all InfoSec news
Cutting through buggy adversarial example defenses: fixing 1 line of code breaks Sabre
May 7, 2024, 4:11 a.m. | Nicholas Carlini
cs.CR updates on arXiv.org arxiv.org
Abstract: Sabre is a defense to adversarial examples that was accepted at IEEE S&P 2024. We first reveal significant flaws in the evaluation that point to clear signs of gradient masking. We then show the cause of this gradient masking: a bug in the original evaluation code. By fixing a single line of code in the original repository, we reduce Sabre's robust accuracy to 0%. In response to this, the authors modify the defense and introduce …
adversarial amp arxiv bug clear code cs.cr cs.lg cutting defense defenses evaluation examples flaws ieee line masking point reveal
More from arxiv.org / cs.CR updates on arXiv.org
A Privacy Preserving System for Movie Recommendations Using Federated Learning
2 days, 12 hours ago |
arxiv.org
Jobs in InfoSec / Cybersecurity
Information Security Engineers
@ D. E. Shaw Research | New York City
Technology Security Analyst
@ Halton Region | Oakville, Ontario, Canada
Senior Cyber Security Analyst
@ Valley Water | San Jose, CA
Senior Security Analyst
@ Oracle | United States
Associate Vulnerability Management Specialist
@ Diebold Nixdorf | Hyderabad, Telangana, India
Cybersecurity Architect, Infrastructure & Technical Security
@ KCB Group | Kenya