Learnijoy
ResearchAI Engineering & DevToolsAI Research

ProMoE-FL Enhances Multimodal Federated Learning with Missing Data

Aavash Chhetri, Bibek Niroula, Eduard Vazquez, Yash Raj Shrestha, Prashnna Gyawali, Loris Bazzani, Binod Bhattarai· July 9, 2026 View original

Summary

This paper introduces ProMoE-FL, a Prototype-conditioned Mixture-of-Experts framework for robust multimodal federated learning, specifically addressing missing modalities. It uses a global client-aware prototype bank and dynamic expert routing to synthesize missing features, outperforming state-of-the-art methods on chest X-ray datasets.

This research addresses a significant challenge in multimodal federated learning: dealing with missing data modalities. Existing approaches often rely on public datasets or simplistic feature synthesis, which can be inadequate. The authors propose ProMoE-FL, a novel framework that leverages a Prototype-conditioned Mixture-of-Experts architecture to robustly synthesize missing features. ProMoE-FL establishes a global prototype bank that is aware of client-specific characteristics, capturing clinically relevant modality priors across different institutions. This prototype bank then conditions a Mixture of Experts, which uses modality indices to enable dynamic, direction-aware routing of experts. This mechanism allows for the intelligent synthesis of missing features. Extensive evaluations on four public chest X-ray datasets (MIMIC-CXR, NIH Open-I, PadChest, and CheXpert) demonstrate that ProMoE-FL consistently surpasses current state-of-the-art methods in both homogeneous and more complex heterogeneous settings.

Why it matters

This advancement is critical for deploying robust AI models in real-world scenarios, especially in healthcare, where data is often incomplete and distributed across multiple institutions, enabling more effective and privacy-preserving AI applications.

How to implement this in your domain

  1. 1Investigate ProMoE-FL for federated learning projects involving multimodal data with potential missing components.
  2. 2Pilot the framework in healthcare settings for diagnostic AI models that combine various imaging or patient data types.
  3. 3Collaborate with research teams to adapt and integrate this approach into existing federated learning pipelines.
  4. 4Assess the privacy implications and benefits of using client-aware prototype banks in distributed learning.

Who benefits

HealthcareMedical ImagingPharmaceuticalsLife Sciences

Key takeaways

  • ProMoE-FL offers a robust solution for multimodal federated learning with missing data.
  • It uses a prototype-conditioned Mixture-of-Experts for intelligent feature synthesis.
  • The framework builds a global client-aware prototype bank to capture modality priors.
  • ProMoE-FL outperforms existing methods, particularly in healthcare imaging applications.

Original post by Aavash Chhetri, Bibek Niroula, Eduard Vazquez, Yash Raj Shrestha, Prashnna Gyawali, Loris Bazzani, Binod Bhattarai

"arXiv:2607.06633v1 Announce Type: cross Abstract: 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.…"

View on X

Originally posted by Aavash Chhetri, Bibek Niroula, Eduard Vazquez, Yash Raj Shrestha, Prashnna Gyawali, Loris Bazzani, Binod Bhattarai on X · view source

Want to go deeper?

Turn these trends into skills with Learnijoy's hands-on AI & tech courses.

Explore courses

More in AI Engineering & DevTools

AI ResearchAI Engineering & DevTools

Transformers Learn Non-Invertible Modular Multiplication via Stratified Fourier Mechanisms.

This research investigates how small transformers learn modular integer multiplication over composite moduli, a non-invertible operation. It proposes the "monoid extension" theory, suggesting models partition input space into hierarchical algebraic regions where Fourier mechanisms apply, explaining how embeddings, attention, and local features contribute to the computation.

Zitong Andrew Chen, Junaid Hasan, Akhil Srinivasan, Hemkesh Bandi, Jarod AlperJul 9, 2026
AI Engineering & DevToolsAI Research

New Interpretable Model Handles Feature Interactions in Tabular Data.

This paper introduces Interaction Aware Interpretable Machine Learning (IAIML), a framework for tabular data that addresses the limitation of traditional interpretable models in capturing feature interactions. IAIML uses adaptive discretization, pairwise interaction scoring, and a partitioned explanation budget to achieve high accuracy while maintaining interpretability.

Srikumar KrishnamoorthyJul 9, 2026
AI ResearchAI Engineering & DevTools

Principles of Deep Feedforward ReLU Networks Unveiled.

This paper systematically studies the mechanisms of deep feedforward ReLU networks, generalizing principles from two-layer networks to deeper architectures. It explains how hidden-layer units form piecewise linear manifolds to divide input space and how paths and their relationships are central to understanding the back-propagation training solution.

Changcun HuangJul 9, 2026