---
title: "Mixture of Clustering Experts with Dual Consistency for Multi-View Clustering"
authors:
  - "Daidai Zhu"
  - "Yang Zhao"
  - "Dandan Ma"
  - "Ganchao Liu"
  - "Zhiyu Jiang"
doi: "10.24963/ijcai.2026/605"
url: "https://ijcai.org/proceedings/2026/605"
pdf: "https://ijcai.org/proceedings/2026/0605.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5431-5439"
track: "Main Track"
keywords:
  - "Machine Learning: Multi-view learning"
  - "Machine Learning: Self-supervised Learning"
  - "Machine Learning: Clustering"
---

# Mixture of Clustering Experts with Dual Consistency for Multi-View Clustering

Daidai Zhu, Yang Zhao, Dandan Ma, Ganchao Liu, Zhiyu Jiang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5431-5439. DOI: [10.24963/ijcai.2026/605](https://doi.org/10.24963/ijcai.2026/605)

## Abstract

Multi-view clustering aims to discover shared semantic structures from multiple complementary data views to improve clustering performance. However, most existing methods rely on a single clustering representation for each semantic cluster, which limits the ability to model complex cluster structures and diverse patterns in real-world data.
To address this limitation, we propose a Mixture of Clustering Experts with Dual Consistency (MoCE-DC). MoCE-DC introduces a set of learnable clustering experts that characterize the same semantic cluster from multiple perspectives within a unified prototype space. A gating mechanism is employed to adaptively select experts for each sample. MoCE-DC decomposes complex semantic clusters into multiple collaborative sub-structures, significantly enhances the ability to model intricate intra-cluster diversity.
To further ensure cross-view semantic consistency, we propose a dual-level alignment mechanism that enforces prediction consistency across views while guiding clustering assignments toward a more discriminative direction. In addition, a gating balance regularization strategy is introduced to mitigate expert collapse and promote balanced expert utilization. Extensive experiments on multiple public multi-view clustering benchmarks demonstrate that MoCE-DC outperforms state-of-the-art methods.