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

In this paper, we propose a robustness benchmark for image-text matchingmodels to assess their vulnerabilities. To this end, we insert adversarialtexts and images into the search pool (i.e., gallery set) and evaluate modelswith the adversarial data. Specifically, we replace a word in the text tochange the meaning of the text and mix images with different images to createperceptible changes in pixels. We assume that such explicit alterations wouldnot deceive a robust model, as they should understand the holistic meaning oftexts and images simultaneously. However, in our evaluations on the proposedbenchmark, many state-of-the-art models show significant performancedegradation, e.g., Recall@1: 81.9% $\rightarrow$ 64.5% in BLIP, 66.1%$\rightarrow$ 37.5% in VSE$\infty$, where the models favor adversarialtexts/images over the original ones. This reveals the current vision-languagemodels may not account for subtle changes or understand the overall context oftexts and images. Our findings can provide insights for improving therobustness of the vision-language models and devising more diverse stress-testmethods in cross-modal retrieval task. Source code and dataset will beavailable at https://github.com/pseulki/rococo.

RoCOCO: Robustness Benchmark of MS-COCO to Stress-test Image-Text Matching Models