Research map / Label-efficient and robust learning / Noisy labels
Sample-wise Label Confidence Incorporation for Learning with Noisy Labels
Chanho Ahn, Kikyung Kim, Ji-won Baek, Jongin Lim, Seungju Han
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 |
|---|---|---|---|
| use two models (noise-robust and noise-free) instead of a single model with both losses (ablation) | label confidence framework · CIFAR-10 sym-80 | accuracy (%): 78.10 -> 81.49 | Improved |
| single model combining cross-entropy and robust loss (ablation) | GCE baseline · CIFAR-100 sym-80 | accuracy (%): 49.68 -> 40.57 | Hurt |
| add augmentation-invariant regularization and label correction to two-model framework (ablation) | Two models · CIFAR-10 sym-80 | accuracy (%): 81.49 -> 91.13 | Improved |
| add augmentation-invariant regularization and label correction to two-model framework (ablation) | Two models · CIFAR-100 sym-80 | accuracy (%): 52.22 -> 61.05 | Improved |
| apply label-confidence framework over GCE robust loss | GCE · CIFAR-10 sym-80 | accuracy (%): 79.19 -> 91.13 | Improved |
| vary cross-entropy influence m between -1 and 1 | label confidence framework · CIFAR-10/100 sym-80 | accuracy (%) | No clear effect |
| set m to 5, raising cross-entropy influence | label confidence framework · CIFAR-10/100 sym-80 | accuracy (%) | Hurt |
| use noise-free model rather than noise-robust model at inference | label confidence framework · CIFAR-10/100 sym-80 | test accuracy | Improved |
| apply method with ResNet-18 backbone vs sample selection methods | Clothing1M · Clothing1M (ResNet-18) | accuracy (%): 70.30 -> 72.97 | Improved |
Closest papers by meaning
- Robust Classification via Regression for Learning with Noisy Labels (ICLR 2024 · similarity 0.93)
- FINE Samples for Learning with Noisy Labels (NeurIPS 2021 · similarity 0.92)
- Two Wrongs Don’t Make a Right: Combating Confirmation Bias in Learning with Label Noise (AAAI 2023 · similarity 0.92)
- Mitigating Label Noise through Data Ambiguation (AAAI 2024 · similarity 0.91)
- On the Role of Label Noise in the Feature Learning Process (ICML 2025 · similarity 0.91)
- Regroup Median Loss for Combating Label Noise (AAAI 2024 · similarity 0.90)
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.
