Research map / Label-efficient and robust learning / Noisy labels
Meta Label Correction for Noisy Label Learning
Read the paperCheck 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 meta-learned label correction instead of MW-Net instance reweighting | Noisy-label learning · CIFAR-10 (mean across noise types and levels) | accuracy: 65.12 -> 86.81 | Helped |
| Use meta-learned label correction instead of GLC | Noisy-label learning · CIFAR-100 (mean across noise types and levels) | accuracy: 50.50 -> 53.68 | Helped |
| Use meta-learned label correction instead of GLC | Noisy-label learning · AG (mean across noise types and levels) | accuracy: 83.88 -> 85.27 | Helped |
| Use meta-learned label correction instead of GLC | Noisy-label learning · Yelp-5 (mean across noise types and levels) | accuracy: 60.12 -> 62.61 | Helped |
| Use meta-learned label correction instead of GLC | Noisy-label learning · Amazon-5 (mean across noise types and levels) | accuracy: 60.31 -> 61.21 | Helped |
| Use meta-learned label correction instead of GLC | Noisy-label learning · Yahoo (mean across noise types and levels) | accuracy: 68.03 -> 73.72 | Helped |
| Use meta-learned label correction instead of GLC | Noisy-label learning · Clothing1M test set | accuracy: 73.69 -> 75.78 | Helped |
| Use larger k-step look-ahead for the meta-gradient (ablation) | MLC · High-noise settings (noise level more than 0.6) | test accuracy | Helped |
| Vary k-step look-ahead for the meta-gradient (ablation) | MLC · Low-noise settings | test accuracy | No clear effect |
| Replace label correction with instance reweighting | Noisy-label learning · Increasing uniform noise levels | test accuracy | Hurt |
Closest papers by meaning
- Revisiting Interpolation for Noisy Label Correction (AAAI 2025 · similarity 0.90)
- Adaptive Integration of Partial Label Learning and Negative Learning for Enhanced Noisy Label Learning (AAAI 2024 · similarity 0.89)
- Learning from Noisy Labels via Self-Taught On-the-Fly Meta Loss Rescaling (AAAI 2025 · similarity 0.88)
- Mitigating Label Noise through Data Ambiguation (AAAI 2024 · similarity 0.87)
- Handling Label Noise via Instance-Level Difficulty Modeling and Dynamic Optimization (NeurIPS 2025 · similarity 0.87)
- NoiseGPT: Label Noise Detection and Rectification through Probability Curvature (NeurIPS 2024 · similarity 0.87)
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.