Papers with Code 아카이브

backdoor defense

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

전체 기법전체 데이터셋

하위 태스크

1

평가표

0

보존된 평가 기록

0

코드

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

https://github.com/rfww/spy-watermark

공식 코드

Spy-Watermark: Robust Invisible Watermarking for Backdoor Attack

https://github.com/rorschachchen/pipd

공식 코드

Progressive Poisoned Data Isolation for Training-time Backdoor Defense

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

공식 코드

ASSET: Robust Backdoor Data Detection Across a Multiplicity of Deep Learning Paradigms

https://github.com/sclbd/backdoorbench

공식 코드

Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor

https://github.com/sclbd/dbd

공식 코드

Backdoor Defense via Decoupling the Training Process

https://github.com/sf-zhai/badt2i

공식 코드

Text-to-Image Diffusion Models can be Easily Backdoored through Multimodal Data Poisoning

https://github.com/shawkui/Proactive_Defensive_Backdoor

공식 코드

Mitigating Backdoor Attack by Injecting Proactive Defensive Backdoor

https://github.com/shawkui/Shared_Adversarial_Unlearning

공식 코드

Shared Adversarial Unlearning: Backdoor Mitigation by Unlearning Shared Adversarial Examples

https://github.com/shymuel/diff-cleanse

공식 코드

Diff-Cleanse: Identifying and Mitigating Backdoor Attacks in Diffusion Models

https://github.com/sycny/zip

공식 코드

Black-box Backdoor Defense via Zero-shot Image Purification

https://github.com/thunlp/ONION

공식 코드

ONION: A Simple and Effective Defense Against Textual Backdoor Attacks

https://github.com/thuyimingli/backdoorbox

공식 코드

REFINE: Inversion-Free Backdoor Defense via Model Reprogramming

https://github.com/tsun/bdmae

공식 코드

Mask and Restore: Blind Backdoor Defense at Test Time with Masked Autoencoder

https://github.com/vita-group/trap-and-replace-backdoor-defense

공식 코드

Trap and Replace: Defending Backdoor Attacks by Trapping Them into an Easy-to-Replace Subnetwork

https://github.com/wanghangpsu/mm-bd

공식 코드

MM-BD: Post-Training Detection of Backdoor Attacks with Arbitrary Backdoor Pattern Types Using a Maximum Margin Statistic

https://github.com/wecanshow/mcl

공식 코드

Model-Contrastive Learning for Backdoor Defense

https://github.com/whitolfchen/refine

공식 코드

REFINE: Inversion-Free Backdoor Defense via Model Reprogramming

https://github.com/xdjph/btu

공식 코드

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models

https://github.com/xiaoyunxxy/parameter_backdoor

공식 코드

Towards Backdoor Stealthiness in Model Parameter Space

https://github.com/ybdai7/trojandam-backdoor-defense

공식 코드

TrojanDam: Detection-Free Backdoor Defense in Federated Learning through Proactive Model Robustification utilizing OOD Data

https://github.com/zaixizhang/cbd

공식 코드

Backdoor Defense via Deconfounded Representation Learning

https://github.com/zrw00/graceful

공식 코드

Gracefully Filtering Backdoor Samples for Generative Large Language Models without Retraining

https://github.com/aisafety-hkust/backdoor_safety_tuning

Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense

https://github.com/aryan-satpathy/backdoor

FIBA: Frequency-Injection based Backdoor Attack in Medical Image Analysis

https://github.com/lancopku/rap

ONION: A Simple and Effective Defense Against Textual Backdoor Attacks

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

FIBA: Frequency-Injection based Backdoor Attack in Medical Image Analysis

https://github.com/yjandali/BackdoorBench-MergeGuard

Backdoor Defense via Decoupling the Training Process

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