FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386076492> ?p ?o ?g. }

• W4386076492 abstract "Although an object may appear in numerous contexts, we often describe it in a limited number of ways. Language allows us to abstract away visual variation to represent and communicate concepts. Building on this intuition, we propose an alternative approach to visual representation learning: using language similarity to sample semantically similar image pairs for contrastive learning. Our approach diverges from image-based contrastive learning by sampling view pairs using language similarity instead of handcrafted augmentations or learned clusters. Our approach also differs from image-text contrastive learning by relying on pre-trained language models to guide the learning rather than directly minimizing a cross-modal loss. Through a series of experiments, we show that language-guided learning yields better features than image-based and image-text representation learning approaches." @default.
• W4386076492 created "2023-08-23" @default.
• W4386076492 creator A5018298417 @default.
• W4386076492 creator A5032488339 @default.
• W4386076492 creator A5063073039 @default.
• W4386076492 date "2023-06-01" @default.
• W4386076492 modified "2023-09-27" @default.
• W4386076492 title "Learning Visual Representations via Language-Guided Sampling" @default.
• W4386076492 cites W1908139891 @default.
• W4386076492 cites W1933349210 @default.
• W4386076492 cites W1956340063 @default.
• W4386076492 cites W1977295328 @default.
• W4386076492 cites W2017814585 @default.
• W4386076492 cites W2025768430 @default.
• W4386076492 cites W2047643928 @default.
• W4386076492 cites W2048679005 @default.
• W4386076492 cites W2051434435 @default.
• W4386076492 cites W2062118960 @default.
• W4386076492 cites W2101105183 @default.
• W4386076492 cites W2108598243 @default.
• W4386076492 cites W2138011018 @default.
• W4386076492 cites W2157364932 @default.
• W4386076492 cites W2166049352 @default.
• W4386076492 cites W2185175083 @default.
• W4386076492 cites W2194775991 @default.
• W4386076492 cites W2250384498 @default.
• W4386076492 cites W2263785794 @default.
• W4386076492 cites W2506483933 @default.
• W4386076492 cites W2533598788 @default.
• W4386076492 cites W2560730294 @default.
• W4386076492 cites W2562153041 @default.
• W4386076492 cites W2798991696 @default.
• W4386076492 cites W2804935296 @default.
• W4386076492 cites W2886641317 @default.
• W4386076492 cites W2962749469 @default.
• W4386076492 cites W2962860144 @default.
• W4386076492 cites W2963115613 @default.
• W4386076492 cites W2963518342 @default.
• W4386076492 cites W2963530300 @default.
• W4386076492 cites W2964194231 @default.
• W4386076492 cites W2964700958 @default.
• W4386076492 cites W2966262636 @default.
• W4386076492 cites W2970641574 @default.
• W4386076492 cites W3035164673 @default.
• W4386076492 cites W3035524453 @default.
• W4386076492 cites W3037492894 @default.
• W4386076492 cites W3103856189 @default.
• W4386076492 cites W3108975329 @default.
• W4386076492 cites W3145450063 @default.
• W4386076492 cites W3153469116 @default.
• W4386076492 cites W3158714121 @default.
• W4386076492 cites W3161438416 @default.
• W4386076492 cites W3166285241 @default.
• W4386076492 cites W3168796319 @default.
• W4386076492 cites W3171007011 @default.
• W4386076492 cites W3172942063 @default.
• W4386076492 cites W3176641147 @default.
• W4386076492 cites W3177096435 @default.
• W4386076492 cites W3182683290 @default.
• W4386076492 cites W3187293622 @default.
• W4386076492 cites W343636949 @default.
• W4386076492 cites W4214481550 @default.
• W4386076492 cites W4214638047 @default.
• W4386076492 cites W4312538130 @default.
• W4386076492 cites W4312980231 @default.
• W4386076492 cites W4313056800 @default.
• W4386076492 cites W4313156423 @default.
• W4386076492 cites W4385572855 @default.
• W4386076492 doi "https://doi.org/10.1109/cvpr52729.2023.01841" @default.
• W4386076492 hasPublicationYear "2023" @default.
• W4386076492 type Work @default.
• W4386076492 citedByCount "0" @default.
• W4386076492 crossrefType "proceedings-article" @default.
• W4386076492 hasAuthorship W4386076492A5018298417 @default.
• W4386076492 hasAuthorship W4386076492A5032488339 @default.
• W4386076492 hasAuthorship W4386076492A5063073039 @default.
• W4386076492 hasConcept C103278499 @default.
• W4386076492 hasConcept C115961682 @default.
• W4386076492 hasConcept C132010649 @default.
• W4386076492 hasConcept C153180895 @default.
• W4386076492 hasConcept C154945302 @default.
• W4386076492 hasConcept C15744967 @default.
• W4386076492 hasConcept C17744445 @default.
• W4386076492 hasConcept C188147891 @default.
• W4386076492 hasConcept C199539241 @default.
• W4386076492 hasConcept C204321447 @default.
• W4386076492 hasConcept C2776359362 @default.
• W4386076492 hasConcept C41008148 @default.
• W4386076492 hasConcept C59404180 @default.
• W4386076492 hasConcept C94625758 @default.
• W4386076492 hasConceptScore W4386076492C103278499 @default.
• W4386076492 hasConceptScore W4386076492C115961682 @default.
• W4386076492 hasConceptScore W4386076492C132010649 @default.
• W4386076492 hasConceptScore W4386076492C153180895 @default.
• W4386076492 hasConceptScore W4386076492C154945302 @default.
• W4386076492 hasConceptScore W4386076492C15744967 @default.
• W4386076492 hasConceptScore W4386076492C17744445 @default.
• W4386076492 hasConceptScore W4386076492C188147891 @default.
• W4386076492 hasConceptScore W4386076492C199539241 @default.
• W4386076492 hasConceptScore W4386076492C204321447 @default.

• next