Research map / Label-efficient and robust learning / Noisy labels
Jo-SRC: A Contrastive Approach for Combating Noisy Labels
Check 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 Jo-SRC instead of JoCoR | 7-layer DNN · CIFAR100N-C Symmetry-20% | accuracy: 53.01 ± 0.04 -> 58.15 ± 0.14 | Helped |
| Use Jo-SRC instead of JoCoR | 7-layer DNN · CIFAR100N-C Symmetry-50% | accuracy: 43.49 ± 0.46 -> 51.26 ± 0.11 | Helped |
| Use Jo-SRC instead of JoCoR | 7-layer DNN · CIFAR100N-C Symmetry-80% | accuracy: 15.49 ± 0.98 -> 23.80 ± 0.05 | Helped |
| Use Jo-SRC instead of JoCoR | 7-layer DNN · CIFAR100N-C Asymmetry-40% | accuracy: 32.70 ± 0.35 -> 38.52 ± 0.20 | Helped |
| Use Jo-SRC instead of Co-teaching | 7-layer DNN · CIFAR80N-O Symmetry-20% | accuracy: 60.38 ± 0.22 -> 65.83 ± 0.13 | Helped |
| Use Jo-SRC instead of Co-teaching | 7-layer DNN · CIFAR80N-O Symmetry-50% | accuracy: 52.42 ± 0.51 -> 58.51 ± 0.08 | Helped |
| Use Jo-SRC instead of Co-teaching | 7-layer DNN · CIFAR80N-O Symmetry-80% | accuracy: 16.59 ± 0.27 -> 29.76 ± 0.09 | Helped |
| Use Jo-SRC instead of Co-teaching+ | 7-layer DNN · CIFAR80N-O Asymmetry-40% | accuracy: 43.01 ± 0.59 -> 53.03 ± 0.25 | Helped |
| Use Jo-SRC instead of JoCoR | ResNet-18 · Clothing1M | accuracy: 70.30 -> 71.78 | Helped |
| Use Jo-SRC instead of DivideMix | ResNet-50 · Clothing1M | accuracy: 74.76 -> 75.93 | Helped |
| Use Jo-SRC instead of DeepSelf | ResNet-50 · Food101N | accuracy: 85.11 -> 86.66 | Helped |
| Add ID noisy samples with mean-teacher relabeling (ablation) | Jo-SRC-C · CIFAR80N-O Symmetry-20% | accuracy: 57.12 ± 0.33 -> 61.32 ± 0.18 | Helped |
Closest papers by meaning
- Combating Noisy Labels by Agreement: A Joint Training Method with Co-Regularization (CVPR 2020 · similarity 0.90)
- Coresets for Robust Training of Deep Neural Networks against Noisy Labels (NeurIPS 2020 · similarity 0.89)
- FINE Samples for Learning with Noisy Labels (NeurIPS 2021 · similarity 0.89)
- Mitigating Memorization of Noisy Labels via Regularization between Representations (ICLR 2023 · similarity 0.87)
- Handling Label Noise via Instance-Level Difficulty Modeling and Dynamic Optimization (NeurIPS 2025 · similarity 0.87)
- Enhancing Noise-Robust Losses for Large-Scale Noisy Data Learning (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.