Research map / Label-efficient and robust learning / Semi-supervised learning
Hyperspherical Consistency Regularization
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 HCR to Self-Tuning | Self-Tuning · Stanford Cars, 15% labels | accuracy (%): 74.99±0.11 -> 78.76±0.08 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · FGVC Aircraft, 15% labels | accuracy (%): 66.68±0.17 -> 70.54±0.02 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · CUB-200-2011, 15% labels | accuracy (%): 64.79±0.06 -> 66.42±0.24 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · CIFAR-100, 2.5K labels | error rate (%): 24.16 -> 23.93 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · CIFAR-100, 400 labels | error rate (%): 47.17 -> 42.71 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · Stanford Cars fine-grained classification | accuracy (%): 92.33±0.10 -> 93.03±0.06 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · FGVC Aircraft fine-grained classification | accuracy (%): 88.96±0.21 -> 90.41±0.03 | Helped |
| Add HCR to Self-Tuning | Self-Tuning · CUB-200-2011 fine-grained classification | accuracy (%): 81.60±0.11 -> 82.63±0.19 | Helped |
| Add HCR to Co-learning | Co-learning · CIFAR-100 with symmetric noise | accuracy (%): 66.58±0.15 -> 70.27±0.32; 55.54±0.43 -> 59.93± 0.25; 35.45±0.79 -> 39.14±0.47 | Helped |
| Add HCR to Co-learning | Co-learning · CIFAR-100 with asymmetric noise | accuracy (%): 65.26±0.76 -> 68.85±0.22; 56.97±1.22 -> 61.94±0.17; 47.62±0.79 -> 50.29±0.69 | Helped |
| Add HCR to Co-learning | Co-learning · CIFAR-100 with instance-dependent noise | accuracy (%): 69.42±0.42 -> 70.03±0.31; 65.45±0.86 -> 66.89±0.41; 60.40±1.37 -> 62.91±0.84 | Helped |
Closest papers by meaning
- HYPO: Hyperspherical Out-Of-Distribution Generalization (ICLR 2024 · similarity 0.86)
- Bridging Mini-Batch and Asymptotic Analysis in Contrastive Learning: From InfoNCE to Kernel-Based Losses (ICML 2024 · similarity 0.86)
- Collapse-Proof Non-Contrastive Self-Supervised Learning (ICML 2025 · similarity 0.85)
- ConR: Contrastive Regularizer for Deep Imbalanced Regression (ICLR 2024 · similarity 0.85)
- Stabilizing Sample Similarity in Representation via Mitigating Random Consistency (ICML 2025 · similarity 0.84)
- Preventing Dimensional Collapse in Self-Supervised Learning via Orthogonality Regularization (NeurIPS 2024 · similarity 0.84)
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.