Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950401034> ?p ?o ?g. }
- W2950401034 abstract "We introduce an inference technique to produce discriminative context-aware image captions (captions that describe differences between images or visual concepts) using only generic context-agnostic training data (captions that describe a concept or an image in isolation). For example, given images and captions of siamese cat and tiger cat, we generate language that describes the siamese cat in a way that distinguishes it from tiger cat. Our key novelty is that we show how to do joint inference over a language model that is context-agnostic and a listener which distinguishes closely-related concepts. We first apply our technique to a justification task, namely to describe why an image contains a particular fine-grained category as opposed to another closely-related category of the CUB-200-2011 dataset. We then study discriminative image captioning to generate language that uniquely refers to one of two semantically-similar images in the COCO dataset. Evaluations with discriminative ground truth for justification and human studies for discriminative image captioning reveal that our approach outperforms baseline generative and speaker-listener approaches for discrimination." @default.
- W2950401034 created "2019-06-27" @default.
- W2950401034 creator A5002713363 @default.
- W2950401034 creator A5017529415 @default.
- W2950401034 creator A5041797310 @default.
- W2950401034 creator A5045719865 @default.
- W2950401034 creator A5050342343 @default.
- W2950401034 date "2017-01-11" @default.
- W2950401034 modified "2023-10-16" @default.
- W2950401034 title "Context-aware Captions from Context-agnostic Supervision" @default.
- W2950401034 cites W1514535095 @default.
- W2950401034 cites W1686810756 @default.
- W2950401034 cites W1706899115 @default.
- W2950401034 cites W1797268635 @default.
- W2950401034 cites W1889624880 @default.
- W2950401034 cites W1958706068 @default.
- W2950401034 cites W1977195311 @default.
- W2950401034 cites W1983927101 @default.
- W2950401034 cites W1996418862 @default.
- W2950401034 cites W2064675550 @default.
- W2950401034 cites W2112055291 @default.
- W2950401034 cites W2120176017 @default.
- W2950401034 cites W2123301721 @default.
- W2950401034 cites W2133564696 @default.
- W2950401034 cites W2136462581 @default.
- W2950401034 cites W2144960104 @default.
- W2950401034 cites W2144968237 @default.
- W2950401034 cites W2202226326 @default.
- W2950401034 cites W2251512949 @default.
- W2950401034 cites W2252033417 @default.
- W2950401034 cites W2261271299 @default.
- W2950401034 cites W2264742718 @default.
- W2950401034 cites W2293700449 @default.
- W2950401034 cites W2337353209 @default.
- W2950401034 cites W2505639562 @default.
- W2950401034 cites W2506483933 @default.
- W2950401034 cites W2602275733 @default.
- W2950401034 cites W2949467366 @default.
- W2950401034 cites W2949555952 @default.
- W2950401034 cites W2949769367 @default.
- W2950401034 cites W2950472486 @default.
- W2950401034 cites W2950761309 @default.
- W2950401034 cites W2951183276 @default.
- W2950401034 cites W2951619830 @default.
- W2950401034 cites W2951805548 @default.
- W2950401034 cites W2951912364 @default.
- W2950401034 cites W2952574180 @default.
- W2950401034 cites W2953119472 @default.
- W2950401034 cites W2964193163 @default.
- W2950401034 doi "https://doi.org/10.48550/arxiv.1701.02870" @default.
- W2950401034 hasPublicationYear "2017" @default.
- W2950401034 type Work @default.
- W2950401034 sameAs 2950401034 @default.
- W2950401034 citedByCount "22" @default.
- W2950401034 countsByYear W29504010342017 @default.
- W2950401034 countsByYear W29504010342018 @default.
- W2950401034 countsByYear W29504010342019 @default.
- W2950401034 countsByYear W29504010342020 @default.
- W2950401034 countsByYear W29504010342021 @default.
- W2950401034 crossrefType "posted-content" @default.
- W2950401034 hasAuthorship W2950401034A5002713363 @default.
- W2950401034 hasAuthorship W2950401034A5017529415 @default.
- W2950401034 hasAuthorship W2950401034A5041797310 @default.
- W2950401034 hasAuthorship W2950401034A5045719865 @default.
- W2950401034 hasAuthorship W2950401034A5050342343 @default.
- W2950401034 hasBestOaLocation W29504010341 @default.
- W2950401034 hasConcept C115961682 @default.
- W2950401034 hasConcept C151730666 @default.
- W2950401034 hasConcept C153180895 @default.
- W2950401034 hasConcept C154945302 @default.
- W2950401034 hasConcept C15744967 @default.
- W2950401034 hasConcept C157657479 @default.
- W2950401034 hasConcept C204321447 @default.
- W2950401034 hasConcept C26517878 @default.
- W2950401034 hasConcept C2776214188 @default.
- W2950401034 hasConcept C2778738651 @default.
- W2950401034 hasConcept C2779343474 @default.
- W2950401034 hasConcept C38652104 @default.
- W2950401034 hasConcept C39890363 @default.
- W2950401034 hasConcept C41008148 @default.
- W2950401034 hasConcept C77805123 @default.
- W2950401034 hasConcept C86803240 @default.
- W2950401034 hasConcept C97931131 @default.
- W2950401034 hasConceptScore W2950401034C115961682 @default.
- W2950401034 hasConceptScore W2950401034C151730666 @default.
- W2950401034 hasConceptScore W2950401034C153180895 @default.
- W2950401034 hasConceptScore W2950401034C154945302 @default.
- W2950401034 hasConceptScore W2950401034C15744967 @default.
- W2950401034 hasConceptScore W2950401034C157657479 @default.
- W2950401034 hasConceptScore W2950401034C204321447 @default.
- W2950401034 hasConceptScore W2950401034C26517878 @default.
- W2950401034 hasConceptScore W2950401034C2776214188 @default.
- W2950401034 hasConceptScore W2950401034C2778738651 @default.
- W2950401034 hasConceptScore W2950401034C2779343474 @default.
- W2950401034 hasConceptScore W2950401034C38652104 @default.
- W2950401034 hasConceptScore W2950401034C39890363 @default.
- W2950401034 hasConceptScore W2950401034C41008148 @default.
- W2950401034 hasConceptScore W2950401034C77805123 @default.
- W2950401034 hasConceptScore W2950401034C86803240 @default.
- W2950401034 hasConceptScore W2950401034C97931131 @default.