# Wasserstein-Aligned Hyperbolic Multi-View Clustering

Source: [Wasserstein-Aligned Hyperbolic Multi-View Clustering](<https://www.sota2.com/research/paper/693a4feb6cfd488761a548cc>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-12-12T05:02:33.257Z

Authors: Rui Wang, Yuting Jiang, Xiaoqing Luo, Xiao-Jun Wu, Nicu Sebe, Ziheng Chen

Published: 2025-12-10T07:56:19.000Z

## Abstract

Multi-view clustering (MVC) aims to uncover the latent structure of multi-view data by learning view-common and view-specific information. Although recent studies have explored hyperbolic representations for better tackling the representation gap between different views, they focus primarily on instance-level alignment and neglect global semantic consistency, rendering them vulnerable to view-specific information (\\textit{e.g.}, noise and cross-view discrepancies). To this end, this paper proposes a novel Wasserstein-Aligned Hyperbolic (WAH) framework for multi-view clustering. Specifically, our method exploits a view-specific hyperbolic encoder for each view to embed features into the Lorentz manifold for hierarchical semantic modeling. Whereafter, a global semantic loss based on the hyperbolic sliced-Wasserstein distance is introduced to align manifold distributions across views. This is followed by soft cluster assignments to encourage cross-view semantic consistency. Extensive experiments on multiple benchmarking datasets show that our method can achieve SOTA clustering performance.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Multi-view Clustering on MNIST-USPS (V=2, N=5000)](<https://www.sota2.com/research/sota/multi-view-clustering-on-mnist-usps-v-2-n-5000.md>): Multi-view Clustering; MNIST-USPS (V=2, N=5000); ACC: 99.88.

- [Multi-view Clustering on COIL-10 V=3, N=720](<https://www.sota2.com/research/sota/multi-view-clustering-on-coil-10-v-3-n-720.md>): Multi-view Clustering; COIL-10 V=3, N=720; ACC: 99.91.

- [Multi-view Clustering on Scene-15 (V=3, N=4485)](<https://www.sota2.com/research/sota/multi-view-clustering-on-scene-15-v-3-n-4485.md>): Multi-view Clustering; Scene-15 (V=3, N=4485); ACC: 74.38.

- [Multi-view Clustering on Amazon V=4 (N=4790)](<https://www.sota2.com/research/sota/multi-view-clustering-on-amazon-v-4-n-4790.md>): Multi-view Clustering; Amazon V=4 (N=4790); ACC: 99.88.

- [Multi-view Clustering on Fashion V=3 N=10000](<https://www.sota2.com/research/sota/multi-view-clustering-on-fashion-v-3-n-10000.md>): Multi-view Clustering; Fashion V=3 N=10000; ACC: 99.74.

- [Multi-view Clustering on YoutubeVideo V=3 (N=101499)](<https://www.sota2.com/research/sota/multi-view-clustering-on-youtubevideo-v-3-n-101499.md>): Multi-view Clustering; YoutubeVideo V=3 (N=101499); ACC: 29.4.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2512.09402>)

[PDF](<https://arxiv.org/pdf/2512.09402.pdf>)
