Research map / Learning theory and algorithms / Clustering
Explicit View-Labels Matter: A Multifacet Complementarity Study of Multi-View Clustering
Chuanxing Geng, Aiyang Han, Songcan Chen
Read the paperCheck 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 |
|---|---|---|---|
| add view-label prediction loss L_cla (ablation) | Baseline L_clu + L_rec · Caltech101-20 | ACC: 55.12 -> 67.34 | Improved |
| add instance-level contrast L_ins (ablation) | Baseline L_clu + L_rec · Caltech101-20 | ACC: 55.12 -> 55.64 | Improved |
| add variance loss L_var (ablation) | Baseline L_clu + L_rec · Caltech101-20 | ACC: 55.12 -> 56.16 | Improved |
| add L_cla on top of L_ins and L_var (ablation) | Baseline + L_ins + L_var · Caltech101-20 | ACC: 56.00 -> 73.77 | Improved |
| explicit view-label loss only (L_clu + L_cla) instead of implicit use via reconstruction (ablation) | Baseline L_clu + L_rec · Caltech101-20 | ACC: 55.12 -> 70.80 | Improved |
| add view-label prediction loss L_cla (ablation) | Baseline · Caltech-5V | ACC: 64.77 -> 76.86 | Improved |
| add instance-level contrast L_ins (ablation) | Baseline · Caltech-5V | ACC: 64.77 -> 78.46 | Improved |
| add variance loss L_var (ablation) | Baseline · Caltech-5V | ACC: 64.77 -> 74.57 | Improved |
| add L_cla on top of L_ins (ablation) | Baseline + L_ins · Caltech-5V | ACC: +5.05% in ACC, +3.27% in NMI, +5.35% in ARI | Improved |
| add view-label prediction loss to MFLVC (MFLVC+) | MFLVC · CCV | ACC: 0.312 -> 0.323 | Improved |
| add variance loss to MFLVC+ (MCMVC+) (ablation) | MFLVC+ · CCV | ACC: 0.323 -> 0.334 | Improved |
| add cluster-assignment enhancement (CE) (ablation) | MCMVC+ · CCV | ACC: 0.301 -> 0.334 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- COMPLETER — COMPLETER: Incomplete Multi-View Clustering via Contrastive Prediction (CVPR 2021) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Enhancing Multi-View Clustering: A Sufficient Information-Theoretic Approach for Consistency Acquisition and Redundancy Elimination (TPAMI 2026 · similarity 0.91)
- Relationship Alignment for View-aware Multi-view Clustering (ICLR 2026 · similarity 0.90)
- 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)
- Robust Multi-view Clustering against Imperfect Information (NeurIPS 2026 · similarity 0.89)
- Incomplete and Unpaired Multi-View Graph Clustering with Cross-View Feature Fusion (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.
