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

Backdoor attack against deep neural networks is currently being profoundlyinvestigated due to its severe security consequences. Current state-of-the-artbackdoor attacks require the adversary to modify the input, usually by adding atrigger to it, for the target model to activate the backdoor. This addedtrigger not only increases the difficulty of launching the backdoor attack inthe physical world, but also can be easily detected by multiple defensemechanisms. In this paper, we present the first triggerless backdoor attackagainst deep neural networks, where the adversary does not need to modify theinput for triggering the backdoor. Our attack is based on the dropouttechnique. Concretely, we associate a set of target neurons that are droppedout during model training with the target label. In the prediction phase, themodel will output the target label when the target neurons are dropped again,i.e., the backdoor attack is launched. This triggerless feature of our attackmakes it practical in the physical world. Extensive experiments show that ourtriggerless backdoor attack achieves a perfect attack success rate with anegligible damage to the model's utility.

Don't Trigger Me! A Triggerless Backdoor Attack Against Deep Neural Networks