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

Although recent personalization methods have democratized high-resolutionimage synthesis by enabling swift concept acquisition with minimal examples andlightweight computation, they also present an exploitable avenue for highaccessible backdoor attacks. This paper investigates a critical and unexploredaspect of text-to-image (T2I) diffusion models - their potential vulnerabilityto backdoor attacks via personalization. Our study focuses on a zero-daybackdoor vulnerability prevalent in two families of personalization methods,epitomized by Textual Inversion and DreamBooth.Compared to traditional backdoorattacks, our proposed method can facilitate more precise, efficient, and easilyaccessible attacks with a lower barrier to entry. We provide a comprehensivereview of personalization in T2I diffusion models, highlighting the operationand exploitation potential of this backdoor vulnerability. To be specific, bystudying the prompt processing of Textual Inversion and DreamBooth, we havedevised dedicated backdoor attacks according to the different ways of dealingwith unseen tokens and analyzed the influence of triggers and concept images onthe attack effect. Through comprehensive empirical study, we endorse theutilization of the nouveau-token backdoor attack due to its impressiveeffectiveness, stealthiness, and integrity, markedly outperforming thelegacy-token backdoor attack.

Personalization as a Shortcut for Few-Shot Backdoor Attack against Text-to-Image Diffusion Models