Simin Tang1,2,3,4, Zhiyong Zhang1,2,3,4,*, Junyan Pan1,2,3,4, Gaoyuan Quan1,2,3,4, Weiguo Wang5, Junchang Jing6
CMC-Computers, Materials & Continua, DOI:10.32604/cmc.2025.073798
Abstract At inference time, deep neural networks are susceptible to backdoor attacks, which can produce attacker-controlled outputs when inputs contain carefully crafted triggers. Existing defense methods often focus on specific attack types or incur high costs, such as data cleaning or model fine-tuning. In contrast, we argue that it is possible to achieve effective and generalizable defense without removing triggers or incurring high model-cleaning costs. From the attacker’s perspective and based on characteristics of vulnerable neuron activation anomalies, we propose an Adaptive Feature Injection (AFI) method for black-box backdoor detection. AFI employs a pre-trained image encoder… More >