Research map / Label-efficient and robust learning / Noisy labels
Content Diversity-guided Ambiguity Mitigation for Open-Set Noisy Label Learning
Zhihao Zhou, Rui Li, Xueying Li
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 content diversity-guided semantic synchronization (CDgss) (ablation) | DivideMix · CIFAR-10 (10k OOD) | accuracy: 92.73 -> 94.66 | Improved |
| add CDgss (ablation) | DivideMix · TinyImageNet OOD (10k) | accuracy: 70.38 -> 77.09 | Improved |
| add region-sensitive distillation regularization (RSDR) on top of CDgss (ablation) | CDgAM · CIFAR-10 (10k OOD) | accuracy: 94.66 -> 95.82 | Improved |
| add RSDR on top of CDgss (ablation) | CDgAM · TinyImageNet OOD (10k) | accuracy: 77.09 -> 77.54 | Improved |
| add RSDR on top of CDgss (ablation) | CDgAM · TinyImageNet OOD (20k) | accuracy: 75.85 -> 77.33 | Improved |
| use GPT-4o class descriptions instead of random text (ablation) | CDgAM · CIFAR-10 (10k CIFAR-100 OOD) | accuracy: 93.86 -> 95.82 | Improved |
| use GPT-4o descriptions instead of class names only (ablation) | CDgAM · CIFAR-10 (10k CIFAR-100 OOD) | accuracy: 95.28 -> 95.82 | Improved |
| use GPT-4o instead of DeepSeek-V2 for descriptions (ablation) | CDgAM · CIFAR-10 (10k CIFAR-100 OOD) | accuracy: 95.65 -> 95.82 | Improved |
| increase number of generated descriptions L beyond 5 (ablation) | CDgAM · CIFAR-10/100 | accuracy | No clear effect |
| raise RSDR coefficient lambda beyond 1 (ablation) | CDgAM · CIFAR-10/100 | accuracy | Hurt |
| use CDgAM instead of SED | SED · CIFAR-100N sym-80% | accuracy: 38.15 -> 48.61 | Improved |
| use CDgAM instead of SED | SED · WebFG-496 Web-car | accuracy: 88.88 -> 90.89 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DivideMix — DivideMix: Learning with Noisy Labels as Semi-supervised Learning (ICLR 2020) · better on 6 of 6 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning with Open-world Noisy Data via Class-independent Margin in Dual Representation Space (AAAI 2025 · similarity 0.88)
- Unlocking the Power of Open Set: A New Perspective for Open-Set Noisy Label Learning (AAAI 2024 · similarity 0.86)
- NoiseGPT: Label Noise Detection and Rectification through Probability Curvature (NeurIPS 2024 · similarity 0.86)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.86)
- Theory-Inspired Deep Multi-View Multi-Label Learning with Incomplete Views and Noisy Labels (CVPR 2025 · similarity 0.86)
- Noisy Label Calibration for Multi-View Classification (AAAI 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.
