Research map / Adversarial robustness and security / Backdoor and poisoning
Rethinking CNN’s Generalization to Backdoor Attack from Frequency Domain
Quanrui Rao, Lin Wang, Wuying Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Studies how CNNs memorize and generalize backdoor-poisoned samples through the frequency domain, examining the role of frequency-domain trigger features in backdoor attacks.
Problem
CNNs are vulnerable to backdoor injection, behaving normally on clean inputs but producing attacker-specified outputs on poisoned ones; prior work focused on spatial-domain trigger changes and their effect on generalization.
Approach
Shifts the analysis to the frequency domain, examining how CNNs memorize poisoned samples and generalize to backdoor triggers in frequency space rather than spatial space.
Closest papers by meaning
- A Dual Stealthy Backdoor: From Both Spatial and Frequency Perspectives (AAAI 2024 · similarity 0.90)
- Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift (CVPR 2025 · similarity 0.90)
- A Theoretical Analysis of Backdoor Poisoning Attacks in Convolutional Neural Networks (ICML 2024 · similarity 0.88)
- BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model (NeurIPS 2025 · similarity 0.88)
- Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness (NeurIPS 2024 · similarity 0.88)
- Demystifying Poisoning Backdoor Attacks from a Statistical Perspective (ICLR 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
