Research map / Learning theory and algorithms / Clustering
Wasserstein-Aligned Hyperbolic Multi-View Clustering
Rui Wang, Yuting Jiang, Xiaoqing Luo, Xiao-Jun Wu, Nicu Sebe, Ziheng 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 semantic contrastive loss Lsem (ablation) | WAH-MVC · Amazon | ACC: 27.57 -> 99.88 | Improved |
| add entropy balance regularizer Lreg (ablation) | WAH-MVC · Fashion | ACC: 71.38 -> 99.74 | Improved |
| add hyperbolic SWD alignment loss LHHSW (ablation) | WAH-MVC · YoutubeVideo | ACC: 25.49 -> 29.40 | Improved |
| add LHHSW alignment (ablation) | WAH-MVC · Amazon | ACC: 99.38 -> 99.88 | Improved |
| replace hyperbolic instance-level contrastive loss with HHSW distribution alignment (ablation) | WAH-MVC · Scene-15 | ACC: 72.27 -> 74.38 | Improved |
| replace instance-level contrastive alignment with HHSW (ablation) | WAH-MVC · YoutubeVideo | ACC: 22.45 -> 29.40 | Improved |
| replace instance-level contrastive alignment with HHSW (ablation) | WAH-MVC · YoutubeVideo | NMI: 23.19 -> 22.74 | Hurt |
| use Busemann (horospherical) projection instead of geodesic projection (ablation) | WAH-MVC · Scene-15 | ACC: 72.77 -> 74.38 | Improved |
| swap encoder to CVCL or CSOT encoder (ablation) | WAH-MVC · Amazon and Fashion | ACC: 99.36 / 99.62 on Amazon | No clear effect |
| use hyperbolic Lorentz alignment instead of Euclidean baselines | WAH-MVC · Scene-15 | ACC: 44.59 -> 74.38 vs best baseline CVCL | Improved |
| use WAH-MVC instead of GCFAgg | WAH-MVC · YoutubeVideo | NMI: 27.31 -> 22.74 | Hurt |
| temperature τ moderate (0.3 to 0.5) versus extreme values (ablation) | WAH-MVC · COIL10, Fashion, Scene, Amazon | ACC and NMI | Improved |
Closest papers by meaning
- Asymmetric Multi-View Clustering with Hyperbolic Uncertainty Modeling (ICML 2026 · similarity 0.92)
- MHCN: A Hyperbolic Neural Network Model for Multi-view Hierarchical Clustering (ICCV 2023 · similarity 0.91)
- Transport Barycenter-Guided Sample-to-Cluster Matching for Unaligned Multi-view Clustering (TPAMI 2026 · similarity 0.88)
- Reconsidering Representation Alignment for Multi-View Clustering (CVPR 2021 · similarity 0.88)
- Robust Multi-view Clustering against Imperfect Information (NeurIPS 2026 · similarity 0.88)
- Incomplete Multi-view Clustering via Hierarchical Semantic Alignment and Cooperative Completion (NeurIPS 2025 · similarity 0.87)
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.
