发表机构
College of Computer and Cyberspace Security, Hebei Normal University; Department of Statistics and Data Science, Southern University of Science and Technology; Key Laboratory of Tibetan Information Processing, Ministry of Education, Qinghai Normal University; School of Computer Science and Technology, Beijing Jiaotong University; State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（河北师范大学计算机与网络空间安全学院; 南方科技大学统计与数据科学系; 青海师范大学藏文信息处理教育部重点实验室; 北京交通大学计算机科学与技术学院; 中国科学院自动化研究所多模态人工智能系统国家重点实验室）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究不完全多视图聚类问题，提出SPORT框架，通过解缠原型组件、引入结构感知对比学习机制和混合插补策略，保留簇级关系结构，联合利用语义原型和流形结构，在多数据集实验中性能优于现有方法。
AI 中文摘要
基于原型的不完全多视图聚类通过利用原型作为缺失视图插补的语义锚点，近来受到越来越多关注。然而，现有方法在三个方面仍有局限。首先，它们通常专注于加强跨视图原型一致性，却忽略原型中嵌入的视图特定信息，限制了多视图表达性。其次，多数方法依赖实例级对比学习，仅对齐跨视图配对样本，无法保留簇级关系结构。第三，缺失视图插补通常仅使用全局原型，未考虑局部几何邻域结构，导致缺失表示恢复不准确。为解决这些局限，我们提出了一种名为SPORT的新框架，明确解缠原型的共享和视图特定组件，同时保留簇级关系结构。具体而言，我们将原型解耦为正交的共享和视图特定组件，仅对齐共享组件以捕获一致语义，同时使视图特定组件去相关以保留互补信息。此外，我们引入了一种结构感知对比学习机制，在跨视图表示学习期间明确建模簇级关系。还采用了混合插补策略，将全局原型匹配与局部邻域匹配相结合，实现联合利用语义原型和流形结构进行缺失视图恢复。在六个基准数据集上的大量实验表明 SPORT 在各种缺失率下均优于现有方法。
英文摘要
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.
