Research map / Label-efficient and robust learning / Semi-supervised learning
Class-Aware Contrastive Semi-Supervised Learning
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 |
|---|---|---|---|
| Add CCSSL to FixMatch | FixMatch · CIFAR100 | Top-1 accuracy: +10.04%/ +3.99%/ +3.26% | Helped |
| Add CCSSL to FixMatch | FixMatch · STL10 | Top-1 accuracy: +14.62% | Helped |
| Add CCSSL to FixMatch | FixMatch · CIFAR10 | Top-1 accuracy: +4.64/ -0.07%/ -0.2% | Mixed |
| Add CCSSL to FixMatch | FixMatch · Semi-iNat 2021, from scratch | Top-1 accuracy: +9.8% | Helped |
| Add CCSSL to FixMatch | FixMatch · Semi-iNat 2021, from MoCo pretrain | Top-1 accuracy: +1.05% | Helped |
| Add image-wise contrastive learning without class-aware clustering (ablation) | FixMatch · CIFAR100 | Accuracy: 72.54 -> 72.37 | Hurt |
| Add image-wise contrastive learning without class-aware clustering (ablation) | FixMatch · Semi-iNat 2021 | Accuracy: 21.41 -> 27.73 | Helped |
| Add re-weighting to contrastive learning (ablation) | FixMatch · Semi-iNat 2021 | Accuracy: 27.73 -> 29.48 | Helped |
| Add class-aware clustering to contrastive learning (ablation) | FixMatch · CIFAR100 | Accuracy: 72.37 -> 75.89 | Helped |
| Add re-weighting to contrastive learning and class-aware clustering (ablation) | FixMatch · Semi-iNat 2021 | Accuracy: 30.49 -> 31.21 | Helped |
| Increase unlabeled-data ratio from μ=5 to μ=7 when T_push=0.9 (ablation) | CCSSL · Semi-iNat 2021 | Accuracy: 26.81 -> 27.46 | Helped |
| Increase unlabeled-data ratio from μ=5 to μ=7 when T_push=0 (ablation) | CCSSL · Semi-iNat 2021 | Accuracy: 25.33 -> 23.14 | Hurt |
Closest papers by meaning
- HyperMatch: Noise-Tolerant Semi-Supervised Learning via Relaxed Contrastive Constraint (CVPR 2023 · similarity 0.91)
- Continuous Contrastive Learning for Long-Tailed Semi-Supervised Recognition (NeurIPS 2024 · similarity 0.90)
- Contrastive Credibility Propagation for Reliable Semi-supervised Learning (AAAI 2024 · similarity 0.89)
- OwMatch: Conditional Self-Labeling with Consistency for Open-World Semi-Supervised Learning (NeurIPS 2024 · similarity 0.89)
- Bi-Level Optimization for Semi-Supervised Learning with Pseudo-Labeling (AAAI 2025 · similarity 0.88)
- CDMAD: Class-Distribution-Mismatch-Aware Debiasing for Class-Imbalanced Semi-Supervised Learning (CVPR 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.