---
title: "Instance-Aligned Semantic Reconstruction for Incomplete Multi-View Clustering"
authors:
  - "Weiqing Yan"
  - "Yongteng Du"
  - "Peng Song"
  - "Chang Tang"
doi: "10.24963/ijcai.2026/572"
url: "https://ijcai.org/proceedings/2026/572"
pdf: "https://ijcai.org/proceedings/2026/0572.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5137-5145"
track: "Main Track"
keywords:
  - "Machine Learning: Clustering"
  - "Machine Learning: Multi-view learning"
  - "Machine Learning: Deep learning architectures"
---

# Instance-Aligned Semantic Reconstruction for Incomplete Multi-View Clustering

Weiqing Yan, Yongteng Du, Peng Song, Chang Tang

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5137-5145. DOI: [10.24963/ijcai.2026/572](https://doi.org/10.24963/ijcai.2026/572)

## Abstract

Incomplete multi-view clustering (IMVC) aims to exploit complementary information from multiple views with missing observations. Recent diffusion-based approaches have shown promise for view completion; however, they often fail to capture instance-aligned global correlations across views and suffer from inefficient inference and loosely coupled optimization. In this paper, we propose IASR, an Instance-Aligned Semantic Reconstruction framework for IMVC. IASR formulates missing-view recovery as a cross-view token alignment generator, in which noisy targets, observed views, and timestep embeddings are jointly represented as tokens and interact to capture long-range cross-view dependencies throughout the denoising trajectory. To stabilize representation learning under generative noise, we further introduce a stable–active dual-encoder representation architecture with a generative-adaptive contrastive learning strategy that tightly couples view completion and clustering. Extensive experiments on eight benchmark datasets demonstrate that IASR consistently outperforms state-of-the-art IMVC methods, especially under high missing-rate settings, while achieving improved inference efficiency.