MedM-VL-2D-3B-en

Introduction

A 2D medical LVLM trained on 2D medical images and English medical texts, enabling tasks such as report generation, VQA, referring expression comprehension (REC), referring expression generation (REG) and image classification.

	Config
Image encoder	google/siglip-base-patch16-256-multilingual
Connector	MLP (2-layer)
LLM	Qwen/Qwen2.5-3B-Instruct
Image resolution	256*256
Sequence length	2048

Evaluation

Benchmark	Med-Flamingo	LLaVA-Med	RadFM	MedM-VL-2D-3B-en
MedMNISTderma	0.012	0.258	0.051	0.810
MedMNISTorgan	0.089	0.668	0.189	0.791
MedPix	0.081	0.151	-	0.087
MIMIC-CXR	0.233	0.204	0.068	0.222
PathVQA	0.334	0.378	0.248	0.634
SAMedidentify	-	0.458	-	0.637
SAMedrefer	-	0.086	-	0.225
SLAKEidentify	-	0.272	-	0.349
SLAKErefer	-	0.041	-	0.261
SLAKEvqa	0.215	0.337	0.817	0.812

Quickstart

Please refer to MedM-VL.

Citation

@inproceedings{shi2025medm,
  title={Medm-vl: What makes a good medical lvlm?},
  author={Shi, Yiming and Yang, Shaoshuai and Zhu, Xun and Wang, Haoyu and Fu, Xiangling and Li, Miao and Wu, Ji},
  booktitle={International Workshop on Agentic AI for Medicine},
  pages={290--299},
  year={2025},
  organization={Springer}
}