Multimodal Federated Learning (MMFL) enables privacy-preserving collaborative training, but real-world clinical applications often suffer from within-modality missingness caused by sensor intermittency or irregular sampling. Existing methods implicitly represent unobserved data via architectural alignment or missing embeddings, often failing to recover the true distribution and yielding sub-optimal performance. We propose CondI, a federated framework explicitly addressing this missingness using conditional diffusion models. CondI employs a two-phase training pipeline: first, imputing unobserved temporal components using available multimodal context and conditional embeddings; second, optimizing modality-specific extractors and joint embedding spaces. During inference, imputed raw data pass through trained extractors to generate robust features, providing a holistic representation for downstream tasks. Explicit data imputation ensures models operate on complete semantic structures, significantly enhancing resilience against severe data incompleteness. Experiments on three clinical datasets (PTB-XL, SLEEP-EDF, MIMIC-IV) demonstrate CondI achieves comparable results to state-of-the-art baselines. Code: https://github.com/ZhengWugeng/CondI
@article{arxiv.2604.23112,
title = {Conditional Imputation for Within-Modality Missingness in Multi-Modal Federated Learning},
author = {Wugeng Zheng and Ziwen Kan and Katie Wang and Chen Chen and Song Wang},
journal= {arXiv preprint arXiv:2604.23112},
year = {2026}
}
Comments
Wugeng Zheng and Ziwen Kan contributed equally to this work. Song Wang is the corresponding author. Accepted to FedVision 2026