Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313013044> ?p ?o ?g. }
- W4313013044 endingPage "3228" @default.
- W4313013044 startingPage "3214" @default.
- W4313013044 abstract "Textbook Question Answering (TQA) task requires answering questions by reasoning based on both the given diagrams and text context. There are mainly two challenges for the task. First, the diagrams are different from the natural images. Similar shapes or color blocks may express different semantics and there is also a large intra-topic variation for diagrams. Hence, the characteristics of visual semantic ambiguity and variable visual appearance make the diagram understanding more challenging. Second, for the text, the specific education domain with terminologies exists a great gap with the general domain. Therefore, it is difficult to represent the text semantics effectively using a text encoder pretrained in the general domain. In this paper, we propose a Spatial-Semantic Collaborative Graph Network (SSCGN) for TQA task, which can help enhance the diagram and text understanding and facilitate multimodal reasoning. Specifically, the Spatial-guided Semantic Enhancing (SSE) module fully exploits the spatial and semantic relationships between visual objects and OCR tokens to collaboratively enhance the diagram semantic understanding. Moreover, based on the semantically enhanced region representations of the SSE module, the Fine-grained Spatial-Aware Graph Network (FSA-GN) can help obtain richer relation-aware region representations for joint reasoning by capturing more fine-grained spatial relationships. We further propose multiple self-supervised auxiliary tasks to enhance the initial diagram and text semantic representations by pretraining the diagram encoder and text encoder. Extensive experiments and ablation studies are conducted to validate the effectiveness of SSCGN." @default.
- W4313013044 created "2023-01-05" @default.
- W4313013044 creator A5000446732 @default.
- W4313013044 creator A5020447579 @default.
- W4313013044 creator A5045145844 @default.
- W4313013044 creator A5070094915 @default.
- W4313013044 creator A5086407377 @default.
- W4313013044 creator A5088544875 @default.
- W4313013044 date "2023-07-01" @default.
- W4313013044 modified "2023-10-17" @default.
- W4313013044 title "Spatial-Semantic Collaborative Graph Network for Textbook Question Answering" @default.
- W4313013044 cites W1933349210 @default.
- W4313013044 cites W1983419592 @default.
- W4313013044 cites W2121406004 @default.
- W4313013044 cites W2194775991 @default.
- W4313013044 cites W2250564385 @default.
- W4313013044 cites W2307512708 @default.
- W4313013044 cites W2745461083 @default.
- W4313013044 cites W2746097825 @default.
- W4313013044 cites W2799088654 @default.
- W4313013044 cites W2897899251 @default.
- W4313013044 cites W2949431215 @default.
- W4313013044 cites W2950382646 @default.
- W4313013044 cites W2952938873 @default.
- W4313013044 cites W2958360136 @default.
- W4313013044 cites W2962959437 @default.
- W4313013044 cites W2963420691 @default.
- W4313013044 cites W2964042428 @default.
- W4313013044 cites W2981165461 @default.
- W4313013044 cites W3025102114 @default.
- W4313013044 cites W3034999214 @default.
- W4313013044 cites W3035106315 @default.
- W4313013044 cites W3094172275 @default.
- W4313013044 cites W3095789240 @default.
- W4313013044 cites W3106278732 @default.
- W4313013044 cites W3114632476 @default.
- W4313013044 cites W3131251978 @default.
- W4313013044 cites W3167387364 @default.
- W4313013044 cites W3176186248 @default.
- W4313013044 cites W3199573137 @default.
- W4313013044 cites W3201541489 @default.
- W4313013044 cites W343636949 @default.
- W4313013044 cites W4206337866 @default.
- W4313013044 cites W4221138453 @default.
- W4313013044 cites W4282936652 @default.
- W4313013044 cites W4304015009 @default.
- W4313013044 doi "https://doi.org/10.1109/tcsvt.2022.3231463" @default.
- W4313013044 hasPublicationYear "2023" @default.
- W4313013044 type Work @default.
- W4313013044 citedByCount "1" @default.
- W4313013044 countsByYear W43130130442023 @default.
- W4313013044 crossrefType "journal-article" @default.
- W4313013044 hasAuthorship W4313013044A5000446732 @default.
- W4313013044 hasAuthorship W4313013044A5020447579 @default.
- W4313013044 hasAuthorship W4313013044A5045145844 @default.
- W4313013044 hasAuthorship W4313013044A5070094915 @default.
- W4313013044 hasAuthorship W4313013044A5086407377 @default.
- W4313013044 hasAuthorship W4313013044A5088544875 @default.
- W4313013044 hasConcept C132525143 @default.
- W4313013044 hasConcept C134306372 @default.
- W4313013044 hasConcept C151730666 @default.
- W4313013044 hasConcept C154945302 @default.
- W4313013044 hasConcept C184337299 @default.
- W4313013044 hasConcept C199360897 @default.
- W4313013044 hasConcept C204321447 @default.
- W4313013044 hasConcept C23123220 @default.
- W4313013044 hasConcept C27511587 @default.
- W4313013044 hasConcept C2779343474 @default.
- W4313013044 hasConcept C33923547 @default.
- W4313013044 hasConcept C36503486 @default.
- W4313013044 hasConcept C41008148 @default.
- W4313013044 hasConcept C44291984 @default.
- W4313013044 hasConcept C80444323 @default.
- W4313013044 hasConcept C86803240 @default.
- W4313013044 hasConceptScore W4313013044C132525143 @default.
- W4313013044 hasConceptScore W4313013044C134306372 @default.
- W4313013044 hasConceptScore W4313013044C151730666 @default.
- W4313013044 hasConceptScore W4313013044C154945302 @default.
- W4313013044 hasConceptScore W4313013044C184337299 @default.
- W4313013044 hasConceptScore W4313013044C199360897 @default.
- W4313013044 hasConceptScore W4313013044C204321447 @default.
- W4313013044 hasConceptScore W4313013044C23123220 @default.
- W4313013044 hasConceptScore W4313013044C27511587 @default.
- W4313013044 hasConceptScore W4313013044C2779343474 @default.
- W4313013044 hasConceptScore W4313013044C33923547 @default.
- W4313013044 hasConceptScore W4313013044C36503486 @default.
- W4313013044 hasConceptScore W4313013044C41008148 @default.
- W4313013044 hasConceptScore W4313013044C44291984 @default.
- W4313013044 hasConceptScore W4313013044C80444323 @default.
- W4313013044 hasConceptScore W4313013044C86803240 @default.
- W4313013044 hasFunder F4320321001 @default.
- W4313013044 hasFunder F4320327609 @default.
- W4313013044 hasFunder F4320330193 @default.
- W4313013044 hasFunder F4320335777 @default.
- W4313013044 hasFunder F4320336567 @default.
- W4313013044 hasIssue "7" @default.
- W4313013044 hasLocation W43130130441 @default.
- W4313013044 hasOpenAccess W4313013044 @default.