method · first seen 31 Oct 2026
adversarial perturbation attack
Attack that adds crafted small input or activation perturbations to make a model behave incorrectly while evading detection.
Mentions per week
PapersCodeDiscussion
Earliest sources
- The Blind Spots of Detection: Adversarial Perturbation Attacks against Watchdogs in Distributed LLM InferencePapers · 31 Oct 2026 · Siu Wai Shi, Xiaochen Zhu, Xinjian Luo, Keke Huang, Bingxue Zhang, Xiaofeng Gao
Latest sources
- The Blind Spots of Detection: Adversarial Perturbation Attacks against Watchdogs in Distributed LLM InferencePapers · 31 Oct 2026 · Siu Wai Shi, Xiaochen Zhu, Xinjian Luo, Keke Huang, Bingxue Zhang, Xiaofeng Gao