Research map / Adversarial robustness and security / Certified and empirical robustness
Your Out-of-Distribution Detection Method is Not Robust!
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 |
|---|---|---|---|
| Use ATD instead of HAT (OpenMax) | OOD detection on CIFAR-10 · CIFAR-10, in and out attacked | AUROC: 0.415 -> 0.693 | Helped |
| Use ATD instead of HAT (MD) | OOD detection on CIFAR-100 · CIFAR-100, in and out attacked | AUROC: 0.363 -> 0.553 | Helped |
| Add a HAT robust feature extractor instead of a standard-trained feature extractor (ablation) | ATD on CIFAR-10 · CIFAR-10, in and out attacked | AUROC: 0.029 -> 0.693 | Helped |
| Add adversarial training to the discriminator (ablation) | ATD on CIFAR-10 · CIFAR-10, in and out attacked | AUROC: 0.511 -> 0.693 | Helped |
| Generate OOD features rather than images, using a feature extractor (ablation) | ATD on CIFAR-10 · CIFAR-10, in and out attacked | AUROC: 0.560 -> 0.693 | Helped |
| Do not adversarially attack generated samples during training (ablation) | ATD on CIFAR-10 · CIFAR-10, in and out attacked | AUROC: 0.546 -> 0.560 | Helped |
Closest papers by meaning
- Splitting & Integrating: Out-of-Distribution Detection via Adversarial Gradient Attribution (ICML 2025 · similarity 0.92)
- Leveraging Perturbation Robustness to Enhance Out-of-Distribution Detection (CVPR 2025 · similarity 0.92)
- Discriminability-Driven Channel Selection for Out-of-Distribution Detection (CVPR 2024 · similarity 0.89)
- Exploiting Discrepancy in Feature Statistic for Out-of-Distribution Detection (AAAI 2024 · similarity 0.89)
- The Best of Both Worlds: On the Dilemma of Out-of-distribution Detection (NeurIPS 2024 · similarity 0.89)
- Out-of-Distribution Detection with Prototypical Outlier Proxy (AAAI 2025 · similarity 0.89)
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.