Research map / Adversarial robustness and security / Black-box and transfer attacks
EMaP: Explainable AI with Manifold-based Perturbations
Minh Nhat Vu, Huy Quang Mai, My T. Thai
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 EMaP manifold-based perturbations in LIME | LIME · Multi-Domain Sentiment (4 review datasets) | precision and recall of explanations | Improved |
| use EMaP perturbations in LIME | LIME+ · MNIST | log-odds | No clear effect |
| use EMaP perturbations in LIME | LIME+ · Fashion-MNIST | log-odds | Improved |
| use EMaP perturbations in LIME | LIME · MNIST and Fashion-MNIST | infidelity | Improved |
| use EMaP perturbations instead of LIME or SHAP perturbations | LIME · MNIST, Fashion-MNIST, Communities and Crime, German Credit | discriminator TP/TN rates | Improved |
| use orthogonal perturbation instead of projection perturbation (ablation) | EMaP · real-world datasets | H0 and H1 Bottleneck distance | Improved |
| use EMaP instead of LIME in run time | LIME · MNIST | seconds per explanation: 0.763 -> 1.311 | Hurt |
Closest papers by meaning
- Mitigating Overthinking in Large Reasoning Models via Manifold Steering (NeurIPS 2025 · similarity 0.84)
- GNNX-BENCH: Unravelling the Utility of Perturbation-based GNN Explainers through In-depth Benchmarking (ICLR 2024 · similarity 0.84)
- Improving Perturbation-based Explanations by Understanding the Role of Uncertainty Calibration (NeurIPS 2025 · similarity 0.83)
- Faithful Model Explanations through Energy-Constrained Conformal Counterfactuals (AAAI 2024 · similarity 0.83)
- Faithful and Efficient Explanations for Neural Networks via Neural Tangent Kernel Surrogate Models (ICLR 2024 · similarity 0.83)
- Even-if Explanations: Formal Foundations, Priorities and Complexity (AAAI 2025 · similarity 0.83)
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.
