Papers with Code 아카이브
Data Poisoning
태스크의 하위 분류와 논문·코드·평가·기법·데이터셋을 탐색하세요.
- 하위 태스크
- 0
- 평가표
- 0
- 보존된 평가 기록
- 0
Data Poisoning is an adversarial attack that tries to manipulate the training dataset in order to control the prediction behavior of a trained model such that the model will label malicious examples into a desired classes (e.g., labeling spam e-mails as safe).
Source: Explaining Vulnerabilities to Adversarial Machine Learning through Visual Analytics
논문
이 태스크와 하위 분야의 관련 논문입니다. 실제 성능 비교는 평가 결과에서 확인하세요.
On the Exploitability of Instruction Tuning
Adversarial Backdoor Attack by Naturalistic Data Poisoning on Trajectory Prediction in Autonomous Driving
OVLA: Neural Network Ownership Verification using Latent Watermarks
Data Poisoning to Fake a Nash Equilibrium in Markov Games
FheFL: Fully Homomorphic Encryption Friendly Privacy-Preserving Federated Learning with Byzantine Users
DeepfakeArt Challenge: A Benchmark Dataset for Generative AI Art Forgery and Data Poisoning Detection
Hyperparameter Learning under Data Poisoning: Analysis of the Influence of Regularization via Multiobjective Bilevel Optimization
Adversarial Clean Label Backdoor Attacks and Defenses on Text Classification Systems
Backdoor Attacks Against Incremental Learners: An Empirical Evaluation Study
Instructions as Backdoors: Backdoor Vulnerabilities of Instruction Tuning for Large Language Models
From Shortcuts to Triggers: Backdoor Defense with Denoised PoE
Differentially-Private Decision Trees and Provable Robustness to Data Poisoning
Faithful and Efficient Explanations for Neural Networks via Neural Tangent Kernel Surrogate Models
Quantifying the robustness of deep multispectral segmentation models against natural perturbations and data poisoning
FedGT: Identification of Malicious Clients in Federated Learning with Secure Aggregation
Evaluating Impact of User-Cluster Targeted Attacks in Matrix Factorisation Recommenders
Text-to-Image Diffusion Models can be Easily Backdoored through Multimodal Data Poisoning
Pick your Poison: Undetectability versus Robustness in Data Poisoning Attacks
Beyond the Model: Data Pre-processing Attack to Deep Learning Models in Android Apps
Interactive System-wise Anomaly Detection
INK: Inheritable Natural Backdoor Attack Against Model Distillation
Defending Against Patch-based Backdoor Attacks on Self-Supervised Learning
Mole Recruitment: Poisoning of Image Classifiers via Selective Batch Sampling
Denoising Autoencoder-based Defensive Distillation as an Adversarial Robustness Algorithm
Learning the Unlearnable: Adversarial Augmentations Suppress Unlearnable Example Attacks
PORE: Provably Robust Recommender Systems against Data Poisoning Attacks
Recursive Euclidean Distance Based Robust Aggregation Technique For Federated Learning
Robust Contrastive Language-Image Pre-training against Data Poisoning and Backdoor Attacks
Naive Bayes Classifiers over Missing Data: Decision and Poisoning
Exploring the Limits of Model-Targeted Indiscriminate Data Poisoning Attacks
CleanCLIP: Mitigating Data Poisoning Attacks in Multimodal Contrastive Learning
WW-FL: Secure and Private Large-Scale Federated Learning
Poisoning Web-Scale Training Datasets is Practical
QTrojan: A Circuit Backdoor Against Quantum Neural Networks
Explainable Label-flipping Attacks on Human Emotion Assessment System
출처: Papers with Code · CC BY-SA 4.0. 아카이브 자료와 새로 수집한 논문은 별도로 관리합니다.