SPORT: Structure-Aware Prototype Disentanglement for Incomplete Multi-View Clustering

2026-07-11

AI Summary

This work addresses critical limitations in existing incomplete multi-view clustering methods, which often neglect view-specific information, lack explicit modeling of cluster-level structures, and perform imputation without preserving local geometric relationships, leading to inaccurate representation recovery. To overcome these issues, the authors propose SPORT, a novel framework that explicitly decouples cluster prototypes into shared and view-specific components under an orthogonality constraint. SPORT further introduces a cluster-structure-aware contrastive learning mechanism to maintain cross-view semantic consistency and integrates global prototype guidance with local manifold structure for hybrid missing-view imputation. Extensive experiments on six benchmark datasets demonstrate that SPORT consistently outperforms state-of-the-art methods across various missing rates, achieving superior clustering performance.

Abstract

Prototype-based Incomplete Multi-view Clustering has recently attracted increasing attention by exploiting prototypes as semantic anchors for missing-view imputation. However, existing approaches are still limited in three aspects. First, they typically focus on enforcing cross-view prototype consistency, while ignoring view-specific information embedded in prototypes, thus limiting multi-view expressiveness. Second, most methods rely on instance-level contrastive learning that only aligns paired samples across views, failing to preserve cluster-level relational structures. Third, missing-view imputation is usually performed using global prototypes alone, without considering local geometric neighborhood structures, leading to inaccurate recovery of missing representations. To address these limitations, we propose a novel framework termed Structure-aware PrOtotype disentanglement foR incomplete multi-view clusTering (SPORT), which explicitly disentangles shared and view-specific components of prototypes while preserving cluster-level relational structures. Specifically, we decouple prototypes into orthogonal shared and view-specific components, aligning only shared components to capture consensus semantics while de-correlating view-specific components to preserve complementary information. Meanwhile, a structure-aware contrastive learning mechanism is incorporated to explicitly model cluster-level relationships during cross-view representation learning. Furthermore, a hybrid imputation strategy integrates global prototype matching with local neighborhood matching, enabling joint exploitation of semantic prototypes and manifold structures for missing-view recovery. Extensive experiments on six benchmark datasets show that SPORT achieves superior performance over state-of-the-art methods under various missing rates.

Problem

• Incomplete Multi-view Clustering
• Prototype Disentanglement
• View-specific Information
• Cluster-level Structure
• Missing-view Imputation

Innovation

• prototype disentanglement
• structure-aware contrastive learning
• incomplete multi-view clustering
• hybrid imputation
• cluster-level relational structure