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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Vision-language pre-training (VLP) on large-scale datasets has shown premierperformance on various downstream tasks. In contrast to plenty of availablebenchmarks with English corpus, large-scale pre-training datasets anddownstream datasets with Chinese corpus remain largely unexplored. In thiswork, we build a large-scale high-quality Chinese Cross-Modal Benchmark namedCCMB for the research community, which contains the currently largest publicpre-training dataset Zero and five human-annotated fine-tuning datasets fordownstream tasks. Zero contains 250 million images paired with 750 million textdescriptions, plus two of the five fine-tuning datasets are also currently thelargest ones for Chinese cross-modal downstream tasks. Along with the CCMB, wealso develop a VLP framework named R2D2, applying a pre-Ranking + Rankingstrategy to learn powerful vision-language representations and a two-waydistillation method (i.e., target-guided Distillation and feature-guidedDistillation) to further enhance the learning capability. With the Zero and theR2D2 VLP framework, we achieve state-of-the-art performance on twelvedownstream datasets from five broad categories of tasks including image-textretrieval, image-text matching, image caption, text-to-image generation, andzero-shot image classification. The datasets, models, and codes are availableat https://github.com/yuxie11/R2D2

CCMB: A Large-scale Chinese Cross-modal Benchmark