نسخة أولية وصول مفتوح
TARE: Weigh a Never-Poisoned Twin Before Reading Backdoor-Defense Costs
Backdoor-defense leaderboards print a clean-accuracy drop and read it as removal cost. Measured on the poisoned victim alone, the drop cannot separate removal from what the defense does to any model, and inherits the victim's start, which for three of BackdoorBench's sixteen attacks is a configuration file: WaNet, BPP …