Research map / Learning theory and algorithms / Clustering
Reconsidering Representation Alignment for Multi-View Clustering
Daniel J. Trosten, Sigurd Lokse, Robert Jenssen, Michael Kampffmeyer
Check your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| Replace adversarial distribution alignment with no alignment in SiMVC | Multi-view clustering · E-MNIST | ACC: 66.8 -> 86.2 | Improved |
| Add selective contrastive alignment to SiMVC | Multi-view clustering · CCV | ACC: 14.4 -> 29.5 | Improved |
| Add selective contrastive alignment to SiMVC | Multi-view clustering · E-MNIST | ACC: 86.2 -> 95.5 | Improved |
| Add selective contrastive alignment to SiMVC | Multi-view clustering · E-FMNIST | NMI: 50.7 -> 52.3 | Mixed |
| Remove adaptive weighting from the contrastive loss (ablation) | CoMVC · E-MNIST | ACC: 95.5 -> 94.7 | Hurt |
| Remove adaptive weighting from the contrastive loss (ablation) | CoMVC · VOC | ACC: 61.9 -> 55.3 | Hurt |
| Remove cluster-aware negative sampling (ablation) | CoMVC · E-MNIST | ACC: 95.5 -> 87.5 | Hurt |
| Remove cluster-aware negative sampling (ablation) | CoMVC · VOC | ACC: 61.9 -> 58.5 | Hurt |
| Add a projection head before cosine similarity (ablation) | CoMVC · E-MNIST | ACC: 95.5 -> 77.1 | Hurt |
| Add a projection head before cosine similarity (ablation) | CoMVC · VOC | ACC: 61.9 -> 62.4 | Mixed |
| Drop clustering-loss terms and retain only L3 (ablation) | SiMVC · E-MNIST | ACC: 86.2 -> 75.2 | Hurt |
| Drop clustering-loss terms and retain only L3 (ablation) | CoMVC · E-MNIST | ACC: 95.5 -> 72.8 | Hurt |
Closest papers by meaning
- Partial Contrastive Learning for Partially View-Aligned Multi-View Clustering (TPAMI 2026 · similarity 0.92)
- Relationship Alignment for View-aware Multi-view Clustering (ICLR 2026 · similarity 0.92)
- On the Effects of Self-Supervision and Contrastive Alignment in Deep Multi-View Clustering (CVPR 2023 · similarity 0.91)
- Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering (CVPR 2025 · similarity 0.91)
- EASEMVC:Efficient Dual Selection Mechanism for Deep Multi-View Clustering (CVPR 2025 · similarity 0.90)
- Incomplete Contrastive Multi-View Clustering with High-Confidence Guiding (AAAI 2024 · similarity 0.89)
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.
