all InfoSec news
Breaking Speaker Recognition with PaddingBack
March 12, 2024, 4:11 a.m. | Zhe Ye, Diqun Yan, Li Dong, Kailai Shen
cs.CR updates on arXiv.org arxiv.org
Abstract: Machine Learning as a Service (MLaaS) has gained popularity due to advancements in Deep Neural Networks (DNNs). However, untrusted third-party platforms have raised concerns about AI security, particularly in backdoor attacks. Recent research has shown that speech backdoors can utilize transformations as triggers, similar to image backdoors. However, human ears can easily be aware of these transformations, leading to suspicion. In this paper, we propose PaddingBack, an inaudible backdoor attack that utilizes malicious operations to …
ai security arxiv attacks backdoor backdoor attacks backdoors breaking can cs.cr cs.sd eess.as eess.sp human image machine machine learning networks neural networks party platforms recognition research security service speaker speaker recognition speech third third-party untrusted
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
Principal Engineer - DLP Endpoint Security
@ Netskope | Bengaluru, Karnataka, India
Security Consultant (m/w/d)
@ Deutsche Telekom | Berlin, Deutschland
Security Engineer
@ IDEMIA | Haarlem, NL, 2031 CC
CyberSecurity Forensics and Incident Response Analyst
@ Bosch Group | Pittsburgh, PA, United States
Cyber MS MDR - Sr Associate
@ KPMG India | Bengaluru, Karnataka, India
Senior Lead Cybersecurity Architect-Threat modeling, Cryptography
@ JPMorgan Chase & Co. | India