SOTAVerified

Adversarial Defense

Competitions with currently unpublished results:

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 403 papers

TitleStatusHype
Bridging Robustness and Generalization Against Word Substitution Attacks in NLP via the Growth Bound Matrix ApproachCode0
Active Adversarial Noise Suppression for Image Forgery Localization—0
Sylva: Tailoring Personalized Adversarial Defense in Pre-trained Models via Collaborative Fine-tuning—0
Towards Effective and Efficient Adversarial Defense with Diffusion Models for Robust Visual TrackingCode0
Are classical deep neural networks weakly adversarially robust?—0
Adversarially Robust AI-Generated Image Detection for Free: An Information Theoretic Perspective—0
A Framework for Adversarial Analysis of Decision Support Systems Prior to Deployment—0
EdgeAgentX: A Novel Framework for Agentic AI at the Edge in Military Communication Networks—0
Accidental Misalignment: Fine-Tuning Language Models Induces Unexpected VulnerabilityCode0
A Few Large Shifts: Layer-Inconsistency Based Minimal Overhead Adversarial Example DetectionCode0
GenoArmory: A Unified Evaluation Framework for Adversarial Attacks on Genomic Foundation ModelsCode1
Lightweight Defense Against Adversarial Attacks in Time Series ClassificationCode0
Hydra: An Agentic Reasoning Approach for Enhancing Adversarial Robustness and Mitigating Hallucinations in Vision-Language Models—0
A Knowledge-guided Adversarial Defense for Resisting Malicious Visual Manipulation—0
AdPO: Enhancing the Adversarial Robustness of Large Vision-Language Models with Preference Optimization—0
Leveraging Generalizability of Image-to-Image Translation for Enhanced Adversarial Defense—0
Robust Deep Reinforcement Learning in Robotics via Adaptive Gradient-Masked Adversarial Attacks—0
One Stone, Two Birds: Enhancing Adversarial Defense Through the Lens of Distributional DiscrepancyCode0
Class-Conditional Neural Polarizer: A Lightweight and Effective Backdoor Defense by Purifying Poisoned Features—0
ShieldLearner: A New Paradigm for Jailbreak Attack Defense in LLMs—0
How vulnerable is my policy? Adversarial attacks on modern behavior cloning policies—0
Adversarial Masked Autoencoder Purifier with Defense Transferability—0
VideoPure: Diffusion-based Adversarial Purification for Video RecognitionCode0
Gradient-Free Adversarial Purification with Diffusion Models—0
Double Visual Defense: Adversarial Pre-training and Instruction Tuning for Improving Vision-Language Model Robustness—0
Towards a constructive framework for control theory—0
Sample Correlation for Fingerprinting Deep Face RecognitionCode0
Proactive Adversarial Defense: Harnessing Prompt Tuning in Vision-Language Models to Detect Unseen Backdoored Images—0
Defensive Dual Masking for Robust Adversarial Defense—0
A Real-Time Defense Against Object Vanishing Adversarial Patch Attacks for Object Detection in Autonomous Vehicles—0
Adversarial Transferability in Deep Denoising Models: Theoretical Insights and Robustness Enhancement via Out-of-Distribution Typical Set Sampling—0
Sustainable Self-evolution Adversarial Training—0
Neural Fingerprints for Adversarial Attack DetectionCode0
Alignment-Based Adversarial Training (ABAT) for Improving the Robustness and Accuracy of EEG-Based BCIsCode0
CausalDiff: Causality-Inspired Disentanglement via Diffusion Model for Adversarial DefenseCode1
Test-time Adversarial Defense with Opposite Adversarial Path and High Attack Time Cost—0
Unitary Multi-Margin BERT for Robust Natural Language ProcessingCode0
Perseus: Leveraging Common Data Patterns with Curriculum Learning for More Robust Graph Neural Networks—0
The Good, the Bad and the Ugly: Watermarks, Transferable Attacks and Adversarial Defenses—0
Secure Video Quality Assessment Resisting Adversarial Attacks—0
Interpretability-Guided Test-Time Adversarial DefenseCode0
Cloud Adversarial Example Generation for Remote Sensing Image Classification—0
Real-world Adversarial Defense against Patch Attacks based on Diffusion ModelCode1
Bypassing DARCY Defense: Indistinguishable Universal Adversarial Triggers—0
Certified Causal Defense with Generalizable Robustness—0
TART: Boosting Clean Accuracy Through Tangent Direction Guided Adversarial Training—0
Improving Out-of-Distribution Data Handling and Corruption Resistance via Modern Hopfield NetworksCode0
Robust Image Classification: Defensive Strategies against FGSM and PGD Adversarial Attacks—0
Privacy-preserving Universal Adversarial Defense for Black-box Models—0
OTAD: An Optimal Transport-Induced Robust Model for Agnostic Adversarial Attack—0
Show:102550
← PrevPage 1 of 9Next →
All datasetsCIFAR-10ImageNet (non-targeted PGD, max perturbation=4)ImageNetImageNet (targeted PGD, max perturbation=16)MNISTCAAD 2018miniImageNet

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1WRN-28-10Accuracy90.03—Unverified
2Diffusion ClassifierAccuracy89.85—Unverified
3Stochastic-LWTA/PGD/WideResNet-34-10Accuracy84.3—Unverified
4Ours (Stochastic-LWTA/PGD/WideResNet-34-5)Accuracy83.4—Unverified
5Ours (Stochastic-LWTA/PGD/WideResNet-34-1)Accuracy81.87—Unverified
6ResNet18 (TRADES-ANCRA/PGD-40)Accuracy81.7—Unverified
7Stochastic-LWTA/PGD/WideResNet-34-5Attack: AutoAttack81.22—Unverified
8PCL (against PGD, white box)Accuracy46.7—Unverified
#ModelMetricClaimedVerifiedStatus
1SAT-EfficientNet-L1Accuracy58.6—Unverified
2LLR-ResNet-152Accuracy47—Unverified
3ResNet-152 free-m=4Accuracy36—Unverified
4ResNet-101 free-m=4Accuracy34.3—Unverified
5ResNet-50 free-m=4Accuracy31.8—Unverified
#ModelMetricClaimedVerifiedStatus
1ResNet101Accuracy99.8—Unverified
2InceptionV3Accuracy98.6—Unverified
3Feature DenoisingAccuracy49.5—Unverified
#ModelMetricClaimedVerifiedStatus
1ResNet-152 DenoiseAccuracy42.8—Unverified
2ResNeXt-101 DenoiseAllAccuracy40.4—Unverified
3ResNet-152Accuracy39—Unverified
#ModelMetricClaimedVerifiedStatus
1Defense GANAccuracy0.85—Unverified
2PuVAEAccuracy0.81—Unverified
#ModelMetricClaimedVerifiedStatus
1Feature DenoisingAccuracy50.6—Unverified
#ModelMetricClaimedVerifiedStatus
1Auto Encoder-Block Switching defense with GradCAMAccuracy 88.54—Unverified