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

Artificial neural networks typically struggle in generalizing toout-of-context examples. One reason for this limitation is caused by havingdatasets that incorporate only partial information regarding the potentialcorrelational structure of the world. In this work, we propose TIDA (TargetedImage-editing Data Augmentation), a targeted data augmentation method focusedon improving models' human-like abilities (e.g., gender recognition) by fillingthe correlational structure gap using a text-to-image generative model. Morespecifically, TIDA identifies specific skills in captions describing images(e.g., the presence of a specific gender in the image), changes the caption(e.g., "woman" to "man"), and then uses a text-to-image model to edit the imagein order to match the novel caption (e.g., uniquely changing a woman to a manwhile maintaining the context identical). Based on the Flickr30K benchmark, weshow that, compared with the original data set, a TIDA-enhanced dataset relatedto gender, color, and counting abilities induces better performance in severalimage captioning metrics. Furthermore, on top of relying on the classical BLEUmetric, we conduct a fine-grained analysis of the improvements of our modelsagainst the baseline in different ways. We compared text-to-image generativemodels and found different behaviors of the image captioning models in terms ofencoding visual encoding and textual decoding.

Targeted Image Data Augmentation Increases Basic Skills Captioning Robustness