JiayiHe/DrVD-Bench_fairness

---
license: mit
task_categories:
- question-answering
- image-to-text
language:
- en
- zh
size_categories:
- 100K<n<1M
---

# DRVD-Bench Fairness

This dataset contains 675,328 examples for medical fairness evaluation.

## Dataset Description

This dataset includes medical questions with fairness attributes injected for bias evaluation in medical AI systems.

## Dataset Structure

### Data Fields

The dataset contains the following fields:

- `source_file`: Dataset field
- `original_index`: Original question before fairness injection
- `unique_id`: Dataset field
- `image_paths`: Dataset field
- `question`: Rewritten question with fairness attribute
- `options`: Dataset field
- `answer`: Dataset field
- `modality`: Dataset field
- `task`: Dataset field
- `options_num`: Dataset field
- `source`: Dataset field
- `original_question`: Original question before fairness injection
- `fairness_attribute_category`: Fairness attribute information
- `fairness_attribute_name`: Fairness attribute information
- `fairness_attribute_value`: Fairness attribute information

## Dataset Statistics

- Total examples: 675,328
- Features: 15

## Usage

```python
from datasets import load_dataset

# Load the dataset
dataset = load_dataset("JiayiHe/DRVD-Bench_fairness")

# Access examples
for example in dataset['train']:
print(example['question'])
print(example['fairness_attribute_name'])
print(example['fairness_attribute_value'])
```

## Citation

If you use this dataset, please cite the original source and mention the fairness evaluation framework.

## License

MIT License

许可证：MIT协议
任务类别：问答、图像转文本
语言：英语、汉语
规模类别：10万 < 样本量 < 100万

# DRVD-Bench 公平性数据集

本数据集共包含675,328条样本，用于医疗AI系统的公平性评估。

## 数据集描述
本数据集收录了注入公平性属性的医疗问题，用于医疗AI系统的偏倚评估。

## 数据集结构
### 数据字段
本数据集包含以下字段：
- `"source_file"`：源文件
- `"original_index"`：公平性注入前的原始问题
- `"unique_id"`：唯一标识符
- `"image_paths"`：图像路径
- `"question"`：注入公平性属性后的重写问题
- `"options"`：选项
- `"answer"`：答案
- `"modality"`：模态
- `"task"`：任务类型
- `"options_num"`：选项数量
- `"source"`：来源
- `"original_question"`：公平性注入前的原始问题
- `"fairness_attribute_category"`：公平性属性类别
- `"fairness_attribute_name"`：公平性属性名称
- `"fairness_attribute_value"`：公平性属性取值

## 数据集统计
- 总样本数：675,328
- 特征维度数：15

## 使用方法
python
from datasets import load_dataset

# 加载数据集
dataset = load_dataset("JiayiHe/DRVD-Bench_fairness")

# 访问样本
for example in dataset['train']:
print(example['question'])
print(example['fairness_attribute_name'])
print(example['fairness_attribute_value'])

## 引用说明
若使用本数据集，请引用其原始来源，并提及本公平性评估框架。

## 许可证
MIT 许可证