{"ID":2856736,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.10534","arxiv_id":"2510.10534","title":"MCE: Towards a General Framework for Handling Missing Modalities under Imbalanced Missing Rates","abstract":"Multi-modal learning has made significant advances across diverse pattern recognition applications. However, handling missing modalities, especially under imbalanced missing rates, remains a major challenge. This imbalance triggers a vicious cycle: modalities with higher missing rates receive fewer updates, leading to inconsistent learning progress and representational degradation that further diminishes their contribution. Existing methods typically focus on global dataset-level balancing, often overlooking critical sample-level variations in modality utility and the underlying issue of degraded feature quality. We propose Modality Capability Enhancement (MCE) to tackle these limitations. MCE includes two synergistic components: i) Learning Capability Enhancement (LCE), which introduces multi-level factors to dynamically balance modality-specific learning progress, and ii) Representation Capability Enhancement (RCE), which improves feature semantics and robustness through subset prediction and cross-modal completion tasks. Comprehensive evaluations on four multi-modal benchmarks show that MCE consistently outperforms state-of-the-art methods under various missing configurations. The final published version is now available at https://doi.org/10.1016/j.patcog.2025.112591. Our code is available at https://github.com/byzhaoAI/MCE.","short_abstract":"Multi-modal learning has made significant advances across diverse pattern recognition applications. However, handling missing modalities, especially under imbalanced missing rates, remains a major challenge. This imbalance triggers a vicious cycle: modalities with higher missing rates receive fewer updates, leading to...","url_abs":"https://arxiv.org/abs/2510.10534","url_pdf":"https://arxiv.org/pdf/2510.10534v2","authors":"[\"Binyu Zhao\",\"Wei Zhang\",\"Zhaonian Zou\"]","published":"2025-10-12T10:26:18Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.LG\",\"cs.MM\"]","methods":"[]","has_code":false,"code_links":[{"ID":608377,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2856736,"paper_url":"https://arxiv.org/abs/2510.10534","paper_title":"MCE: Towards a General Framework for Handling Missing Modalities under Imbalanced Missing Rates","repo_url":"https://github.com/byzhaoAI/MCE","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}