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

Neural image classifiers are known to undergo severe performance degradationwhen exposed to inputs that exhibit covariate shifts with respect to thetraining distribution. A general interventional data augmentation(IDA)mechanism that simulates arbitrary interventions over spurious variableshas often been conjectured as a theoretical solution to this problem andapproximated to varying degrees of success. In this work, we study how wellmodern Text-to-Image (T2I) generators and associated image editing techniquescan solve the problem of IDA. We experiment across a diverse collection ofbenchmarks in domain generalization, ablating across key dimensions of T2Igeneration, including interventional prompts, conditioning mechanisms, andpost-hoc filtering, showing that it substantially outperforms previouslystate-of-the-art image augmentation techniques independently of how eachdimension is configured. We discuss the comparative advantages of using T2I forimage editing versus synthesis, also finding that a simple retrieval baselinepresents a surprisingly effective alternative, which raises interestingquestions about how generative models should be evaluated in the context ofdomain generalization.

Not Just Pretty Pictures: Toward Interventional Data Augmentation Using Text-to-Image Generators