📜 BRIDGE-Open Dataset

The BRIDGE-Open dataset is an open-access subset of the datasets used by the BRIDGE benchmark. This dataset includes the 55 tasks of the open-access datasets that can be released to the public. Each dataset should be used in accordance with the license of its original release (the source of each dataset is listed at our BRIDGE paper, Supplementary Section 5.)

Due to privacy and security considerations of clinical data, regulated-access datasets can not be directly published. However, all detailed task descriptions and their corresponding data sources are available in our BRIDGE paper, Supplementary Section 5. Importantly, all 87 datasets have been verified to be either fully open-access or publicly accessible via reasonable request in our systematic review.

📂 Dataset Structure

The BRIDGE-Open dataset contains the following two folders:

• Dataset: This folder contains all cases used to evaluate the performance of the LLMs included by our BRIGDE benchmark.
• Example: This folder contains the 5 additionally and randomly selected cases that are used for the 5-shot inference of our BRIDGE benchmark.

Each file under the two above folders are JSON files, with the files under the Dataset folder named as dataset_task_name.json, and the files under the Example folder named as dataset_task_name.example.json. Each JSON file has the following fields:

• task: The name of the task.
• language: The language of the dataset.
• task type: The type of the task (e.g., Text Classification, Event Extraction, etc.).
• id: The unique identifier of the samples.
• split: The split of the dataset. All samples under the Dataset folder will have test in this field. All samples under the Example folder will have train in this field.
• instruction: The system prompt for the task. This is the instruction that will be given to the LLMs.
• input: The input text. This is the text that will be given to the LLMs.
• output (only for the example JSON files): The expected output. This is the true label of the sample.
• pred (only for the dataset JSON files): The actual output of the LLM. This field will be left blank for all datasets, and will be populated by the LLM after the model inference.

📖 Dataset Descriptions

The detailed descriptions of the datasets can be found in our BRIDGE paper, Supplementary Section 5.

🏆 BRIDGE Leaderboard

Our BRIDGE leaderboard is publically available at here, which ranks the performance of different LLMs on all 87 tasks described in the paper. The leaderboard is updated regularly when new models and datatsets are added to the benchmark.

🤝 Contributing

We welcome and greatly value contributions and collaborations from the community! If you have clinical text datasets that you would like to add to the BRIDEG benchmark, please fill in the Google Form and let us know!

We are committed to expanding BRIDGE while strictly adhering to appropriate data use agreements and ethical guidelines. Let's work together to advance the responsible application of LLMs in medicine!

💻 BRIDGE Code Repository

The code repository for the BRIDGE benchmark is publically available at GitHub. For relevant information about the repository, please refer to the BRIDGE leaderboard.

📚 Citation

If you use the BRIDGE benchmark or any of the datasets in your research, please cite the following papers:

@article{BRIDGE-benchmark,
    title={BRIDGE: Benchmarking Large Language Models for Understanding Real-world Clinical Practice Text},
    author={Wu, Jiageng and Gu, Bowen and Zhou, Ren and Xie, Kevin and Snyder, Doug and Jiang, Yixing and Carducci, Valentina and Wyss, Richard and Desai, Rishi J and Alsentzer, Emily and Celi, Leo Anthony and Rodman, Adam and Schneeweiss, Sebastian and Chen, Jonathan H. and Romero-Brufau, Santiago and Lin, Kueiyu Joshua and Yang, Jie},
    year={2025},
    journal={arXiv preprint arXiv: 2504.19467},
    archivePrefix={arXiv},
    primaryClass={cs.CL},
    url={https://arxiv.org/abs/2504.19467},
}
@article{clinical-text-review,
    title={Clinical text datasets for medical artificial intelligence and large language models—a systematic review},
    author={Wu, Jiageng and Liu, Xiaocong and Li, Minghui and Li, Wanxin and Su, Zichang and Lin, Shixu and Garay, Lucas and Zhang, Zhiyun and Zhang, Yujie and Zeng, Qingcheng and Shen, Jie and Yuan, Changzheng and Yang, Jie},
    journal={NEJM AI},
    volume={1},
    number={6},
    pages={AIra2400012},
    year={2024},
    publisher={Massachusetts Medical Society}
}