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/

Global: Zendy Tools annual

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

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

In this paper, we make a simple observation that questions about images oftencontain premises - objects and relationships implied by the question - and thatreasoning about premises can help Visual Question Answering (VQA) modelsrespond more intelligently to irrelevant or previously unseen questions. Whenpresented with a question that is irrelevant to an image, state-of-the-art VQAmodels will still answer purely based on learned language biases, resulting innon-sensical or even misleading answers. We note that a visual question isirrelevant to an image if at least one of its premises is false (i.e. notdepicted in the image). We leverage this observation to construct a dataset forQuestion Relevance Prediction and Explanation (QRPE) by searching for falsepremises. We train novel question relevance detection models and show thatmodels that reason about premises consistently outperform models that do not.We also find that forcing standard VQA models to reason about premises duringtraining can lead to improvements on tasks requiring compositional reasoning.

The Promise of Premise: Harnessing Question Premises in Visual Question Answering