SOTAVerified

Interpretability Techniques for Deep Learning

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 25 papers

TitleStatusHype
Time series saliency maps: explaining models across multiple domainsCode1
Dissecting and Mitigating Diffusion Bias via Mechanistic InterpretabilityCode1
IBO: Inpainting-Based Occlusion to Enhance Explainable Artificial Intelligence Evaluation in HistopathologyCode0
Explainable Deep Learning: A Visual Analytics Approach with Transition MatricesCode0
CausalGym: Benchmarking causal interpretability methods on linguistic tasksCode2
Less is More: Fewer Interpretable Region via Submodular Subset SelectionCode2
TraceFL: Interpretability-Driven Debugging in Federated Learning via Neuron ProvenanceCode1
Improving Interpretability via Regularization of Neural Activation Sensitivity—0
Making Sense of Dependence: Efficient Black-box Explanations Using Dependence MeasureCode1
Learning the Dynamics of Physical Systems from Sparse Observations with Finite Element NetworksCode1
Show:102550
← PrevPage 1 of 3Next →
All datasetsCausalGymCelebA

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1RISEInsertion AUC score0.57—Unverified
2HSIC-AttributionInsertion AUC score0.57—Unverified
3Kernel SHAPInsertion AUC score0.52—Unverified
4LIMEInsertion AUC score0.52—Unverified
5SaliencyInsertion AUC score0.46—Unverified
6Grad-CAMInsertion AUC score0.37—Unverified
7Integrated GradientsInsertion AUC score0.36—Unverified