AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Debugging Tests for Model Explanations

Julius Adebayo, Michael Muelly, Ilaria Liccardi, Been Kim

NeurIPS 2020 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation · 7 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
inject spurious background and check attributions vs ideal maskfeature attributions · spurious sky/bamboo backgroundSSIM-GT2: 0.78 to 0.98Improved
compare attributions of mislabelled vs correctly labelled training inputfeature attributions · 10% flipped labelsSSIM: 0.73-0.99No clear effect
successively re-initialize VGG-16 weights from top layermodified backprop methods (GBP, DConvNet, PNet, PAttribution, LRP-SPAF) · VGG-16 ImageNetattribution changeNo clear effect
compare attributions from out-of-domain model with in-domain modelGradient-style attributions · Fashion MNIST to ImageNet VGG-16SSIM / rank correlation: SSIM 0.87 for VGrad, rank correlation 0.028Mixed
compare FMNIST attributions on MNIST model vs FMNIST modelGuided BackProp · FMNIST to MNIST modelrank correlation: SSIM 0.81, rank correlation 0.51Mixed
Pattern Attribution on out-of-domain VGG-16PAttribution · FMNIST to VGG-16SSIM: 0.77 -> 0.12Hurt
normal vs spurious model human ratingshuman study · Likert recommendation, Gradientmedian Likert score: 4 -> 2Hurt
normal vs spurious model human ratingshuman study · Likert recommendation, Integrated Gradientsmedian Likert score: 3 -> 3No clear effect
mislabelled training examples shown to usershuman study · Likert recommendation, Gradientmedian Likert score: 4 -> 1Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy