Research map / Label-efficient and robust learning / Noisy labels
Confidence Scores Make Instance-dependent Label-noise Learning Possible
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 instance-level forward correction instead of Co-teaching | Noisy-label learning · Clothing1M | accuracy: 70.11 -> 73.35 | Helped |
| Use instance-level forward correction instead of LQ | Noisy-label learning · Clothing1M | accuracy: 67.65 -> 73.35 | Helped |
| Use instance-level forward correction instead of Forward correction | Noisy-label learning · Clothing1M | accuracy: 60.62 -> 73.35 | Helped |
| Use instance-level forward correction instead of MAE | Noisy-label learning · Clothing1M | accuracy: 60.02 -> 73.35 | Helped |
| Add zero-mean Gaussian noise to confidence scores, with standard deviation up to 0.6 and clipping to [0,1] (ablation) | ILFC · Synthetic dataset | test accuracy | No clear effect |
| Use ILFC instead of the baselines under high-level instance-dependent noise | Noisy-label learning · Synthetic dataset | test accuracy | Helped |
| Use ILFC instead of the baselines under high-level instance-dependent noise | Noisy-label learning · SVHN | test accuracy | Helped |
| Use ILFC instead of the baselines under high-level instance-dependent noise | Noisy-label learning · CIFAR10 | test accuracy | Helped |
Closest papers by meaning
- An Information Fusion Approach to Learning with Instance-Dependent Label Noise (ICLR 2022 · similarity 0.93)
- A Second-Order Approach to Learning With Instance-Dependent Label Noise (CVPR 2021 · similarity 0.93)
- Label Noise: Ignorance Is Bliss (NeurIPS 2024 · similarity 0.90)
- Learning Causal Transition Matrix for Instance-dependent Label Noise (AAAI 2025 · similarity 0.89)
- Handling Label Noise via Instance-Level Difficulty Modeling and Dynamic Optimization (NeurIPS 2025 · similarity 0.88)
- Learning the Latent Causal Structure for Modeling Label Noise (NeurIPS 2024 · similarity 0.88)
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.