English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Zendy Plus

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Zendy Tools

Zendy Open

Multimodal knowledge bases (MMKBs) provide cross-modal aligned knowledgecrucial for multimodal tasks. However, the images in existing MMKBs aregenerally collected for entities in encyclopedia knowledge graphs. Therefore,detailed groundings of visual semantics with linguistic concepts are lacking,which are essential for the visual concept cognition ability of multimodalmodels. Addressing this gap, we introduce M^2ConceptBase, the firstconcept-centric MMKB. M^2ConceptBase models concepts as nodes with associatedimages and detailed textual descriptions. We propose a context-aware multimodalsymbol grounding approach to align concept-image and concept-description pairsusing context information from image-text datasets. Comprising 951K images and152K concepts, M^2ConceptBase links each concept to an average of 6.27 imagesand a single description, ensuring comprehensive visual and textual semantics.Human studies confirm more than 95% alignment accuracy, underscoring itsquality. Additionally, our experiments demonstrate that M^2ConceptBasesignificantly enhances VQA model performance on the OK-VQA task. M^2ConceptBasealso substantially improves the fine-grained concept understanding capabilitiesof multimodal large language models through retrieval augmentation in twoconcept-related tasks, highlighting its value.

M^2ConceptBase: A Fine-Grained Aligned Concept-Centric Multimodal Knowledge Base