SOTAVerified

Backdoor Attack

Backdoor attacks inject maliciously constructed data into a training set so that, at test time, the trained model misclassifies inputs patched with a backdoor trigger as an adversarially-desired target class.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 291–300 of 523 papers

TitleStatusHype
A general approach to enhance the survivability of backdoor attacks by decision path couplingCode0
SynGhost: Invisible and Universal Task-agnostic Backdoor Attack via Syntactic TransferCode0
Model Pairing Using Embedding Translation for Backdoor Attack Detection on Open-Set Classification TasksCode0
Low-Frequency Black-Box Backdoor Attack via Evolutionary Algorithm—0
VL-Trojan: Multimodal Instruction Backdoor Attacks against Autoregressive Visual Language Models—0
Whispers in Grammars: Injecting Covert Backdoors to Compromise Dense Retrieval SystemsCode0
Defending Against Weight-Poisoning Backdoor Attacks for Parameter-Efficient Fine-Tuning—0
Backdoor Attack against One-Class Sequential Anomaly Detection ModelsCode0
OrderBkd: Textual backdoor attack through repositioningCode0
The last Dance : Robust backdoor attack via diffusion models and bayesian approach—0
Show:102550
← PrevPage 30 of 53Next →

No leaderboard results yet.