tatdqa_test

## Dataset Description

This is the test set taken from the [TAT-DQA dataset](https://nextplusplus.github.io/TAT-DQA/). TAT-DQA is a large-scale Document VQA dataset that was constructed from publicly available real-world financial reports. It focuses on rich tabular and textual content requiring numerical reasoning. Questions and answers were manually annotated by human experts in finance.

Example of data (see viewer)

### Data Curation
Unlike other 'academic' datasets, we kept the full test set as this dataset closely represents our use case of document retrieval. There are 1,663 image-query pairs.

### Load the dataset

```python
from datasets import load_dataset
ds = load_dataset("vidore/tatdqa_test", split="test")
```

### Dataset Structure

Here is an example of a dataset instance structure:

```json
features:
- name: questionId
dtype: string
- name: query
dtype: string
- name: question_types
dtype: 'null'
- name: image
dtype: image
- name: docId
dtype: int64
- name: image_filename
dtype: string
- name: page
dtype: string
- name: answer
dtype: 'null'
- name: data_split
dtype: string
- name: source
dtype: string
```

## Citation Information

If you use this dataset in your research, please cite the original dataset as follows:

```latex
@inproceedings{zhu-etal-2021-tat,
title = "{TAT}-{QA}: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance",
author = "Zhu, Fengbin and
Lei, Wenqiang and
Huang, Youcheng and
Wang, Chao and
Zhang, Shuo and
Lv, Jiancheng and
Feng, Fuli and
Chua, Tat-Seng",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.254",
doi = "10.18653/v1/2021.acl-long.254",
pages = "3277--3287"
}

@inproceedings{zhu2022towards,
title={Towards complex document understanding by discrete reasoning},
author={Zhu, Fengbin and Lei, Wenqiang and Feng, Fuli and Wang, Chao and Zhang, Haozhou and Chua, Tat-Seng},
booktitle={Proceedings of the 30th ACM International Conference on Multimedia},
pages={4857--4866},
year={2022}
}

```

## 数据集说明
本测试集取自[TAT-DQA数据集(TAT-DQA dataset)](https://nextplusplus.github.io/TAT-DQA/)。TAT-DQA是一个大规模文档视觉问答（Document VQA）数据集，其构建素材均来自公开可得的真实金融报告。该数据集聚焦于包含丰富表格与文本内容、需要数值推理的任务场景，问答对均由金融领域的人类专家手动标注完成。

数据示例（详见查看器）

### 数据集整理
与其他"学术类"数据集不同，我们保留了完整的测试集，因该数据集高度贴合我们的文档检索应用场景。该数据集共包含1663组图像-查询对。

### 加载数据集
python
from datasets import load_dataset
ds = load_dataset("vidore/tatdqa_test", split="test")

### 数据集结构
以下为数据集实例的结构示例：
json
features:
- name: questionId
dtype: string
- name: query
dtype: string
- name: question_types
dtype: 'null'
- name: image
dtype: image
- name: docId
dtype: int64
- name: image_filename
dtype: string
- name: page
dtype: string
- name: answer
dtype: 'null'
- name: data_split
dtype: string
- name: source
dtype: string

## 引用信息
若您在研究中使用该数据集，请按以下方式引用原始数据集：
latex
@inproceedings{zhu-etal-2021-tat,
title = "{TAT}-{QA}: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance",
author = "Zhu, Fengbin and
Lei, Wenqiang and
Huang, Youcheng and
Wang, Chao and
Zhang, Shuo and
Lv, Jiancheng and
Feng, Fuli and
Chua, Tat-Seng",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.254",
doi = "10.18653/v1/2021.acl-long.254",
pages = "3277--3287"
}

@inproceedings{zhu2022towards,
title={Towards complex document understanding by discrete reasoning},
author={Zhu, Fengbin and Lei, Wenqiang and Feng, Fuli and Wang, Chao and Zhang, Haozhou and Chua, Tat-Seng},
booktitle={Proceedings of the 30th ACM International Conference on Multimedia},
pages={4857--4866},
year={2022}
}