Research map / Label-efficient and robust learning / OOD and anomaly detection
Supervision Adaptation Balancing In-Distribution Generalization and Out-of-Distribution Detection
Zhilin Zhao, Longbing Cao, Kun-Yu Lin
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 |
|---|---|---|---|
| add supervision adaptation (SA) with OOD training samples | Baseline ResNet18 · CIFAR10 | AUROC (%): 91.4 -> 98.9 | Improved |
| add SA | Baseline ResNet18 · CIFAR100 | AUROC (%): 80.1 -> 96.3 | Improved |
| use SA instead of FIG | FIG · CIFAR10 | AUROC (%): 97.1 -> 98.9 | Improved |
| add SA to MobileNetV2 | Baseline MobileNetV2 · CIFAR10 | AUROC (%): 87.9 -> 97.6 | Improved |
| add SA (ID accuracy) | Baseline ResNet18 · CIFAR100 | accuracy (%): 77.7 -> 78.6 | Improved |
| add outlier exposure (OE) with manual labels | Baseline ResNet18 · CIFAR10 | accuracy (%): 95.0 -> 94.9 | Hurt |
| add domain adaptation method DANN | Baseline ResNet18 · CIFAR10 | AUROC (%): 91.4 -> 87.1 | Hurt |
| add label smoothing (LS) | Baseline ResNet18 · CIFAR10 | AUROC (%): 91.4 -> 86.7 | Hurt |
| train SA with LSUN instead of TinyImageNet as OOD data | SA ResNet18 · CIFAR10 | AUROC (%): 98.1 -> 98.9 | Improved |
| train SA with CelebA instead of TinyImageNet as OOD data | SA ResNet18 · CIFAR10 | AUROC (%): 98.1 -> 96.1 | Hurt |
| increase component parameter epsilon (OOD proportion) (ablation) | SA ResNet18 · CIFAR10 | AUROC (%) | Improved |
| add MSMI to MBCE-only training (full SA vs MBCE) (ablation) | SA ResNet18 · CIFAR100 | accuracy (%) | Improved |
Closest papers by meaning
- Meta OOD Learning For Continuously Adaptive OOD Detection (ICCV 2023 · similarity 0.88)
- OoD-Bench: Quantifying and Understanding Two Dimensions of Out-of-Distribution Generalization (CVPR 2022 · similarity 0.88)
- Rethinking Out-of-Distribution Detection on Imbalanced Data Distribution (NeurIPS 2024 · similarity 0.88)
- Splitting & Integrating: Out-of-Distribution Detection via Adversarial Gradient Attribution (ICML 2025 · similarity 0.87)
- Can We Ignore Labels in Out of Distribution Detection? (ICLR 2025 · similarity 0.87)
- Beyond Clean Training Data: A Versatile and Model-Agnostic Framework for Out-of-Distribution Detection with Contaminated Training Data (CVPR 2025 · 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.
