Saved in:
Bibliographic Details
Main Authors: He, Yuting, You, Chenyu, Li, Shuo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.21861
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914585354698752
author He, Yuting
You, Chenyu
Li, Shuo
author_facet He, Yuting
You, Chenyu
Li, Shuo
contents Multi-modality medical vision (MV) foundation models (FM) are fundamentally challenged by pronounced Non-IID feature statistics across heterogeneous imaging modalities. Monolithic self-supervised optimization on such data induces conflicting gradients, driving representations to collapse toward modality-dominant shortcuts. This work reframes this failure as an imbalance between specialization and coordination in emergent modularity, and proposes Director-Experts (DEX), a modular network that explicitly regulates these dynamics in stacked modules. Each DEX module comprises a pool of experts, dynamically adapted by our image-wise activation strategy, autonomously specializing in modality-dominant statistics, together with a director, updated via our group exponential moving average, which distills multi-expert knowledge into a shared space for semantic integration across modalities, thus driving the emergence of modular representations. We curate a new benchmark, Medical Vision Universe, over 4 million images across 10 modalities, which provides a FM-level pre-training with the broadest coverage of distinct imaging modalities to our DEX. Extensive evaluations on 26 downstream tasks demonstrate improved optimization behavior and transferability, indicating DEX as a principled step toward general-purpose multi-modality medical AI. Our code and dataset will be opened at https://github.com/YutingHe-list/DEX.
format Preprint
id arxiv_https___arxiv_org_abs_2605_21861
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Learning Emergent Modular Representations in Multi-modality Medical Vision Foundation Models
He, Yuting
You, Chenyu
Li, Shuo
Computer Vision and Pattern Recognition
Artificial Intelligence
Multi-modality medical vision (MV) foundation models (FM) are fundamentally challenged by pronounced Non-IID feature statistics across heterogeneous imaging modalities. Monolithic self-supervised optimization on such data induces conflicting gradients, driving representations to collapse toward modality-dominant shortcuts. This work reframes this failure as an imbalance between specialization and coordination in emergent modularity, and proposes Director-Experts (DEX), a modular network that explicitly regulates these dynamics in stacked modules. Each DEX module comprises a pool of experts, dynamically adapted by our image-wise activation strategy, autonomously specializing in modality-dominant statistics, together with a director, updated via our group exponential moving average, which distills multi-expert knowledge into a shared space for semantic integration across modalities, thus driving the emergence of modular representations. We curate a new benchmark, Medical Vision Universe, over 4 million images across 10 modalities, which provides a FM-level pre-training with the broadest coverage of distinct imaging modalities to our DEX. Extensive evaluations on 26 downstream tasks demonstrate improved optimization behavior and transferability, indicating DEX as a principled step toward general-purpose multi-modality medical AI. Our code and dataset will be opened at https://github.com/YutingHe-list/DEX.
title Learning Emergent Modular Representations in Multi-modality Medical Vision Foundation Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2605.21861