all InfoSec news
Imperio: Language-Guided Backdoor Attacks for Arbitrary Model Control
March 18, 2024, 4:11 a.m. | Ka-Ho Chow, Wenqi Wei, Lei Yu
cs.CR updates on arXiv.org arxiv.org
Abstract: Natural language processing (NLP) has received unprecedented attention. While advancements in NLP models have led to extensive research into their backdoor vulnerabilities, the potential for these advancements to introduce new backdoor threats remains unexplored. This paper proposes Imperio, which harnesses the language understanding capabilities of NLP models to enrich backdoor attacks. Imperio provides a new model control experience. Demonstrated through controlling image classifiers, it empowers the adversary to manipulate the victim model with arbitrary output …
arxiv attacks attention backdoor backdoor attacks capabilities control cs.cr cs.lg language led natural natural language natural language processing nlp research threats understanding unprecedented vulnerabilities
More from arxiv.org / cs.CR updates on arXiv.org
Jobs in InfoSec / Cybersecurity
CyberSOC Technical Lead
@ Integrity360 | Sandyford, Dublin, Ireland
Cyber Security Strategy Consultant
@ Capco | New York City
Cyber Security Senior Consultant
@ Capco | Chicago, IL
Sr. Product Manager
@ MixMode | Remote, US
Corporate Intern - Information Security (Year Round)
@ Associated Bank | US WI Remote
Senior Offensive Security Engineer
@ CoStar Group | US-DC Washington, DC