Research map / Label-efficient and robust learning / OOD and anomaly detection
Revealing the Distributional Vulnerability of Discriminators by Implicit Generators
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 |
|---|---|---|---|
| fine-tune discriminator on its own implicit-generator OOD samples (FIG) | ResNet18 baseline detector · SVHN, CIFAR10, CIFAR100 | AUROC (%): 1.22% to 23.49% improvement | Improved |
| replace Gaussian-noise OOD samples with implicit-generator samples | GS fine-tuning · SVHN, CIFAR10, CIFAR100 | AUROC (%): 5.69% improvement | Improved |
| fine-tune with FIG on higher-resolution data | ResNet18 · MiniImageNet | AUROC (%): 7.87% average AUROC improvement | Improved |
| replace DCC* with FIG | ResNet18 · MiniImageNet | AUROC (%): 81.4 -> 84.4 | Improved |
| replace DCC* with FIG | ResNet18 · CIFAR100 | AUROC (%): 88.7 -> 89.3 | Improved |
| add FIG fine-tuning to ODIN detector | ResNet18 · CIFAR100 | AUROC (%): 84.6 -> 90.2 | Improved |
| add FIG fine-tuning to DRF detector | ResNet18 · CIFAR100 | AUROC (%): 71.2 -> 74.4 | Improved |
| use FIG instead of JCL / DCC* for balance of ID accuracy and OOD detection | ResNet18 · CIFAR10 | harmonic mean: 46.8 -> 47.5 | Improved |
| use adversarial samples (AD) instead of standard training | ResNet18 · CIFAR100 | harmonic mean: 39.9 -> 35.9 | Hurt |
| raise low-entropy percentage K above 0.1 (ablation) | FIG · SVHN, CIFAR10, CIFAR100 | AUROC (%) | Hurt |
| raise low-entropy percentage K from 0 (no generated samples) up to 0.1 (ablation) | FIG · SVHN, CIFAR10, CIFAR100 | AUROC (%) | Improved |
Closest papers by meaning
- Your Out-of-Distribution Detection Method is Not Robust! (NeurIPS 2022 · similarity 0.86)
- Neural Mean Discrepancy for Efficient Out-of-Distribution Detection (CVPR 2022 · similarity 0.85)
- Adversarially Robust Out-of-Distribution Detection Using Lyapunov-Stabilized Embeddings (ICLR 2025 · similarity 0.85)
- Expecting The Unexpected: Towards Broad Out-Of-Distribution Detection (NeurIPS 2024 · similarity 0.84)
- Discriminability-Driven Channel Selection for Out-of-Distribution Detection (CVPR 2024 · similarity 0.84)
- Mysteries of the Deep: Role of Intermediate Representations in Out of Distribution Detection (NeurIPS 2025 · similarity 0.84)
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.
