MedAD-38K

<div align="center"> # MedAD-38K ### Dataset for *MedAD-R1: Consistency-Reinforced Policy Optimization for Interpretable Medical Anomaly Detection* [](https://arxiv.org/abs/2602.01081) [](https://github.com/zhtstar/MedAD-R1) [](https://huggingface.co/datasets/zhtstar/MedAD-38K) [](https://modelscope.cn/datasets/zhtstar/MedAD-38K) [](https://huggingface.co/zhtstar/MedAD-R1) [](https://modelscope.cn/models/zhtstar/MedAD-R1) </div> > 数据集目前采用申请访问方式。论文接收后，我们计划在源数据集许可和再分发条款允许的范围内公开项目资源。 ## Dataset overview **MedAD-38K** is a medical-image visual question answering benchmark with quality-controlled diagnostic reasoning annotations. The released dataset contains **38,422 medical images** and **130,146 VQA instances**. <div align="center"> <img src="https://raw.githubusercontent.com/zhtstar/MedAD-R1/main/dataset-tasks.png" alt="MedAD-38K examples" width="780" /> </div> 数据构造、标注流程、数据来源、任务定义和评测协议请参阅论文及其补充材料。 ## Project resources | Resource | Links | | :-- | :-- | | Paper | [MedAD-R1: Consistency-Reinforced Policy Optimization for Interpretable Medical Anomaly Detection](https://arxiv.org/abs/2602.01081) | | Code | [GitHub](https://github.com/zhtstar/MedAD-R1) | | Dataset: MedAD-38K | [Hugging Face](https://huggingface.co/datasets/zhtstar/MedAD-38K) · [ModelScope](https://modelscope.cn/datasets/zhtstar/MedAD-38K) | | Model: MedAD-R1 | [Hugging Face](https://huggingface.co/zhtstar/MedAD-R1) · [ModelScope](https://modelscope.cn/models/zhtstar/MedAD-R1) | ## Citation ```bibtex @article{zhang2026medadr1, title = {MedAD-R1: Consistency-Reinforced Policy Optimization for Interpretable Medical Anomaly Detection}, author = {Zhang, Haitao and Wang, Yingying and Wang, Jiaxiang and Xu, Haote and Zhang, Hongyang and Chen, Yirong and Huang, Yue and Ding, Xinghao}, journal = {arXiv preprint arXiv:2602.01081}, year = {2026} } ```