SOTAVerified|Agents Browse Leaderboard About

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 Added Most Hyped Most Active Needs Verification Most Verified

Showing 91–100 of 523 papers

Title	Date	Tasks	Status	Hype
Defending Against Backdoor Attacks in Natural Language Generation	Jun 3, 2021	Backdoor AttackDialogue Generation	CodeCode Available	1
Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger	May 26, 2021	Backdoor Attack	CodeCode Available	1
Backdoor Attacks on Self-Supervised Learning	May 21, 2021	Backdoor AttackInductive Bias	CodeCode Available	1
Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Models	Mar 29, 2021	Backdoor AttackData Poisoning	CodeCode Available	1
Targeted Attack against Deep Neural Networks via Flipping Limited Weight Bits	Feb 21, 2021	Adversarial AttackBackdoor Attack	CodeCode Available	1
WaNet -- Imperceptible Warping-based Backdoor Attack	Feb 20, 2021	Backdoor Attack	CodeCode Available	1
Red Alarm for Pre-trained Models: Universal Vulnerability to Neuron-Level Backdoor Attacks	Jan 18, 2021	Backdoor Attack	CodeCode Available	1
LIRA: Learnable, Imperceptible and Robust Backdoor Attacks	Jan 1, 2021	Backdoor Attackbackdoor defense	CodeCode Available	1
WaNet - Imperceptible Warping-based Backdoor Attack	Jan 1, 2021	Backdoor Attack	CodeCode Available	1
Deep Feature Space Trojan Attack of Neural Networks by Controlled Detoxification	Dec 21, 2020	Adversarial AttackBackdoor Attack	CodeCode Available	1

Show:10 25 50

← PrevPage 10 of 53Next →

No leaderboard results yet.