Matches in SemOpenAlex for { <https://semopenalex.org/work/W4306891337> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4306891337 abstract "Multimodal pre-training breaks down the modality barriers and allows the individual modalities to be mutually augmented with information, resulting in significant advances in representation learning. However, graph modality, as a very general and important form of data, cannot be easily interacted with other modalities because of its non-regular nature. In this paper, we propose MMGA (Multimodal learning with Graph Alignment), a novel multimodal pre-training framework to incorporate information from graph (social network), image and text modalities on social media to enhance user representation learning. In MMGA, a multi-step graph alignment mechanism is proposed to add the self-supervision from graph modality to optimize the image and text encoders, while using the information from the image and text modalities to guide the graph encoder learning. We conduct experiments on the dataset crawled from Instagram. The experimental results show that MMGA works well on the dataset and improves the fans prediction task's performance. We release our dataset, the first social media multimodal dataset with graph, of 60,000 users labeled with specific topics based on 2 million posts to facilitate future research." @default.
- W4306891337 created "2022-10-21" @default.
- W4306891337 creator A5045543861 @default.
- W4306891337 creator A5049692788 @default.
- W4306891337 date "2022-10-18" @default.
- W4306891337 modified "2023-09-27" @default.
- W4306891337 title "MMGA: Multimodal Learning with Graph Alignment" @default.
- W4306891337 doi "https://doi.org/10.48550/arxiv.2210.09946" @default.
- W4306891337 hasPublicationYear "2022" @default.
- W4306891337 type Work @default.
- W4306891337 citedByCount "0" @default.
- W4306891337 crossrefType "posted-content" @default.
- W4306891337 hasAuthorship W4306891337A5045543861 @default.
- W4306891337 hasAuthorship W4306891337A5049692788 @default.
- W4306891337 hasBestOaLocation W43068913371 @default.
- W4306891337 hasConcept C111919701 @default.
- W4306891337 hasConcept C118505674 @default.
- W4306891337 hasConcept C119857082 @default.
- W4306891337 hasConcept C132525143 @default.
- W4306891337 hasConcept C136764020 @default.
- W4306891337 hasConcept C144024400 @default.
- W4306891337 hasConcept C154945302 @default.
- W4306891337 hasConcept C17744445 @default.
- W4306891337 hasConcept C199539241 @default.
- W4306891337 hasConcept C204321447 @default.
- W4306891337 hasConcept C23123220 @default.
- W4306891337 hasConcept C2776359362 @default.
- W4306891337 hasConcept C2779903281 @default.
- W4306891337 hasConcept C2780226545 @default.
- W4306891337 hasConcept C2780660688 @default.
- W4306891337 hasConcept C36289849 @default.
- W4306891337 hasConcept C41008148 @default.
- W4306891337 hasConcept C518677369 @default.
- W4306891337 hasConcept C59404180 @default.
- W4306891337 hasConcept C80444323 @default.
- W4306891337 hasConcept C94625758 @default.
- W4306891337 hasConceptScore W4306891337C111919701 @default.
- W4306891337 hasConceptScore W4306891337C118505674 @default.
- W4306891337 hasConceptScore W4306891337C119857082 @default.
- W4306891337 hasConceptScore W4306891337C132525143 @default.
- W4306891337 hasConceptScore W4306891337C136764020 @default.
- W4306891337 hasConceptScore W4306891337C144024400 @default.
- W4306891337 hasConceptScore W4306891337C154945302 @default.
- W4306891337 hasConceptScore W4306891337C17744445 @default.
- W4306891337 hasConceptScore W4306891337C199539241 @default.
- W4306891337 hasConceptScore W4306891337C204321447 @default.
- W4306891337 hasConceptScore W4306891337C23123220 @default.
- W4306891337 hasConceptScore W4306891337C2776359362 @default.
- W4306891337 hasConceptScore W4306891337C2779903281 @default.
- W4306891337 hasConceptScore W4306891337C2780226545 @default.
- W4306891337 hasConceptScore W4306891337C2780660688 @default.
- W4306891337 hasConceptScore W4306891337C36289849 @default.
- W4306891337 hasConceptScore W4306891337C41008148 @default.
- W4306891337 hasConceptScore W4306891337C518677369 @default.
- W4306891337 hasConceptScore W4306891337C59404180 @default.
- W4306891337 hasConceptScore W4306891337C80444323 @default.
- W4306891337 hasConceptScore W4306891337C94625758 @default.
- W4306891337 hasLocation W43068913371 @default.
- W4306891337 hasLocation W43068913372 @default.
- W4306891337 hasOpenAccess W4306891337 @default.
- W4306891337 hasPrimaryLocation W43068913371 @default.
- W4306891337 hasRelatedWork W2470996195 @default.
- W4306891337 hasRelatedWork W2474574787 @default.
- W4306891337 hasRelatedWork W2952745240 @default.
- W4306891337 hasRelatedWork W2959445501 @default.
- W4306891337 hasRelatedWork W3118761370 @default.
- W4306891337 hasRelatedWork W3199271201 @default.
- W4306891337 hasRelatedWork W4283811427 @default.
- W4306891337 hasRelatedWork W4301143707 @default.
- W4306891337 hasRelatedWork W4306891337 @default.
- W4306891337 hasRelatedWork W4312851439 @default.
- W4306891337 isParatext "false" @default.
- W4306891337 isRetracted "false" @default.
- W4306891337 workType "article" @default.