Research map / Label-efficient and robust learning / Recognition and long-tail
Contrastive Clustering
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- and cluster-level contrastive learning together rather than either head alone (ablation) | CC · CIFAR-10 | NMI: 0.699 -> 0.705; 0.592 -> 0.705 | Helped |
| use instance- and cluster-level contrastive learning together rather than either head alone (ablation) | CC · ImageNet-10 | NMI: 0.838 -> 0.859; 0.850 -> 0.859 | Helped |
| use two independently augmented views instead of replacing one view with the raw image (ablation) | CC · CIFAR-10 | NMI: 0.630 -> 0.705 | Helped |
| use two independently augmented views instead of replacing one view with the raw image (ablation) | CC · ImageNet-10 | NMI: 0.852 -> 0.859 | No clear effect |
| add data augmentation rather than using two identical raw views (ablation) | CC · CIFAR-10 | NMI: 0.045 -> 0.705 | Helped |
| add data augmentation rather than using two identical raw views (ablation) | CC · ImageNet-10 | NMI: 0.063 -> 0.859 | Helped |
| replace ResNet34 with ResNet18 (ablation) | CC · CIFAR-10 | NMI: 0.705 -> 0.650 | Hurt |
| replace ResNet34 with ResNet50 (ablation) | CC · CIFAR-10 | NMI: 0.705 -> 0.663 | Hurt |
| replace ResNet34 with ResNet18 (ablation) | CC · ImageNet-10 | NMI: 0.859 -> 0.851 | No clear effect |
| replace ResNet34 with ResNet50 (ablation) | CC · ImageNet-10 | NMI: 0.859 -> 0.859 | No clear effect |
| use CC rather than PICA | CIFAR-10 · clustering | NMI: 0.591 -> 0.705 | Helped |
| use CC rather than PICA | CIFAR-100 · clustering | NMI: 0.310 -> 0.431 | Helped |
Closest papers by meaning
- GeCC: Generalized Contrastive Clustering with Domain Shifts Modeling (AAAI 2025 · similarity 0.89)
- Multi-view Granular-ball Contrastive Clustering (AAAI 2025 · similarity 0.88)
- THESAURUS: Contrastive Graph Clustering by Swapping Fused Gromov-Wasserstein Couplings (AAAI 2025 · similarity 0.88)
- Contrastive Learning is Spectral Clustering on Similarity Graph (ICLR 2024 · similarity 0.88)
- Deep Contrastive Graph Learning with Clustering-Oriented Guidance (AAAI 2024 · similarity 0.88)
- Contrastive Multi-view Subspace Clustering via Tensor Transformers Autoencoder (AAAI 2025 · 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.