TCM-Vision-Benchmark

# <span>📚 Introduction</span> This is the text benchmark for [ShizhenGPT](https://github.com/FreedomIntelligence/ShizhenGPT), a multimodal LLM for **Traditional Chinese Medicine (TCM)**. For details, see our [paper](https://arxiv.org/abs/2508.14706) and [GitHub repository](https://github.com/FreedomIntelligence/ShizhenGPT). # <span>📊 Benchmark Overview</span> The benchmark is composed of 7 sections, each compiled from different authoritative TCM illustrated books. | | Samples | | ------------------------------------ | ------------------------------ | | TCM Patent | 1119 | | TCM Material | 1020 | | TCM Herb | 1100 | | Tongue | 768 | | Palm | 640 | | Holism | 1011 | | Tuina | 831 | | Eye | 715 | ## ⚒️ Data Construction ```json { "image": [ "tcm_bench_images/0001.jpg" ], "question": "请根据这张图片，判断它属于下面哪一种？", "options": { "A": "巴戟肉", "B": "盐巴戟天", "C": "制白附子片", "D": "生白附子片" }, "answer": "巴戟肉", "answer_idx": "A", "category": "TCM Patent" } ``` # <span>📖 Citation</span> If you find our data useful, please consider citing our work! ``` @misc{chen2025shizhengptmultimodalllmstraditional, title={ShizhenGPT: Towards Multimodal LLMs for Traditional Chinese Medicine}, author={Junying Chen and Zhenyang Cai and Zhiheng Liu and Yunjin Yang and Rongsheng Wang and Qingying Xiao and Xiangyi Feng and Zhan Su and Jing Guo and Xiang Wan and Guangjun Yu and Haizhou Li and Benyou Wang}, year={2025}, eprint={2508.14706}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2508.14706}, } ```# 📚 引言 本数据集为面向**中医药（Traditional Chinese Medicine, TCM）**的多模态大语言模型（Large Language Model, LLM）时珍GPT（ShizhenGPT，项目链接：https://github.com/FreedomIntelligence/ShizhenGPT）所配套的文本基准测试集。详细信息可参阅我们的[研究论文](https://arxiv.org/abs/2508.14706)与[GitHub开源仓库](https://github.com/FreedomIntelligence/ShizhenGPT)。 # 📊 基准测试集概览 本基准测试集共包含7个模块，所有数据均源自不同权威中医药图解典籍。 | | 样本数量 | | ------------------------------------ | ------------------------------ | | 中成药（TCM Patent） | 1119 | | 中药材（TCM Material） | 1020 | | 草本药材（TCM Herb） | 1100 | | 舌象（Tongue） | 768 | | 掌象（Palm） | 640 | | 整体辨证（Holism） | 1011 | | 推拿（Tuina） | 831 | | 眼象（Eye） | 715 | ## ⚒️ 数据构建示例 json { "图像路径": [ "tcm_bench_images/0001.jpg" ], "问题": "请根据这张图片，判断它属于下面哪一种？", "选项": { "A": "巴戟肉", "B": "盐巴戟天", "C": "制白附子片", "D": "生白附子片" }, "正确答案": "巴戟肉", "答案索引": "A", "分类": "中成药（TCM Patent）" } # 📖 引用说明 若您认为本数据集对您的研究有所帮助，请考虑引用我们的工作： bibtex @misc{chen2025shizhengptmultimodalllmstraditional, title={ShizhenGPT: Towards Multimodal LLMs for Traditional Chinese Medicine}, author={Junying Chen and Zhenyang Cai and Zhiheng Liu and Yunjin Yang and Rongsheng Wang and Qingying Xiao and Xiangyi Feng and Zhan Su and Jing Guo and Xiang Wan and Guangjun Yu and Haizhou Li and Benyou Wang}, year={2025}, eprint={2508.14706}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2508.14706}, }

搜集汇总

背景与挑战

背景概述

TCM-Vision-Benchmark是一个专为传统中医多模态大语言模型ShizhenGPT设计的基准测试数据集，包含TCM专利、药材等7个类别，数据源自权威中医图解书籍，以图像和问答对形式构建。