Papers with Code 아카이브

Backdoor Attack

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

전체 기법 전체 데이터셋

하위 태스크: 0
평가표: 0
보존된 평가 기록: 0

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.

코드

관련 논문과 평가 결과에 연결된 구현입니다. 출처에서 확인된 저자 구현에는 공식 코드 표시가 있습니다.

https://github.com/nura-j/attack_fl

Genetic Algorithm-Based Dynamic Backdoor Attack on Federated Learning-Based Network Traffic Classification

https://github.com/radiant0726/cbw

Backdoor Attack against Speaker Verification

https://github.com/reds-lab/narcissus

Narcissus: A Practical Clean-Label Backdoor Attack with Limited Information

https://github.com/ruoxi-jia-group/narcissus

Narcissus: A Practical Clean-Label Backdoor Attack with Limited Information

https://github.com/shuchiwu/rda

BadEncoder: Backdoor Attacks to Pre-trained Encoders in Self-Supervised Learning

https://github.com/thunlp/DictSKB

Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger

https://github.com/wegodev2/virtual-prompt-injection

Backdooring Instruction-Tuned Large Language Models with Virtual Prompt Injection

https://github.com/xandery-geek/BackdoorAttacks

A new Backdoor Attack in CNNs by training set corruption without label poisoning

https://github.com/xqx12/daily-info

Neurotoxin: Durable Backdoors in Federated Learning

https://github.com/yangyunfei16/DBIA

DBIA: Data-free Backdoor Injection Attack against Transformer Networks

https://github.com/zhang-henry/inactive

Towards Imperceptible Backdoor Attack in Self-supervised Learning

출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.