" @default.
• W2803481662 crossrefType "posted-content" @default.
• W2803481662 hasAuthorship W2803481662A5026099748 @default.
• W2803481662 hasAuthorship W2803481662A5040025256 @default.
• W2803481662 hasAuthorship W2803481662A5040795718 @default.
• W2803481662 hasAuthorship W2803481662A5082936206 @default.
• W2803481662 hasConcept C11413529 @default.
• W2803481662 hasConcept C119857082 @default.
• W2803481662 hasConcept C134306372 @default.
• W2803481662 hasConcept C151730666 @default.
• W2803481662 hasConcept C154945302 @default.
• W2803481662 hasConcept C170858558 @default.
• W2803481662 hasConcept C177148314 @default.
• W2803481662 hasConcept C202444582 @default.
• W2803481662 hasConcept C2776214188 @default.
• W2803481662 hasConcept C2778112365 @default.
• W2803481662 hasConcept C2779343474 @default.
• W2803481662 hasConcept C33923547 @default.
• W2803481662 hasConcept C41008148 @default.
• W2803481662 hasConcept C48103436 @default.
• W2803481662 hasConcept C54355233 @default.
• W2803481662 hasConcept C80444323 @default.
• W2803481662 hasConcept C86803240 @default.
• W2803481662 hasConcept C9652623 @default.
• W2803481662 hasConceptScore W2803481662C11413529 @default.
• W2803481662 hasConceptScore W2803481662C119857082 @default.
• W2803481662 hasConceptScore W2803481662C134306372 @default.
• W2803481662 hasConceptScore W2803481662C151730666 @default.
• W2803481662 hasConceptScore W2803481662C154945302 @default.
• W2803481662 hasConceptScore W2803481662C170858558 @default.
• W2803481662 hasConceptScore W2803481662C177148314 @default.
• W2803481662 hasConceptScore W2803481662C202444582 @default.
• W2803481662 hasConceptScore W2803481662C2776214188 @default.
• W2803481662 hasConceptScore W2803481662C2778112365 @default.
• W2803481662 hasConceptScore W2803481662C2779343474 @default.
• W2803481662 hasConceptScore W2803481662C33923547 @default.
• W2803481662 hasConceptScore W2803481662C41008148 @default.
• W2803481662 hasConceptScore W2803481662C48103436 @default.
• W2803481662 hasConceptScore W2803481662C54355233 @default.
• W2803481662 hasConceptScore W2803481662C80444323 @default.
• W2803481662 hasConceptScore W2803481662C86803240 @default.
• W2803481662 hasConceptScore W2803481662C9652623 @default.
• W2803481662 hasLocation W28034816621 @default.
• W2803481662 hasOpenAccess W2803481662 @default.
• W2803481662 hasPrimaryLocation W28034816621 @default.
• W2803481662 hasRelatedWork W2006988199 @default.
• W2803481662 hasRelatedWork W2248934801 @default.
• W2803481662 hasRelatedWork W2468625129 @default.
• W2803481662 hasRelatedWork W2766438578 @default.
• W2803481662 hasRelatedWork W2799004920 @default.
• W2803481662 hasRelatedWork W2805880533 @default.

• next