FedTaste: Topology-Aware Structural Transfer for Multimodal Federated Learning with Missing Modalities

LAUA: Handling Missing Modalities and Unpaired Data in Multimodal Federated Learning

Multimodal federated learning enables multiple clients to collaboratively train models from distributed multimodal data while preserving data privacy. In realistic federated settings, multimodal samples are often missing or unpaired, and cross-modal heterogeneity across clients can hinder stable optimization. Many existing federated multimodal methods attempt to mitigate modality missingness by generating synthetic paired data through data augmentation or generative models. However, they typically rely on paired supervision or treat client updates uniformly, making them brittle under modality missingness and client-level variability. To address these challenges, we propose a federated multimodal learning framework (LAUA) for learning from a mixture of unimodal and multimodal clients. On clients, LAUA aligns representations in a shared variational latent space, where KL regularization yields a principled and lightweight confidence signal for estimating uncertainty. Unimodal clients learn transferable representations via self-supervised objectives, while multimodal clients additionally leverage task supervision and incorporate an internal distillation component to enhance cross-modal consistency and stabilize local optimization. On the server, LAUA performs uncertainty-weighted aggregation that adaptively down-weights unreliable client updates. Experiments on various datasets show that LAUA substantially mitigates performance degradation under modality missingness across retrieval and regression tasks, attaining up to 20% relative improvement in MRR for retrieval and up to 24.3% relative improvement in MSE for regression.

FedAMB: Adaptive Modality Balancing for Dominance-Robust Multimodal Federated Distillation

Multimodal Federated Learning under Dual-Axis Modality Missingness

Multimodal federated learning (FL) supports collaborative modeling in privacy-sensitive health-sensing and medical settings, but realistic deployments often exhibit dual-axis modality missingness: clients have different modality sets, and individual samples may contain only subsets of the modalities available locally. Existing methods typically address these two axes separately. We propose Flux, a multimodal federated learning framework built around two complementary components. First, modality-aware confidence tempering learns sample-specific confidence for each modality through mask-aware unimodal supervision and fuses the confidence estimates from observed modalities into a sample-adaptive temperature that adjusts predictive sharpness according to evidence quality and completeness. Second, gradient-decoupled private adaptation applies this temperature only to a client-private prediction pathway, while training the shared federated model with a standard, untempered objective. This enables sample-specific, client-local confidence adaptation without allowing confidence-dependent gradients to perturb shared representation learning. Across four multimodal datasets, Flux achieves the highest average macro-F1 on every dataset, outperforming the strongest dataset-specific baseline by 0.8~2.2 points and by 1.6 points on average. Additional analyses demonstrate favorable calibration, temperature sensitivity to both modality missingness and input corruption, and more stable shared optimization under private-only tempering. Our code is available at https://github.com/AdibaOrz/Flux.

FedADB: Class Anchor-Driven Dual-Branch Federated Learning for Mitigating Forgetting

FedADB, a Class Anchor-Driven Dual-Branch FL framework, a dual-branch collaborative training mechanism designed for clients that achieves significant improvements in both accuracy and convergence speed.

Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-Tuning

Federated fine-tuning of Multimodal Large Language Models (MLLMs) across distributed networks enables privacy-sensitive adaptation to evolving data streams, yet a fundamental obstacle prevents robust deployment in dynamic environments: catastrophic forgetting, wherein sequential task updates erase previously acquired knowledge across visual, linguistic, and cross-modal representations. Addressing this challenge is especially critical for autonomous networked AI operating in safety-sensitive domains, such as content moderation, where reliable retention of prior knowledge underpins system integrity. To overcome this, we propose Federated Continual Multimodal Learning (FedCMM), a framework that embeds continual-learning safeguards into the federated optimization loop at three complementary levels. At the parameter level, modality-aware elastic weight consolidation computes separate Fisher information matrices for the vision encoder, language backbone, and cross-modal projector, providing granular, asymmetry-aware protection against modality-specific forgetting. At the data level, each client trains a lightweight local generative replay module to synthesize raw-data-free embedding-level multimodal replay tuples without any raw data sharing. At the aggregation level, Task-similarity-aware gradient aggregation autonomously filters and reweights client updates by gradient cosine similarity, suppressing conflicting directions and stabilizing the global learning trajectory. Extensive experiments on two benchmarks demonstrate that FedCMM consistently outperforms recent baselines on accuracy and backward transfer, confirming that holistic, modality-aware optimization enables robust evolutive adaptation across heterogeneous networked AI deployments.

ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities

In this paper, we address the problem of multimodal federated learning with missing modality. Existing methods utilize an additional public dataset or perform naive feature synthesis that is based solely on the available modality. To address these limitations, we propose ProMoE-FL, a Prototype-conditioned Mixture-of-Experts framework for robust missing-modality feature synthesis in multimodal federated learning. ProMoE-FL builds a global client-aware prototype bank that captures clinically meaningful modality priors across institutions. Our Mixture of Experts is conditioned on these prototypes and modality indices to enable direction-aware expert routing for dynamically synthesizing missing features. We perform extensive quantitative and qualitative evaluations on four public chest X-ray datasets (MIMIC-CXR, NIH Open-I, PadChest, and CheXpert) and demonstrate that ProMoE-FL consistently outperforms state-of-the-art methods in both homogeneous as well as the more challenging heterogeneous settings.