ORCA：面向多视图聚类的正交残差共识对齐方法
ORCA:Orthogonal Residual Consensus Alignment for Multi-View Clustering
AI 导读
研究者提出 ORCA（Orthogonal Residual Consensus Alignment）方法，解决深度多视图聚类中信息冗余累积与表征层级无差别对齐问题。该方法包含语义增量循环融合和共识校准层级对比对齐两个模块，在五个基准上对比十五种方法，在 100leaves 数据集上准确率超过最强基线 5.25%。消融实验验证了各模块有效性，代码已开源。
正文 · 原文
Abstract:Deep multi-view clustering often suffers from redundant information accumulation during fusion and indiscriminate alignment across representation levels. We propose Orthogonal Residual Consensus Alignment (ORCA) to address these challenges. Its semantic incremental recurrent fusion module extracts cross-view orthogonal residuals as non-redundant increments and adaptively weights recurrent states through divergence-guided gating. Its consensus-calibrated hierarchical contrastive alignment module uses the global consensus to weight contrastive alignment of view-level, pairwise, and recurrent representations. Together, these modules balance view-specific structure with cross-view consistency. Experiments on five benchmarks against fifteen competing methods demonstrate ORCA's effectiveness, including a 5.25 percent accuracy improvement over the strongest baseline on 100leaves. Ablation studies validate both modules. Code is available at this https URL.
| Comments: | 24 pages,5 figures,2 tables |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.16749 [cs.CV] |
| (or arXiv:2611.16749v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2611.16749 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lin Zhao [view email]
[v1]
Thu, 12 Nov 2026 13:00:39 UTC (4,371 KB)
来源：arXiv cs.AI · arxiv.org