AI Paper Atlas

Research map / Adversarial robustness and security / Backdoor and poisoning

BackdoorBench: A Comprehensive Benchmark of Backdoor Learning

Baoyuan Wu, Hongrui Chen, Mingda Zhang, Zihao Zhu, Shaokui Wei, Danni Yuan, Chao Shen

NeurIPS 2022 · Adversarial robustness and security › Backdoor and poisoning · approach: Trigger-based backdoor attacks · 41 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
raise poisoning ratio from 5% to 10%backdoor attacks vs FT/FP/NAD/NC/ABL/ANP defenses · CIFAR-10, PreAct-ResNet18attack success rateMixed
higher poisoning ratio makes ANP prune more neurons (SIG attack)ANP defense · CIFAR-10pruned neurons: 328 -> 466Mixed
ABL at 5% vs 10% poisoning ratio (LC attack)ABL defense · CIFAR-10poisoned samples among 500 isolated: 0 -> 500 poisoned of 500Mixed
change architecture from PreAct-ResNet18 to EfficientNet-B3BadNets, LC, SSBA, WaNet attacks · CIFAR-10attack successHurt
more classes (CIFAR-100, Tiny ImageNet vs CIFAR-10, GTSRB)backdoor defenses · R-Acc vs ASRrobust accuracy recoveryHurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy