{"paper":{"title":"Deep Incomplete Multi-View Clustering via Hierarchical Imputation and Alignment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Hierarchical imputation of missing cluster assignments via cross-view similarity followed by feature reconstruction from intra-cluster statistics allows accurate shared clustering from partially observed multi-view data.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jian Li, Lusi Li, Rui Ning, Yiming Du, Ziyu Wang","submitted_at":"2026-01-14T00:46:00Z","abstract_excerpt":"Incomplete multi-view clustering (IMVC) aims to discover shared cluster structures from multi-view data with partial observations. The core challenges lie in accurately imputing missing views without introducing bias, while maintaining semantic consistency across views and compactness within clusters. To address these challenges, we propose DIMVC-HIA, a novel deep IMVC framework that integrates hierarchical imputation and alignment with four key components: (1) view-specific autoencoders for latent feature extraction, coupled with a view-shared clustering predictor to produce soft cluster assi"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on benchmarks demonstrate that our framework achieves superior performance under varying levels of missingness.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That estimating missing cluster assignments from cross-view contrastive similarity and reconstructing features from intra-cluster statistics introduces no systematic bias and preserves semantic consistency.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"DIMVC-HIA is a deep method for clustering multi-view data with missing views by hierarchically imputing assignments and features then aligning clusters for consistency and compactness.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Hierarchical imputation of missing cluster assignments via cross-view similarity followed by feature reconstruction from intra-cluster statistics allows accurate shared clustering from partially observed multi-view data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"88df2c3613eb12fb48faf5b75d4a10e906ec6f1a82cc471f8690756d6f9557ab"},"source":{"id":"2601.09051","kind":"arxiv","version":1},"verdict":{"id":"554d4b7a-e896-4864-b98d-1f40040d2b24","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T15:19:29.995266Z","strongest_claim":"Experiments on benchmarks demonstrate that our framework achieves superior performance under varying levels of missingness.","one_line_summary":"DIMVC-HIA is a deep method for clustering multi-view data with missing views by hierarchically imputing assignments and features then aligning clusters for consistency and compactness.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That estimating missing cluster assignments from cross-view contrastive similarity and reconstructing features from intra-cluster statistics introduces no systematic bias and preserves semantic consistency.","pith_extraction_headline":"Hierarchical imputation of missing cluster assignments via cross-view similarity followed by feature reconstruction from intra-cluster statistics allows accurate shared clustering from partially observed multi-view data."},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"803c316b11e183ac046186bf78c1a68a03997a658fc41a518f074b9873769016"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}