Research map / Label-efficient and robust learning / Recognition and long-tail
State-Relabeling Adversarial Active Learning
Check 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 SRAAL rather than VAAL for active sample selection | active learning · CIFAR-100 | accuracy | Helped |
| use SRAAL rather than LL4AL for active sample selection | active learning · CIFAR-10 | accuracy | Mixed |
| use SRAAL rather than prior methods for active sample selection | active learning · Caltech-101 | accuracy | Helped |
| use SRAAL rather than VAAL for sample selection | semantic segmentation active learning · Cityscapes | mean IoU | Helped |
| initialize the labeled pool with k-center rather than random sampling | SRAAL · CIFAR-10 | accuracy | Helped |
| replace the proposed uncertainty score with entropy (ablation) | SRAAL · CIFAR-100, sampling ratios 20%–40% | accuracy: 55.0 -> 54.0; 59.1 -> 58.2; 62.3 -> 61.1; 65.0 -> 64.5; 66.9 -> 65.7 | Hurt |
| replace the proposed uncertainty score with standard deviation (ablation) | SRAAL · CIFAR-100, sampling ratios 20%–40% | accuracy: 55.0 -> 54.1; 59.1 -> 57.0; 62.3 -> 59.4; 65.0 -> 63.3; 66.9 -> 64.1 | Hurt |
| add online uncertainty state relabeling instead of using binary unlabeled-state labels (ablation) | SRAAL · CIFAR-100 | accuracy | Helped |
| add the supervised target learner and annotation-embedded representation (ablation) | SRAAL · CIFAR-100 | accuracy | Helped |
| add both online uncertainty state relabeling and the supervised target learner (ablation) | SRAAL · CIFAR-100 | accuracy | Helped |
Closest papers by meaning
- Multi-Classifier Adversarial Optimization for Active Learning (AAAI 2023 · similarity 0.88)
- Annealing Self-Distillation Rectification Improves Adversarial Training (ICLR 2024 · similarity 0.86)
- Soften to Defend: Towards Adversarial Robustness via Self-Guided Label Refinement (CVPR 2024 · similarity 0.85)
- Towards Cost-Effective Learning: A Synergy of Semi-Supervised and Active Learning (CVPR 2025 · similarity 0.84)
- Improved Algorithm for Deep Active Learning under Imbalance via Optimal Separation (ICML 2025 · similarity 0.84)
- Inconsistency-Based Data-Centric Active Open-Set Annotation (AAAI 2024 · 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.