Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963525826> ?p ?o ?g. }
- W2963525826 abstract "A widespread approach to processing spoken language is to first automatically transcribe it into text. An alternative is to use an end-to-end approach: recent works have proposed to learn semantic embeddings of spoken language from images with spoken captions, without an intermediate transcription step. We propose to use multitask learning to exploit existing transcribed speech within the end-to-end setting. We describe a three-task architecture which combines the objectives of matching spoken captions with corresponding images, speech with text, and text with images. We show that the addition of the speech/text task leads to substantial performance improvements on image retrieval when compared to training the speech/image task in isolation. We conjecture that this is due to a strong inductive bias transcribed speech provides to the model, and offer supporting evidence for this." @default.
- W2963525826 created "2019-07-30" @default.
- W2963525826 creator A5022698890 @default.
- W2963525826 date "2019-01-01" @default.
- W2963525826 modified "2023-10-18" @default.
- W2963525826 title "Symbolic Inductive Bias for Visually Grounded Learning of Spoken Language" @default.
- W2963525826 cites W1494198834 @default.
- W2963525826 cites W1924770834 @default.
- W2963525826 cites W1978660892 @default.
- W2963525826 cites W2094035326 @default.
- W2963525826 cites W2106731506 @default.
- W2963525826 cites W2107917162 @default.
- W2963525826 cites W2117539524 @default.
- W2963525826 cites W2516255829 @default.
- W2963525826 cites W2531207078 @default.
- W2963525826 cites W2556468274 @default.
- W2963525826 cites W2556930864 @default.
- W2963525826 cites W2586148577 @default.
- W2963525826 cites W2592170186 @default.
- W2963525826 cites W2603679025 @default.
- W2963525826 cites W2758849341 @default.
- W2963525826 cites W2765507192 @default.
- W2963525826 cites W2784231336 @default.
- W2963525826 cites W2809324505 @default.
- W2963525826 cites W2888882903 @default.
- W2963525826 cites W2889077436 @default.
- W2963525826 cites W2890907172 @default.
- W2963525826 cites W2913340405 @default.
- W2963525826 cites W2962753610 @default.
- W2963525826 cites W2962835968 @default.
- W2963525826 cites W2962862718 @default.
- W2963525826 cites W2962982474 @default.
- W2963525826 cites W2963303028 @default.
- W2963525826 cites W2963330681 @default.
- W2963525826 cites W2963496089 @default.
- W2963525826 cites W2963842982 @default.
- W2963525826 cites W2963902314 @default.
- W2963525826 cites W2963983719 @default.
- W2963525826 cites W2964001192 @default.
- W2963525826 cites W3105148948 @default.
- W2963525826 cites W68733909 @default.
- W2963525826 doi "https://doi.org/10.18653/v1/p19-1647" @default.
- W2963525826 hasPublicationYear "2019" @default.
- W2963525826 type Work @default.
- W2963525826 sameAs 2963525826 @default.
- W2963525826 citedByCount "30" @default.
- W2963525826 countsByYear W29635258262019 @default.
- W2963525826 countsByYear W29635258262020 @default.
- W2963525826 countsByYear W29635258262021 @default.
- W2963525826 countsByYear W29635258262022 @default.
- W2963525826 countsByYear W29635258262023 @default.
- W2963525826 crossrefType "proceedings-article" @default.
- W2963525826 hasAuthorship W2963525826A5022698890 @default.
- W2963525826 hasBestOaLocation W29635258261 @default.
- W2963525826 hasConcept C138885662 @default.
- W2963525826 hasConcept C154945302 @default.
- W2963525826 hasConcept C162324750 @default.
- W2963525826 hasConcept C165696696 @default.
- W2963525826 hasConcept C179926584 @default.
- W2963525826 hasConcept C187736073 @default.
- W2963525826 hasConcept C197352929 @default.
- W2963525826 hasConcept C204321447 @default.
- W2963525826 hasConcept C2776230583 @default.
- W2963525826 hasConcept C2780451532 @default.
- W2963525826 hasConcept C28006648 @default.
- W2963525826 hasConcept C28490314 @default.
- W2963525826 hasConcept C38652104 @default.
- W2963525826 hasConcept C41008148 @default.
- W2963525826 hasConcept C41895202 @default.
- W2963525826 hasConceptScore W2963525826C138885662 @default.
- W2963525826 hasConceptScore W2963525826C154945302 @default.
- W2963525826 hasConceptScore W2963525826C162324750 @default.
- W2963525826 hasConceptScore W2963525826C165696696 @default.
- W2963525826 hasConceptScore W2963525826C179926584 @default.
- W2963525826 hasConceptScore W2963525826C187736073 @default.
- W2963525826 hasConceptScore W2963525826C197352929 @default.
- W2963525826 hasConceptScore W2963525826C204321447 @default.
- W2963525826 hasConceptScore W2963525826C2776230583 @default.
- W2963525826 hasConceptScore W2963525826C2780451532 @default.
- W2963525826 hasConceptScore W2963525826C28006648 @default.
- W2963525826 hasConceptScore W2963525826C28490314 @default.
- W2963525826 hasConceptScore W2963525826C38652104 @default.
- W2963525826 hasConceptScore W2963525826C41008148 @default.
- W2963525826 hasConceptScore W2963525826C41895202 @default.
- W2963525826 hasLocation W29635258261 @default.
- W2963525826 hasLocation W29635258262 @default.
- W2963525826 hasLocation W29635258263 @default.
- W2963525826 hasLocation W29635258264 @default.
- W2963525826 hasOpenAccess W2963525826 @default.
- W2963525826 hasPrimaryLocation W29635258261 @default.
- W2963525826 hasRelatedWork W1483367581 @default.
- W2963525826 hasRelatedWork W1579705726 @default.
- W2963525826 hasRelatedWork W1986021162 @default.
- W2963525826 hasRelatedWork W2081647779 @default.
- W2963525826 hasRelatedWork W2896387979 @default.
- W2963525826 hasRelatedWork W2906407728 @default.
- W2963525826 hasRelatedWork W2946863128 @default.
- W2963525826 hasRelatedWork W2964233404 @default.
- W2963525826 hasRelatedWork W3168410956 @default.
- W2963525826 hasRelatedWork W3185852197 @default.