MedM-VL: What Makes a Good Medical LVLM?
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916946885214208 |
|---|---|
| author | Shi, Yiming Yang, Shaoshuai Zhu, Xun Wang, Haoyu Fu, Xiangling Li, Miao Wu, Ji |
| author_facet | Shi, Yiming Yang, Shaoshuai Zhu, Xun Wang, Haoyu Fu, Xiangling Li, Miao Wu, Ji |
| contents | Medical image analysis is essential in modern healthcare. Deep learning has redirected research focus toward complex medical multimodal tasks, including report generation and visual question answering. Traditional task-specific models often fall short in handling these challenges. Large vision-language models (LVLMs) offer new solutions for solving such tasks. In this study, we build on the popular LLaVA framework to systematically explore model architectures and training strategies for both 2D and 3D medical LVLMs. We present extensive empirical findings and practical guidance. To support reproducibility and future research, we release a modular codebase, MedM-VL, and two pre-trained models: MedM-VL-2D for 2D medical image analysis and MedM-VL-CT-Chest for 3D CT-based applications. The code is available at: https://github.com/MSIIP/MedM-VL |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_04323 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MedM-VL: What Makes a Good Medical LVLM? Shi, Yiming Yang, Shaoshuai Zhu, Xun Wang, Haoyu Fu, Xiangling Li, Miao Wu, Ji Computer Vision and Pattern Recognition Medical image analysis is essential in modern healthcare. Deep learning has redirected research focus toward complex medical multimodal tasks, including report generation and visual question answering. Traditional task-specific models often fall short in handling these challenges. Large vision-language models (LVLMs) offer new solutions for solving such tasks. In this study, we build on the popular LLaVA framework to systematically explore model architectures and training strategies for both 2D and 3D medical LVLMs. We present extensive empirical findings and practical guidance. To support reproducibility and future research, we release a modular codebase, MedM-VL, and two pre-trained models: MedM-VL-2D for 2D medical image analysis and MedM-VL-CT-Chest for 3D CT-based applications. The code is available at: https://github.com/MSIIP/MedM-VL |
| title | MedM-VL: What Makes a Good Medical LVLM? |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2504.04323 |