SOTAVerified

Adversarial Robustness

Adversarial Robustness evaluates the vulnerabilities of machine learning models under various types of adversarial attacks.

Papers

Showing 14011450 of 1746 papers

TitleStatusHype
Buffer Zone based Defense against Adversarial Examples in Image Classification0
What are effective labels for augmented data? Improving robustness with AutoLabel0
Perturbation Type Categorization for Multiple _p Bounded Adversarial Robustness0
Disentangling Adversarial Robustness in Directions of the Data ManifoldCode0
Intriguing class-wise properties of adversarial training0
No Feature Is An Island: Adaptive Collaborations Between Features Improve Adversarial Robustness0
Test-Time Adaptation and Adversarial Robustness0
Collective Robustness Certificates0
Improving Adversarial Robustness in Weight-quantized Neural Networks0
A Simple Fine-tuning Is All You Need: Towards Robust Deep Learning Via Adversarial Fine-tuning0
Gradient-Free Adversarial Attacks for Bayesian Neural NetworksCode0
Self-Progressing Robust TrainingCode0
Sample Complexity of Adversarially Robust Linear Classification on Separated Data0
Adversarially Robust Estimate and Risk Analysis in Linear Regression0
On the human-recognizability phenomenon of adversarially trained deep image classifiersCode0
Generate and Verify: Semantically Meaningful Formal Analysis of Neural Network Perception Systems0
A case for new neural network smoothness constraints0
Achieving Adversarial Robustness Requires An Active Teacher0
Improving Adversarial Robustness via Probabilistically Compact Loss with Logit ConstraintsCode0
Learning Energy-Based Models With Adversarial TrainingCode0
On 1/n neural representation and robustnessCode0
Overcomplete Representations Against Adversarial VideosCode0
Evaluating adversarial robustness in simulated cerebellum0
Unsupervised Adversarially-Robust Representation Learning on Graphs0
Stochastic Gradient Descent with Nonlinear Conjugate Gradient-Style Adaptive Momentum0
Interpretable Graph Capsule Networks for Object Recognition0
FAT: Federated Adversarial Training0
How Robust are Randomized Smoothing based Defenses to Data Poisoning?0
Fast Adversarial Robustness Certification of Nearest Prototype Classifiers for Arbitrary SeminormsCode0
Adversarial Robustness Across Representation Spaces0
Robust and Private Learning of Halfspaces0
Architectural Adversarial Robustness: The Case for Deep Pursuit0
Incorporating Hidden Layer representation into Adversarial Attacks and Defences0
A Study on the Uncertainty of Convolutional Layers in Deep Neural Networks0
Exposing the Robustness and Vulnerability of Hybrid 8T-6T SRAM Memory Architectures to Adversarial Attacks in Deep Neural Networks0
aw_nas: A Modularized and Extensible NAS framework0
A More Biologically Plausible Local Learning Rule for ANNs0
On Adversarial Robustness of 3D Point Cloud Classification under Adaptive Attacks0
Adversarial Examples for k-Nearest Neighbor Classifiers Based on Higher-Order Voronoi DiagramsCode0
An Experimental Study of Semantic Continuity for Deep Learning Models0
Effective, Efficient and Robust Neural Architecture Search0
Contextual Fusion For Adversarial Robustness0
Robustified Domain Adaptation0
Probing Predictions on OOD Images via Nearest CategoriesCode0
Towards Understanding the Regularization of Adversarial Robustness on Neural Networks0
Bridging the Performance Gap between FGSM and PGD Adversarial TrainingCode0
Recent Advances in Understanding Adversarial Robustness of Deep Neural Networks0
Beyond cross-entropy: learning highly separable feature distributions for robust and accurate classification0
Robustness May Be at Odds with Fairness: An Empirical Study on Class-wise Accuracy0
Towards Robust Neural Networks via Orthogonal DiversityCode0
Show:102550
← PrevPage 29 of 35Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DeBERTa (single model)Accuracy0.61Unverified
2ALBERT (single model)Accuracy0.59Unverified
3T5 (single model)Accuracy0.57Unverified
4SMART_RoBERTa (single model)Accuracy0.54Unverified
5FreeLB (single model)Accuracy0.5Unverified
6RoBERTa (single model)Accuracy0.5Unverified
7InfoBERT (single model)Accuracy0.46Unverified
8ELECTRA (single model)Accuracy0.42Unverified
9BERT (single model)Accuracy0.34Unverified
10SMART_BERT (single model)Accuracy0.3Unverified
#ModelMetricClaimedVerifiedStatus
1Mixed classifierAccuracy95.23Unverified
2Stochastic-LWTA/PGD/WideResNet-34-10Accuracy92.26Unverified
3Stochastic-LWTA/PGD/WideResNet-34-5Accuracy91.88Unverified
4GLOT-DRAccuracy84.13Unverified
5TRADES-ANCRA/ResNet18Accuracy81.7Unverified
#ModelMetricClaimedVerifiedStatus
1ResNet-50 (SGD, Cosine)Accuracy77.4Unverified
2ResNet-50 (SGD, Step)Accuracy76.9Unverified
3DeiT-S (AdamW, Cosine)Accuracy76.8Unverified
4ResNet-50 (AdamW, Cosine)Accuracy76.4Unverified
#ModelMetricClaimedVerifiedStatus
1DeiT-S (AdamW, Cosine)Accuracy12.2Unverified
2ResNet-50 (SGD, Cosine)Accuracy3.3Unverified
3ResNet-50 (SGD, Step)Accuracy3.2Unverified
4ResNet-50 (AdamW, Cosine)Accuracy3.1Unverified
#ModelMetricClaimedVerifiedStatus
1ResNet-50 (AdamW, Cosine)mean Corruption Error (mCE)59.3Unverified
2ResNet-50 (SGD, Step)mean Corruption Error (mCE)57.9Unverified
3ResNet-50 (SGD, Cosine)mean Corruption Error (mCE)56.9Unverified
4DeiT-S (AdamW, Cosine)mean Corruption Error (mCE)48Unverified
#ModelMetricClaimedVerifiedStatus
1DeiT-S (AdamW, Cosine)Accuracy13Unverified
2ResNet-50 (SGD, Cosine)Accuracy8.4Unverified
3ResNet-50 (SGD, Step)Accuracy8.3Unverified
4ResNet-50 (AdamW, Cosine)Accuracy8.1Unverified
#ModelMetricClaimedVerifiedStatus
1Mixed ClassifierClean Accuracy85.21Unverified
2ResNet18/MART-ANCRAClean Accuracy60.1Unverified