List your company

Backdoor Defense on SST-2

91.71CACCVanilla
69.40275.193580.98586.7765Feb 19, 2024Jun 24, 2024Oct 28, 2024Mar 3, 2025Jul 7, 2025Nov 10, 2025Mar 16, 2026

Evaluation results

MethodLinks
VanillaAttack=StyleBkd, Target LM=BERT-base2024.0291.7177.19———
VanillaTarget LM=BERT-Base, Attack=StyleBkd2024.0291.7177.19———
RAPTarget LM=BERT-Base, Attack=Addsent2024.0291.7127.19———
No-DefenseAttack=Sentbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0391.5899.23———
No-DefenseAttack=Wordbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0391.4995.92———
No-DefenseAttack=Stylebkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0391.4270.11———
STRIPTarget LM=BERT-Base, Attack=Addsent2024.0291.3928.62———
VanillaAttack=Badnets, Target LM=BERT-base2024.0291.2794.63———
VanillaAttack=HiddenKiller, Target LM=BERT-base2024.0291.193.53———
VanillaTarget LM=BERT-Base, Attack=HiddenKiller2024.0291.193.53———
BadLLM-TGAttack=Wordbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0391.0914.76———
VanillaAttack=Addsent, Target LM=BERT-base2024.0290.9999.89———
VanillaTarget LM=BERT-Base, Attack=Addsent2024.0290.9999.89———
CUBETarget LM=BERT-Base, Attack=StyleBkd2024.0290.8322.43———
RAPAttack=Wordbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0390.8173.95———
BKITarget LM=BERT-Base, Attack=Addsent2024.0290.7233.05———
No-DefenseAttack=Synbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0390.5594.56———
STRIPTarget LM=BERT-Base, Attack=HiddenKiller2024.0290.3990.57———
BKITarget LM=BERT-Base, Attack=StyleBkd2024.0290.3482.46———
PrefixAttack=Badnets, Target LM=BERT-base2024.0290.1786.73———
RAPTarget LM=BERT-Base, Attack=StyleBkd2024.0290.1779.38———
PrefixAttack=StyleBkd, Target LM=BERT-base2024.0290.0663.6———
RAPAttack=Sentbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0390.0498.44———
STRIPTarget LM=BERT-Base, Attack=StyleBkd2024.0289.8978.62———
BadActsAttack=Sentbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0389.650.55———
PrefixAttack=Addsent, Target LM=BERT-base2024.0289.4699.89———
BadActsAttack=Wordbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0389.418.38———
BadActsAttack=Stylebkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0389.3556.84———
RAPAttack=Stylebkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0389.3167.21———
RAPAttack=Synbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0389.1494.43———
LoRAAttack=Badnets, Target LM=BERT-base2024.0289.0765.57———
LoRAAttack=Addsent, Target LM=BERT-base2024.0288.9696.16———
BadLLM-TGAttack=Stylebkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0388.9625.13———
LoRAAttack=StyleBkd, Target LM=BERT-base2024.0288.9157.24———
LoRAAttack=HiddenKiller, Target LM=BERT-base2024.0288.5852.96———
PrefixAttack=HiddenKiller, Target LM=BERT-base2024.0288.5268.64———
BKITarget LM=BERT-Base, Attack=HiddenKiller2024.0288.4194.85———
RAPTarget LM=BERT-Base, Attack=HiddenKiller2024.0288.2589.14———
BadLLM-TGAttack=Sentbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0388.240.86———
BadActsAttack=Synbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0388.1982.3———
MuScleLoRAAttack=StyleBkd, Target LM=BERT-base2024.0287.8133.22———
MuScleLoRATarget LM=BERT-Base, Attack=StyleBkd2024.0287.8133.22———
ONIONAttack=Sentbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0387.8186.75———
CUBETarget LM=BERT-Base, Attack=Addsent2024.0287.737.94———
MuScleLoRAAttack=HiddenKiller, Target LM=BERT-base2024.0287.6425.11———
MuScleLoRATarget LM=BERT-Base, Attack=HiddenKiller2024.0287.6425.11———
ONIONAttack=Stylebkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0387.3771.09———
ONIONAttack=Wordbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0387.2819.43———
AdapterAttack=Badnets, Target LM=BERT-base2024.0287.2655.26———
AdapterAttack=StyleBkd, Target LM=BERT-base2024.0287.160.96———
ONIONTarget LM=BERT-Base, Attack=Addsent2024.0287.0449.78———
AdapterAttack=Addsent, Target LM=BERT-base2024.0286.8887.83———
MuScleLoRAAttack=Addsent, Target LM=BERT-base2024.0286.7718.97———
MuScleLoRATarget LM=BERT-Base, Attack=Addsent2024.0286.7718.97———
AdapterAttack=HiddenKiller, Target LM=BERT-base2024.0286.645.39———
MuScleLoRAAttack=Badnets, Target LM=BERT-base2024.0286.5412.94———
CUBETarget LM=BERT-Base, Attack=HiddenKiller2024.0285.545.61———
ONIONTarget LM=BERT-Base, Attack=StyleBkd2024.0285.4581.76———
ONIONAttack=Synbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0385.3694.78———
ONIONTarget LM=BERT-Base, Attack=HiddenKiller2024.0285.2396.05———
MuScleLoRAAttack=Synbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0384.430.61———
MuScleLoRAAttack=Sentbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0384.3134.5———
MuScleLoRAAttack=Stylebkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0383.7137.11———
MuScleLoRAAttack=Wordbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0383.5417.96———
BadLLM-TGAttack=Synbkd, Victim Model=BERT-base-uncased, Poisoning Rate=20%2026.0370.2614.36———
MDPAttack=BadNets2023.09————0.99
MDPAttack=AddSent2023.09————0.99
MDPAttack=LWP2023.09————0.98
MDPAttack=EP2023.09————1
MDPAttack=SOS2023.09————1
ONIONAttack=BadNets2023.09————0.64
ONIONAttack=AddSent2023.09————0.54
ONIONAttack=LWP2023.09————0.72
ONIONAttack=EP2023.09————0.67
ONIONAttack=SOS2023.09————0.61
RAPAttack=BadNets2023.09————0.53
RAPAttack=AddSent2023.09————0.52
RAPAttack=LWP2023.09————0.83
RAPAttack=EP2023.09————0.56
RAPAttack=SOS2023.09————0.51
STRIPAttack=BadNets2023.09————0.66
STRIPAttack=AddSent2023.09————0.51
STRIPAttack=LWP2023.09————0.6
STRIPAttack=EP2023.09————0.84
STRIPAttack=SOS2023.09————0.82