SOTAVerified

DNN Testing

Testing the reliability of DNNs.

Papers

Showing 125 of 27 papers

TitleStatusHype
Revisiting Neuron Coverage for DNN Testing: A Layer-Wise and Distribution-Aware CriterionCode1
Perception Matters: Detecting Perception Failures of VQA Models Using Metamorphic TestingCode1
Distribution-Aware Testing of Neural Networks Using Generative ModelsCode1
RBT4DNN: Requirements-based Testing of Neural NetworksCode0
GAN-enhanced Simulation-driven DNN Testing in Absence of Ground Truth0
DANDI: Diffusion as Normative Distribution for Deep Neural Network Input0
Bridging the Gap between Real-world and Synthetic Images for Testing Autonomous Driving Systems0
Targeted Deep Learning System Boundary TestingCode0
DeepSample: DNN sampling-based testing for operational accuracy assessment0
DeepKnowledge: Generalisation-Driven Deep Learning Testing0
Rethinking Diversity in Deep Neural Network Testing0
CGDTest: A Constrained Gradient Descent Algorithm for Testing Neural Networks0
An Overview of Structural Coverage Metrics for Testing Neural NetworksCode0
NPC: Neuron Path Coverage via Characterizing Decision Logic of Deep Neural Networks0
CatchBackdoor: Backdoor Detection via Critical Trojan Neural Path Fuzzing0
Black-Box Testing of Deep Neural Networks Through Test Case DiversityCode0
Provably Valid and Diverse Mutations of Real-World Media Data for DNN Testing0
Exposing Previously Undetectable Faults in Deep Neural Networks0
Neuron Coverage-Guided Domain Generalization0
IWA: Integrated Gradient based White-box Attacks for Fooling Deep Neural Networks0
Towards Practical Robustness Analysis for DNNs based on PAC-Model LearningCode0
Automatic Test Suite Generation for Key-Points Detection DNNs using Many-Objective Search (Experience Paper)0
Comparing Offline and Online Testing of Deep Neural Networks: An Autonomous Car Case Study0
Boosting Operational DNN Testing Efficiency through ConditioningCode0
Testing DNN Image Classifiers for Confusion & Bias ErrorsCode0
Show:102550
← PrevPage 1 of 2Next →

No leaderboard results yet.