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

Recent years have witnessed the sustained evolution of misinformation thataims at manipulating public opinions. Unlike traditional rumors or fake newseditors who mainly rely on generated and/or counterfeited images, text andvideos, current misinformation creators now more tend to use out-of-contextmultimedia contents (e.g. mismatched images and captions) to deceive the publicand fake news detection systems. This new type of misinformation increases thedifficulty of not only detection but also clarification, because everyindividual modality is close enough to true information. To address thischallenge, in this paper we explore how to achieve interpretable cross-modalde-contextualization detection that simultaneously identifies the mismatchedpairs and the cross-modal contradictions, which is helpful for fact-checkwebsites to document clarifications. The proposed model first symbolicallydisassembles the text-modality information to a set of fact queries based onthe Abstract Meaning Representation of the caption and then forwards thequery-image pairs into a pre-trained large vision-language model select the``evidences" that are helpful for us to detect misinformation. Extensiveexperiments indicate that the proposed methodology can provide us with muchmore interpretable predictions while maintaining the accuracy same as thestate-of-the-art model on this task.

Interpretable Detection of Out-of-Context Misinformation with Neural-Symbolic-Enhanced Large Multimodal Model