packages of attack and defense

attack Methods

BadNet	Badnets: Identifying vulnerabilities in the machine learning model supply chain.
Blended	Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
Blind	Blind Backdoors in Deep Learning Models
Bpp	BppAttack: Stealthy and Efficient Trojan Attacks Against Deep Neural Networks via Image Quantization and Contrastive Adversarial Learning
CTRL	An Embarrassingly Simple Backdoor Attack on Self-supervised Learning
InputAware	Input-aware dynamic backdoor attack
LabelConsistent	Label-Consistent Backdoor Attacks
LowFrequency	Rethinking the backdoor attacks' triggers: A frequency perspective
LIRA	LIRA: Learnable, Imperceptible and Robust Backdoor Attacks
PoisonInk	Poison ink: Robust and invisible backdoor attack
NormalCase	Normal training case (Train a clean model with clean data)
Refool	Reflection Backdoor: A Natural Backdoor Attack on Deep Neural Networks
SIG	A new backdoor attack in CNNs by training set corruption without label poisoning
SSBA	Invisible backdoor attack with sample-specific triggers
TrojanNN	Trojaning Attack on Neural Networks
Wanet	WaNet - Imperceptible Warping-based Backdoor Attack

defense Methods

abl	Anti-backdoor learning: Training clean models on poisoned data.
ac	Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering
anp	Adversarial Neuron Pruning Purifies Backdoored Deep Models
bnp	Pre-activation Distributions Expose Backdoor Neurons
clp	Data-free backdoor removal based on channel lipschitzness
d_br	Effective backdoor defense by exploiting sensitivity of poisoned samples
d_st	Effective backdoor defense by exploiting sensitivity of poisoned samples
dbd	Backdoor Defense Via Decoupling The Training Process
ep	Pre-activation Distributions Expose Backdoor Neurons
fp	Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks
ft	Basic class for ft defense method.
ft_sam	Enhancing Fine-Tuning Based Backdoor Defense with Sharpness-Aware Minimization
i_bau	Adversarial unlearning of backdoors via implicit hypergradient
mbns	Pre-activation Distributions Expose Backdoor Neurons
mcr	Bridging mode connectivity in loss landscapes and adversarial robustness
nab	Beating Backdoor Attack at Its Own Game
nad	Neural Attention Distillation: Erasing Backdoor Triggers From Deep Neural Networks
nc	Neural Cleanse: Identifying And Mitigating Backdoor Attacks In Neural Networks
npd	Neural polarizer: A lightweight and effective backdoor defense via purifying poisoned features
rnp	Reconstructive Neuron Pruning for Backdoor Defense
sau	Shared adversarial unlearning: Backdoor mitigation by unlearning shared adversarial examples
spectral	Spectral Signatures in Backdoor Attacks

inference-time detection Methods

strip	STRIP: A Defence Against Trojan Attacks on Deep Neural Networks
teco	Detecting Backdoors During the Inference Stage Based on Corruption Robustness Consistency

pretraining detection Methods

strip	STRIP: A Defence Against Trojan Attacks on Deep Neural Networks
beatrix	The Beatrix Resurrections: Robust Backdoor Detection via Gram Matrices
scan	Demon in the Variant: Statistical Analysis of DNNs for Robust Backdoor Contamination Detection
spectre	SPECTRE: Defending Against Backdoor Attacks Using Robust Statistics
spectral	Spectral Signatures in Backdoor Attacks
ac	Detecting Backdoor Attacks on Deep Neural Networks by Activation Clustering