# Med-MoE: Mixture of Domain-Specific Experts for Lightweight Medical Vision-Language Models

arXiv: https://arxiv.org/abs/2404.10237 · HTML: https://mlsanity.com/paper/2404.10237/

## Key facts

- Cited by 46 papers in the map (data through 14 Oct 2026).
- Placed in Medical vision-language models, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 31 of its 63 references that are in the map.
- First posted on arXiv 16 Apr 2024.

## Placed in

- Multimodal LLMs / vision-language models / [Medical vision-language models](https://mlsanity.com/mllm/medical-vision-language-models/)
- Mixture-of-experts / [Medical and biomedical MoE](https://mlsanity.com/moe/medical-and-biomedical-moe/)

## Builds on (most cited first, 30 of 31)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [DeepSeekMoE: Towards Ultimate Expert Specialization in Mixture-of-Experts Language Models](https://mlsanity.com/paper/2401.06066/) (2024)
- [MedMNIST v2 -- A large-scale lightweight benchmark for 2D and 3D biomedical image classification](https://mlsanity.com/paper/2110.14795/) (2021)
- [PathVQA: 30000+ Questions for Medical Visual Question Answering](https://mlsanity.com/paper/2003.10286/) (2020)
- [GaLore: Memory-Efficient LLM Training by Gradient Low-Rank Projection](https://mlsanity.com/paper/2403.03507/) (2024)
- [SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering](https://mlsanity.com/paper/2102.09542/) (2021)
- [Med-Flamingo: a Multimodal Medical Few-shot Learner](https://mlsanity.com/paper/2307.15189/) (2023)
- [Learning Factored Representations in a Deep Mixture of Experts](https://mlsanity.com/paper/1312.4314/) (2013)
- [MedMNIST Classification Decathlon: A Lightweight AutoML Benchmark for Medical Image Analysis](https://mlsanity.com/paper/2010.14925/) (2020)
- [The Computational Limits of Deep Learning](https://mlsanity.com/paper/2007.05558/) (2020)
- [Auto-Keras: An Efficient Neural Architecture Search System](https://mlsanity.com/paper/1806.10282/) (2018)
- [SPHINX-X: Scaling Data and Parameters for a Family of Multi-modal Large Language Models](https://mlsanity.com/paper/2402.05935/) (2024)
- [BLIVA: A Simple Multimodal LLM for Better Handling of Text-Rich Visual Questions](https://mlsanity.com/paper/2308.09936/) (2023)
- [FastMoE: A Fast Mixture-of-Expert Training System](https://mlsanity.com/paper/2103.13262/) (2021)
- [Mixture of Cluster-conditional LoRA Experts for Vision-language Instruction Tuning](https://mlsanity.com/paper/2312.12379/) (2023)
- [To Repeat or Not To Repeat: Insights from Scaling LLM under Token-Crisis](https://mlsanity.com/paper/2305.13230/) (2023)
- [LLaVA-MoLE: Sparse Mixture of LoRA Experts for Mitigating Data Conflicts in Instruction Finetuning MLLMs](https://mlsanity.com/paper/2401.16160/) (2024)
- [Transparency of Deep Neural Networks for Medical Image Analysis: A Review of Interpretability Methods](https://mlsanity.com/paper/2111.02398/) (2021)
- [SlimPajama-DC: Understanding Data Combinations for LLM Training](https://mlsanity.com/paper/2309.10818/) (2023)
- [When Do We Not Need Larger Vision Models?](https://mlsanity.com/paper/2403.13043/) (2024)
- [Open-Ended Medical Visual Question Answering Through Prefix Tuning of Language Models](https://mlsanity.com/paper/2303.05977/) (2023)
- [MoAI: Mixture of All Intelligence for Large Language and Vision Models](https://mlsanity.com/paper/2403.07508/) (2024)
- [An Empirical Study of Scaling Instruct-Tuned Large Multimodal Models](https://mlsanity.com/paper/2309.09958/) (2023)
- [Q2ATransformer: Improving Medical VQA via an Answer Querying Decoder](https://mlsanity.com/paper/2304.01611/) (2023)
- [Visual Question Answering: A Survey on Techniques and Common Trends in Recent Literature](https://mlsanity.com/paper/2305.11033/) (2023)
- [Towards an empirical understanding of MoE design choices](https://mlsanity.com/paper/2402.13089/) (2024)

