Skip to content
arXiv cs.CV · Papers

Corrupting Attention: Evasion-Based Adversarial Attacks on Encoder Attention in Detection Transformers

arXiv:2608.06674v1 Announce Type: new Abstract: Adversarial vulnerabilities remain a major concern for the safe deployment of neural networks, particularly in object detection, a core task embedded in many safety-critical systems. Detection transformers have emerged as leading object detectors, yet their adversarial ro