Research map / Learning theory and algorithms / Clustering
Dual-Branch Representations with Dynamic Gated Fusion and Triple-Granularity Alignment for Deep Multi-View Clustering
Wenyuan Kong, Zhibin Gu, Bing Li
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 semantics encoding branch (ablation) | DREAM · UCI | ACC: 87.05 -> 95.90 | Improved |
| add structure encoding branch (ablation) | DREAM · UCI | ACC: 75.90 -> 95.90 | Improved |
| add structure encoding branch (ablation) | DREAM · BBC | ACC: 29.20 -> 90.07 | Improved |
| add semantics encoding branch (ablation) | DREAM · BBC | ACC: 91.53 -> 90.07 | Hurt |
| add gated fusion (vs simple averaging) (ablation) | DREAM · UCI | ACC: 82.70 -> 95.90 | Improved |
| add gated fusion (vs simple averaging) (ablation) | DREAM · HW | ACC: 92.85 -> 97.80 | Improved |
| add triple-granularity feature alignment (ablation) | DREAM · UCI | ACC: 88.35 -> 95.90 | Improved |
| add feature alignment (ablation) | DREAM · NGS | ACC: 62.40 -> 97.80 | Improved |
| use GCN encoder instead of ground-truth adjacency with linear layer (ablation) | DREAM · BBC | ACC: 51.24 -> 90.07 | Improved |
| use GCN encoder instead of ground-truth adjacency with linear layer (ablation) | DREAM · Yale | ACC: 77.58 -> 78.18 | No clear effect |
| use structural cues instead of semantic cues in gated fusion (ablation) | DREAM · NGS | ACC: 86.00 -> 97.80 | Improved |
| use cosine-similarity graph instead of Euclidean graph (ablation) | DREAM · BBC | ACC: 24.23 -> 90.07 | Improved |
Closest papers by meaning
- Reconsidering Representation Alignment for Multi-View Clustering (CVPR 2021 · similarity 0.91)
- Medusa: A Multi-Scale High-order Contrastive Dual-Diffusion Approach for Multi-View Clustering (CVPR 2025 · similarity 0.90)
- Geometry-Aware Variational Information Maximization for Deep Incomplete Multi-view Clustering (AAAI 2026 · similarity 0.90)
- EASEMVC:Efficient Dual Selection Mechanism for Deep Multi-View Clustering (CVPR 2025 · similarity 0.88)
- SURER: Structure-Adaptive Unified Graph Neural Network for Multi-View Clustering (AAAI 2024 · similarity 0.88)
- Enhanced then Progressive Fusion with View Graph for Multi-View Clustering (CVPR 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.
