Research map / Label-efficient and robust learning / Recognition and long-tail
Rethinking Interpretation: Input-Agnostic Saliency Mapping of Deep Visual Classifiers
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Generate concept visualizations from VGG-16 and evaluate them with the same classifier | Input-agnostic saliency · ImageNet (10 concepts) | M_score: 0.99 | Helped |
| Generate concept visualizations from ResNet-50 and evaluate them with the same classifier | Input-agnostic saliency · ImageNet (10 concepts) | M_score: 0.99 | Helped |
| Evaluate VGG-16-source visualizations with ResNet-50 instead of VGG-16 | Input-agnostic saliency · ImageNet (10 concepts) | M_score: 0.99 -> 0.71 | Hurt |
| Evaluate ResNet-50-source visualizations with VGG-16 instead of ResNet-50 | Input-agnostic saliency · ImageNet (10 concepts) | M_score: 0.99 -> 0.69 | Hurt |
| Apply input-agnostic saliency to compromised classifiers to visualize a target class | Compromised VGG-16 · ImageNet backdoor detection | trigger-pattern visibility | Helped |
Closest papers by meaning
- Sanity Checks for Saliency Maps (NeurIPS 2018 · similarity 0.91)
- DiffCAM: Data-Driven Saliency Maps by Capturing Feature Differences (CVPR 2025 · similarity 0.87)
- Towards Fine-Grained Interpretability: Counterfactual Explanations for Misclassification with Saliency Partition (CVPR 2025 · similarity 0.87)
- Explainable Saliency: Articulating Reasoning with Contextual Prioritization (CVPR 2025 · similarity 0.86)
- DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging (NeurIPS 2026 · similarity 0.86)
- Saliency-R1: Enforcing Interpretable and Faithful Vision-language Reasoning via Saliency-map Alignment Reward (CVPR 2026 · similarity 0.86)
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.