Research map / Label-efficient and robust learning / OOD and anomaly detection
Harnessing Out-Of-Distribution Examples via Augmenting Content and Style
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 |
|---|---|---|---|
| Add content-style disentanglement (ablation) | HOOD · OOD detection (CIFAR10 ID; LSUN OOD) | AUROC (%): 84.94\pm1.3 -> 86.12\pm0.6 | Helped |
| Add benign OOD data (ablation) | HOOD · Open-set SSL (CIFAR10, 400 labels/class) | accuracy (%): 83.32\pm2.0 -> 86.22\pm2.7 | Helped |
| Add malign OOD data (ablation) | HOOD · OOD detection (CIFAR10 ID; LSUN OOD) | AUROC (%): 82.50\pm2.2 -> 86.12\pm0.6 | Helped |
| Add benign and malign OOD augmentations (ablation) | HOOD · OOD detection (CIFAR10 ID; LSUN OOD) | AUROC (%): 80.83\pm0.8 -> 86.12\pm0.6 | Helped |
| Add content-style disentanglement (ablation) | HOOD · Open-set SSL (CIFAR10, 400 labels/class) | accuracy (%): 82.55\pm1.1 -> 86.22\pm2.7 | Helped |
| Add benign OOD data (ablation) | HOOD · Open-set DA (VisDA) | accuracy (%): 66.3\pm2.5 -> 72.4\pm1.6 | Helped |
| Add malign OOD data (ablation) | HOOD · Open-set SSL (CIFAR10, 400 labels/class) | accuracy (%): 85.40\pm0.8 -> 86.22\pm2.7 | Helped |
| Add content-style disentanglement (ablation) | HOOD · Open-set DA (VisDA) | accuracy (%): 64.6\pm0.9 -> 72.4\pm1.6 | Helped |
| Add benign OOD data (ablation) | HOOD · OOD detection (CIFAR10 ID; LSUN OOD) | AUROC (%): 85.95\pm1.8 -> 86.12\pm0.6 | Helped |
| Add malign OOD data (ablation) | HOOD · Open-set DA (VisDA) | accuracy (%): 71.8\pm1.2 -> 72.4\pm1.6 | Helped |
| Increase augmentation count from the paper's preferred four to too few or too many (ablation) | HOOD · CIFAR10 (augmentation-number analysis) | performance | Hurt |
| Use HOOD instead of OpenMatch | Open-set SSL · CIFAR10, 400 labels/class; Clean Acc. | accuracy (%): 87.92\pm1.0 -> 86.22\pm2.7 | Hurt |
Closest papers by meaning
- Image Background Serves as Good Proxy for Out-of-distribution Data (ICLR 2024 · similarity 0.89)
- The Best of Both Worlds: On the Dilemma of Out-of-distribution Detection (NeurIPS 2024 · similarity 0.89)
- Rethinking the Evaluation of Out-of-Distribution Detection: A Sorites Paradox (NeurIPS 2024 · similarity 0.87)
- ImageNet-OOD: Deciphering Modern Out-of-Distribution Detection Algorithms (ICLR 2024 · similarity 0.87)
- Out-of-Distribution Detection with Prototypical Outlier Proxy (AAAI 2025 · similarity 0.87)
- MultiOOD: Scaling Out-of-Distribution Detection for Multiple Modalities (NeurIPS 2024 · 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.