## Built on by (most cited first, 30 of 46)

- [Efficient Multimodal Large Language Models: A Survey](https://mlsanity.com/paper/2405.10739/) (2024)
- [Unlocking Multimodal Mathematical Reasoning via Process Reward Model](https://mlsanity.com/paper/2501.04686/) (2025)
- [JarvisIR: Elevating Autonomous Driving Perception with Intelligent Image Restoration](https://mlsanity.com/paper/2504.04158/) (2025)
- [Vision-Language Models for Edge Networks: A Comprehensive Survey](https://mlsanity.com/paper/2502.07855/) (2025)
- [Modality-Fair Preference Optimization for Trustworthy MLLM Alignment](https://mlsanity.com/paper/2410.15334/) (2024)
- [3D-RAD: A Comprehensive 3D Radiology Med-VQA Dataset with Multi-Temporal Analysis and Diverse Diagnostic Tasks](https://mlsanity.com/paper/2506.11147/) (2025)
- [CodeV: Code with Images for Faithful Visual Reasoning via Tool-Aware Policy Optimization](https://mlsanity.com/paper/2511.19661/) (2025)
- [A Survey on Efficient Vision-Language Models](https://mlsanity.com/paper/2504.09724/) (2025)
- [Multimodal Large Language Models for Medicine: A Comprehensive Survey](https://mlsanity.com/paper/2504.21051/) (2025)
- [MedThink: Explaining Medical Visual Question Answering via Multimodal Decision-Making Rationale](https://mlsanity.com/paper/2404.12372/) (2024)
- [DeRS: Towards Extremely Efficient Upcycled Mixture-of-Experts Models](https://mlsanity.com/paper/2503.01359/) (2025)
- [OmniCT: Towards a Unified Slice-Volume LVLM for Comprehensive CT Analysis](https://mlsanity.com/paper/2602.16110/) (2026)
- [PRS-Med: Position Reasoning Segmentation in Medical Imaging](https://mlsanity.com/paper/2505.11872/) (2025)
- [Med-U1: Incentivizing Unified Medical Reasoning in LLMs via Large-scale Reinforcement Learning](https://mlsanity.com/paper/2506.12307/) (2025)
- [FreqMoE: Dynamic Frequency Enhancement for Neural PDE Solvers](https://mlsanity.com/paper/2505.06858/) (2025)
- [Med-GLIP: Advancing Medical Language-Image Pre-training with Large-scale Grounded Dataset](https://mlsanity.com/paper/2508.10528/) (2025)
- [From large language models to multimodal AI: A scoping review on the potential of generative AI in medicine](https://mlsanity.com/paper/2502.09242/) (2025)
- [NeuroMoE: A Transformer-Based Mixture-of-Experts Framework for Multi-Modal Neurological Disorder Classification](https://mlsanity.com/paper/2506.14970/) (2025)
- [Uncertainty-Driven Expert Control: Enhancing the Reliability of Medical Vision-Language Models](https://mlsanity.com/paper/2507.09209/) (2025)
- [TranSQL+: Serving Large Language Models with SQL on Low-Resource Hardware](https://mlsanity.com/paper/2502.02818/) (2025)
- [HeartcareGPT: A Unified Multimodal ECG Suite for Dual Signal-Image Modeling and Understanding](https://mlsanity.com/paper/2506.05831/) (2025)
- [MoSE: Skill-by-Skill Mixture-of-Experts Learning for Embodied Autonomous Machines](https://mlsanity.com/paper/2507.07818/) (2025)
- [CodePercept: Code-Grounded Visual STEM Perception for MLLMs](https://mlsanity.com/paper/2603.10757/) (2026)
- [Beyond Medical Diagnostics: How Medical Multimodal Large Language Models Think in Space](https://mlsanity.com/paper/2603.13800/) (2026)
- [THOR-MoE: Hierarchical Task-Guided and Context-Responsive Routing for Neural Machine Translation](https://mlsanity.com/paper/2505.14173/) (2025)
- [Knowing or Guessing? Robust Medical Visual Question Answering via Joint Consistency and Contrastive Learning](https://mlsanity.com/paper/2508.18687/) (2025)
- [The Rise of Sparse Mixture-of-Experts: A Survey from Algorithmic Foundations to Decentralized Architectures and Vertical Domain Applications](https://mlsanity.com/paper/2602.08019/) (2026)
- [SegMoTE: Token-Level Mixture of Experts for Medical Image Segmentation](https://mlsanity.com/paper/2602.19213/) (2026)
- [Democratizing MLLMs in Healthcare: TinyLLaVA-Med for Efficient Healthcare Diagnostics in Resource-Constrained Settings](https://mlsanity.com/paper/2409.12184/) (2024)
- [MicarVLMoE: A Modern Gated Cross-Aligned Vision-Language Mixture of Experts Model for Medical Image Captioning and Report Generation](https://mlsanity.com/paper/2504.20343/) (2025)

## Abstract (arXiv)

Recent advancements in general-purpose or domain-specific multimodal large language models (LLMs) have witnessed remarkable progress for medical decision-making. However, they are designated for specific classification or generative tasks, and require model training or finetuning on large-scale datasets with sizeable parameters and tremendous computing, hindering their clinical utility across diverse resource-constrained scenarios in practice. In this paper, we propose a novel and lightweight framework Med-MoE (Mixture-of-Experts) that tackles both discriminative and generative multimodal medical tasks. The learning of Med-MoE consists of three steps: multimodal medical alignment, instruction tuning and routing, and domain-specific MoE tuning. After aligning multimodal medical images with LLM tokens, we then enable the model for different multimodal medical tasks with instruction tuning, together with a trainable router tailored for expert selection across input modalities. Finally, the model is tuned by integrating the router with multiple domain-specific experts, which are selectively activated and further empowered by meta expert. Comprehensive experiments on both open- and close-end medical question answering (Med-VQA) and image classification tasks across datasets such as VQA-RAD, SLAKE and Path-VQA demonstrate that our model can achieve performance superior to or on par with state-of-the-art baselines, while only requiring approximately 30%-50% of activated model parameters. Extensive analysis and ablations corroborate the effectiveness and practical utility of our method.