Research map / Label-efficient and robust learning / Noisy labels
ProMeCD: Unifying Long-Tailed and Noisy Label Learning via White-Box Control
Yu Zhu
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 ProMeCD white-box PI control instead of FMW-Net | ResNet-32 · CIFAR-100-LT IF=100 (100 epochs) | top-1 accuracy: 42.32 -> 47.52 (+5.2%) | Improved |
| train ProMeCD for 200 instead of 100 epochs | ProMeCD · CIFAR-100-LT IF=100 | top-1 accuracy: 47.52 -> 53.12 | Improved |
| use ProMeCD instead of PaCo | ResNet-32 · CIFAR-100-LT IF=100 | top-1 accuracy: 52.00 -> 53.12 (+1.12%) | Improved |
| use ProMeCD instead of standard CE | ResNet-50 · iNaturalist 2018 | top-1 accuracy: 57.16 -> 72.94 | Improved |
| use ProMeCD instead of PaCo on iNaturalist | ResNet-50 · iNaturalist 2018 | top-1 accuracy: 73.20 -> 72.94 | Hurt |
| use ProMeCD instead of FMW-Net under flip noise | ResNet-32 · CIFAR-10-N Flip 40% | test accuracy: 68.80 -> 91.12 | Improved |
| use ProMeCD instead of FMW-Net (DBG) | ResNet-50 · mini WebVision | top-1 accuracy: 72.44 -> 73.28 | Improved |
| use ProMeCD instead of D-SINK | ResNet-18 · LTNLL CIFAR-100 IF=100 NR=60% (Sym.) | test accuracy: 30.42 -> 32.88 | Improved |
| use ProMeCD instead of D-SINK | ResNet-18 · LTNLL CIFAR-10 IF=100 NR=60% (Sym.) | test accuracy: 57.80 -> 60.15 | Improved |
| use ProMeCD instead of MW-Net | ResNet-32 · CIFAR-100-N Sym. 40% | test accuracy: 45.79 -> 65.41 | Improved |
Closest papers by meaning
- Foster Adaptivity and Balance in Learning with Noisy Labels (ECCV 2024 · similarity 0.87)
- Keep It on a Leash: Controllable Pseudo-label Generation Towards Realistic Long-Tailed Semi-Supervised Learning (NeurIPS 2025 · similarity 0.87)
- Disentangling Sampling and Labeling Bias for Learning in Large-output Spaces (ICML 2021 · similarity 0.86)
- Label Noise: Ignorance Is Bliss (NeurIPS 2024 · similarity 0.85)
- Harnessing Hierarchical Label Distribution Variations in Test Agnostic Long-tail Recognition (ICML 2024 · similarity 0.85)
- Rectifying Soft-Label Entangled Bias in Long-Tailed Dataset Distillation (NeurIPS 2025 · similarity 0.85)
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.
