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 Plus annual

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

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We introduce VisoGender, a novel dataset for benchmarking gender bias invision-language models. We focus on occupation-related biases within ahegemonic system of binary gender, inspired by Winograd and Winogender schemas,where each image is associated with a caption containing a pronoun relationshipof subjects and objects in the scene. VisoGender is balanced by genderrepresentation in professional roles, supporting bias evaluation in two ways:i) resolution bias, where we evaluate the difference between pronoun resolutionaccuracies for image subjects with gender presentations perceived as masculineversus feminine by human annotators and ii) retrieval bias, where we compareratios of professionals perceived to have masculine and feminine genderpresentations retrieved for a gender-neutral search query. We benchmark severalstate-of-the-art vision-language models and find that they demonstrate bias inresolving binary gender in complex scenes. While the direction and magnitude ofgender bias depends on the task and the model being evaluated, captioningmodels are generally less biased than Vision-Language Encoders. Dataset andcode are available at https://github.com/oxai/visogender

VisoGender: A dataset for benchmarking gender bias in image-text pronoun resolution