metadata
library_name: transformers
tags:
- medical image
- VLM
- MedVision
base_model: Qwen/Qwen2.5-VL-32B-Instruct
base_model_relation: merge
datasets: YongchengYAO/MedVision
license: cc-by-nc-4.0
arxiv: 2511.18676
Model Card for Model ID
This is a supervised finetuned version of Qwen/Qwen2.5-VL-32B-Instruct.
Model Details
Model Description
- License: CC-BY-NC-4.0
- Finetuned from model:
Qwen/Qwen2.5-VL-32B-Instruct
Uses
- Direct use: check
Qwen/Qwen2.5-VL-32B-Instruct - The model is evaluated in the MedVision benchmark
Training Details
Training Data
The model is trained on the MedVision dataset. Specifically, this list defines the training data.
Training Procedure
- 🌟 Tutorial: Supervised Fine-Tuning (SFT) for VLMs on Medical Image Data
- 🧑🏻💻 Script: the training script handles dependencies installation, data processing, and SFT training.
Evaluation
See the angle/distance (A/D) measurement results for Qwen2.5VL (32B, SFT_{5K}) on the MedVision project page.
Citation
@misc{yao2025medvisiondatasetbenchmarkquantitative,
title={MedVision: Dataset and Benchmark for Quantitative Medical Image Analysis},
author={Yongcheng Yao and Yongshuo Zong and Raman Dutt and Yongxin Yang and Sotirios A Tsaftaris and Timothy Hospedales},
year={2025},
eprint={2511.18676},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2511.18676},
}