Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378498666> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W4378498666 abstract "Weakly supervised vision-and-language pre-training (WVLP), which learns cross-modal representations with limited cross-modal supervision, has been shown to effectively reduce the data cost of pre-training while maintaining decent performance on downstream tasks. However, current WVLP methods use only local descriptions of images, i.e., object tags, as cross-modal anchors to construct weakly-aligned image-text pairs for pre-training. This affects the data quality and thus the effectiveness of pre-training. In this paper, we propose to directly take a small number of aligned image-text pairs as anchors, and represent each unaligned image and text by its similarities to these anchors, i.e., relative representations. We build a WVLP framework based on the relative representations, namely RELIT, which collects high-quality weakly-aligned image-text pairs from large-scale image-only and text-only data for pre-training through relative representation-based retrieval and generation. Experiments on four downstream tasks show that RELIT achieves new state-of-the-art results under the weakly supervised setting." @default.
- W4378498666 created "2023-05-27" @default.
- W4378498666 creator A5023363049 @default.
- W4378498666 creator A5046448314 @default.
- W4378498666 creator A5071458554 @default.
- W4378498666 creator A5073190439 @default.
- W4378498666 date "2023-05-24" @default.
- W4378498666 modified "2023-09-25" @default.
- W4378498666 title "Weakly Supervised Vision-and-Language Pre-training with Relative Representations" @default.
- W4378498666 doi "https://doi.org/10.48550/arxiv.2305.15483" @default.
- W4378498666 hasPublicationYear "2023" @default.
- W4378498666 type Work @default.
- W4378498666 citedByCount "0" @default.
- W4378498666 crossrefType "posted-content" @default.
- W4378498666 hasAuthorship W4378498666A5023363049 @default.
- W4378498666 hasAuthorship W4378498666A5046448314 @default.
- W4378498666 hasAuthorship W4378498666A5071458554 @default.
- W4378498666 hasAuthorship W4378498666A5073190439 @default.
- W4378498666 hasBestOaLocation W43784986661 @default.
- W4378498666 hasConcept C111472728 @default.
- W4378498666 hasConcept C115961682 @default.
- W4378498666 hasConcept C121332964 @default.
- W4378498666 hasConcept C138885662 @default.
- W4378498666 hasConcept C153180895 @default.
- W4378498666 hasConcept C153294291 @default.
- W4378498666 hasConcept C154945302 @default.
- W4378498666 hasConcept C169590947 @default.
- W4378498666 hasConcept C17744445 @default.
- W4378498666 hasConcept C185592680 @default.
- W4378498666 hasConcept C188027245 @default.
- W4378498666 hasConcept C199360897 @default.
- W4378498666 hasConcept C199539241 @default.
- W4378498666 hasConcept C204321447 @default.
- W4378498666 hasConcept C2776359362 @default.
- W4378498666 hasConcept C2777211547 @default.
- W4378498666 hasConcept C2778755073 @default.
- W4378498666 hasConcept C2779530757 @default.
- W4378498666 hasConcept C2780801425 @default.
- W4378498666 hasConcept C2781238097 @default.
- W4378498666 hasConcept C41008148 @default.
- W4378498666 hasConcept C51632099 @default.
- W4378498666 hasConcept C62520636 @default.
- W4378498666 hasConcept C71139939 @default.
- W4378498666 hasConcept C77660490 @default.
- W4378498666 hasConcept C94625758 @default.
- W4378498666 hasConceptScore W4378498666C111472728 @default.
- W4378498666 hasConceptScore W4378498666C115961682 @default.
- W4378498666 hasConceptScore W4378498666C121332964 @default.
- W4378498666 hasConceptScore W4378498666C138885662 @default.
- W4378498666 hasConceptScore W4378498666C153180895 @default.
- W4378498666 hasConceptScore W4378498666C153294291 @default.
- W4378498666 hasConceptScore W4378498666C154945302 @default.
- W4378498666 hasConceptScore W4378498666C169590947 @default.
- W4378498666 hasConceptScore W4378498666C17744445 @default.
- W4378498666 hasConceptScore W4378498666C185592680 @default.
- W4378498666 hasConceptScore W4378498666C188027245 @default.
- W4378498666 hasConceptScore W4378498666C199360897 @default.
- W4378498666 hasConceptScore W4378498666C199539241 @default.
- W4378498666 hasConceptScore W4378498666C204321447 @default.
- W4378498666 hasConceptScore W4378498666C2776359362 @default.
- W4378498666 hasConceptScore W4378498666C2777211547 @default.
- W4378498666 hasConceptScore W4378498666C2778755073 @default.
- W4378498666 hasConceptScore W4378498666C2779530757 @default.
- W4378498666 hasConceptScore W4378498666C2780801425 @default.
- W4378498666 hasConceptScore W4378498666C2781238097 @default.
- W4378498666 hasConceptScore W4378498666C41008148 @default.
- W4378498666 hasConceptScore W4378498666C51632099 @default.
- W4378498666 hasConceptScore W4378498666C62520636 @default.
- W4378498666 hasConceptScore W4378498666C71139939 @default.
- W4378498666 hasConceptScore W4378498666C77660490 @default.
- W4378498666 hasConceptScore W4378498666C94625758 @default.
- W4378498666 hasLocation W43784986661 @default.
- W4378498666 hasOpenAccess W4378498666 @default.
- W4378498666 hasPrimaryLocation W43784986661 @default.
- W4378498666 hasRelatedWork W1549289070 @default.
- W4378498666 hasRelatedWork W2092957489 @default.
- W4378498666 hasRelatedWork W2133574959 @default.
- W4378498666 hasRelatedWork W2536452361 @default.
- W4378498666 hasRelatedWork W3126602623 @default.
- W4378498666 hasRelatedWork W3153415088 @default.
- W4378498666 hasRelatedWork W3177008965 @default.
- W4378498666 hasRelatedWork W88325386 @default.
- W4378498666 hasRelatedWork W2178972535 @default.
- W4378498666 hasRelatedWork W2529086386 @default.
- W4378498666 isParatext "false" @default.
- W4378498666 isRetracted "false" @default.
- W4378498666 workType "article" @default.