Research map / Image generation / Autoregressive and latent generation
Explaining Image Classifiers With Multiscale Directional Image Representation
Stefan Kolek, Robert Windesheim, Hector Andrade-Loarca, Gitta Kutyniok, Ron Levie
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use a shearlet rather than wavelet representation for the mask (ablation) | WaveletX · ImageNet validation explanations | conciseness-preciseness scores | Improved |
| Add a spatial penalty to CartoonX (ablation) | CartoonX · ImageNet explanations | explanation clarity | Improved |
| Add a spatial penalty to CartoonX (ablation) | CartoonX · ImageNet explanations | irrelevant blurry areas | Improved |
| Remove smoothness constraints from the pixel mask (ablation) | Smooth pixel mask · ImageNet validation explanations | hallucination score | Hurt |
| Use a smooth pixel mask rather than an unconstrained pixel mask (ablation) | Pixel mask · ImageNet validation explanations | hallucination score | Improved |
| Replace shearlets with wavelets (ablation) | ShearletX · ImageNet validation explanations | conciseness-preciseness scores | Hurt |
| Use unconstrained pixel masks instead of ShearletX (ablation) | ShearletX · ImageNet validation explanations | CP-Entropy | Mixed |
| Use shearlet-domain masking instead of smooth pixel masking | Smooth pixel mask · ImageNet validation explanations | conciseness-preciseness scores | Improved |
| Use shearlet-domain masking instead of smooth pixel masking | Smooth pixel mask · ImageNet validation explanations | hallucination score | Hurt |
| Use shearlet-domain masking instead of unconstrained pixel masking | Pixel mask without smoothness constraints · ImageNet validation explanations | hallucination score | Improved |
| Use ShearletX rather than Smooth Mask | Smooth Mask · ImageNet (MobileNetV3Small) | explanation time: 11.61s -> 54.26s | Hurt |
Closest papers by meaning
- Explaining in Diffusion: Explaining a Classifier with Diffusion Semantics (CVPR 2025 · similarity 0.85)
- A Simple Interpretable Transformer for Fine-Grained Image Classification and Analysis (ICLR 2024 · similarity 0.84)
- Finer-CAM: Spotting the Difference Reveals Finer Details for Visual Explanation (CVPR 2025 · similarity 0.82)
- Interpretable Image Classification via Non-parametric Part Prototype Learning (CVPR 2025 · similarity 0.82)
- I-CEE: Tailoring Explanations of Image Classification Models to User Expertise (AAAI 2024 · similarity 0.82)
- Comparing the Decision-Making Mechanisms by Transformers and CNNs via Explanation Methods (CVPR 2024 · similarity 0.82)
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.
