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

Deep learning models are increasingly used in mobile applications as criticalcomponents. Unlike the program bytecode whose vulnerabilities and threats havebeen widely-discussed, whether and how the deep learning models deployed in theapplications can be compromised are not well-understood since neural networksare usually viewed as a black box. In this paper, we introduce a highlypractical backdoor attack achieved with a set of reverse-engineering techniquesover compiled deep learning models. The core of the attack is a neuralconditional branch constructed with a trigger detector and several operatorsand injected into the victim model as a malicious payload. The attack iseffective as the conditional logic can be flexibly customized by the attacker,and scalable as it does not require any prior knowledge from the originalmodel. We evaluated the attack effectiveness using 5 state-of-the-art deeplearning models and real-world samples collected from 30 users. The resultsdemonstrated that the injected backdoor can be triggered with a success rate of93.5%, while only brought less than 2ms latency overhead and no more than 1.4%accuracy decrease. We further conducted an empirical study on real-world mobiledeep learning apps collected from Google Play. We found 54 apps that werevulnerable to our attack, including popular and security-critical ones. Theresults call for the awareness of deep learning application developers andauditors to enhance the protection of deployed models.

DeepPayload: Black-box Backdoor Attack on Deep Learning Models through Neural Payload Injection