Papers with Code 아카이브

Adversarial Attack

태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.

전체 기법 전체 데이터셋

하위 태스크: 4
평가표: 3
보존된 평가 기록: 9

An Adversarial Attack is a technique to find a perturbation that changes the prediction of a machine learning model. The perturbation can be very small and imperceptible to human eyes.

Source: Recurrent Attention Model with Log-Polar Mapping is Robust against Adversarial Attacks

코드

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

https://github.com/gongzhitaao/adversarial-classifier

공식 코드

Adversarial and Clean Data Are Not Twins

https://github.com/google-research/retvec

공식 코드

RETVec: Resilient and Efficient Text Vectorizer

https://github.com/google-research/scenic/tree/main/scenic/projects/adversarialtraining

공식 코드

Pyramid Adversarial Training Improves ViT Performance

https://github.com/google/unisim

공식 코드

RETSim: Resilient and Efficient Text Similarity

https://github.com/graph-axis/gcleaner

공식 코드

"No Matter What You Do": Purifying GNN Models via Backdoor Unlearning

https://github.com/grasses/poisonprompt

공식 코드

PoisonPrompt: Backdoor Attack on Prompt-based Large Language Models

https://github.com/greencubic/lookhere

공식 코드

LookHere: Vision Transformers with Directed Attention Generalize and Extrapolate

https://github.com/grispeut/Feature-Alignment

공식 코드

Using Feature Alignment Can Improve Clean Average Precision and Adversarial Robustness in Object Detection

https://github.com/gtolias/tma

공식 코드

Targeted Mismatch Adversarial Attack: Query with a Flower to Retrieve the Tower

https://github.com/guijiejie/aadn

공식 코드

Fooling the Image Dehazing Models by First Order Gradient

https://github.com/guowei-cn/cca_ud-universal-training-level-defence

공식 코드

Universal Detection of Backdoor Attacks via Density-based Clustering and Centroids Analysis

https://github.com/gzhu-dvl/lhs-ba

공식 코드

Query-Efficient Adversarial Attack Based on Latin Hypercube Sampling

https://github.com/hamidbostani2021/evadedroid

공식 코드

EvadeDroid: A Practical Evasion Attack on Machine Learning for Black-box Android Malware Detection

https://github.com/hanbocai/badspeech_soe

공식 코드

Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound

https://github.com/hanssuny/diffam

공식 코드

DiffAM: Diffusion-based Adversarial Makeup Transfer for Facial Privacy Protection

https://github.com/haofanwang/Score-CAM

공식 코드

Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks

https://github.com/haowang-cqu/transtroj

공식 코드

Model Supply Chain Poisoning: Backdooring Pre-trained Models via Embedding Indistinguishability

https://github.com/harrywuhust2022/Adv-Crowd-analysis

공식 코드

Towards Adversarial Patch Analysis and Certified Defense against Crowd Counting

https://github.com/havelhuang/EKiML-embed-knowledge-into-ML-model

공식 코드

Embedding and Extraction of Knowledge in Tree Ensemble Classifiers

https://github.com/hazardfy/fiba

공식 코드

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

https://github.com/hbrachemi/energy_backdoor

공식 코드

Energy Backdoor Attack to Deep Neural Networks

https://github.com/hbrachemi/vlm_defense-attack

공식 코드

VIP: Visual Information Protection through Adversarial Attacks on Vision-Language Models

https://github.com/hendrycks/natural-adv-examples

공식 코드

Natural Adversarial Examples

https://github.com/henryzxu/lane-verification

공식 코드

Model-Agnostic Defense for Lane Detection against Adversarial Attack

https://github.com/hfawaz/ijcnn19attacks

공식 코드

Adversarial Attacks on Deep Neural Networks for Time Series Classification

https://github.com/hgh0545/NA2

공식 코드

Query-Efficient Adversarial Attack Against Vertical Federated Graph Learning

https://github.com/hgh0545/graph-fraudster

공식 코드

Graph-Fraudster: Adversarial Attacks on Graph Neural Network Based Vertical Federated Learning

https://github.com/hku-tasr/anywheredoor

공식 코드

AnywhereDoor: Multi-Target Backdoor Attacks on Object Detection

https://github.com/hmdolatabadi/AdvFlow

공식 코드

Black-box Adversarial Example Generation with Normalizing Flows

https://github.com/hncszyq/ShadowAttack

공식 코드

Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon

https://github.com/hong-xian/backdoor-attack

공식 코드

Generalization Bound and New Algorithm for Clean-Label Backdoor Attack

https://github.com/hongw579/agkd-bml

공식 코드

AGKD-BML: Defense Against Adversarial Attack by Attention Guided Knowledge Distillation and Bi-directional Metric Learning

https://github.com/hqa-attack/hqaattack-demo

공식 코드

HQA-Attack: Toward High Quality Black-Box Hard-Label Adversarial Attack on Text

https://github.com/hsndbk4/arc

공식 코드

Adversarial Vulnerability of Randomized Ensembles

https://github.com/hsouri/Sleeper-Agent

공식 코드

Sleeper Agent: Scalable Hidden Trigger Backdoors for Neural Networks Trained from Scratch

https://github.com/hsouri/gdp

공식 코드

Generating Potent Poisons and Backdoors from Scratch with Guided Diffusion

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