TechQA-RAG-Eval

## Dataset Description:
TechQA-RAG-Eval is a reduced version of the original TechQA ([IBM’s GitHub Page](https://github.com/ibm/techqa), [HuggingFace](https://huggingface.co/datasets/PrimeQA/TechQA/tree/main)) dataset specifically for evaluating Retrieval-Augmented Generation (RAG) systems. The dataset consists of technical support questions and their answers, sourced from real IBM developer forums where acceptable answers included links to reference technical documentation.

This dataset is ready for commercial/non-commercial use.

## Dataset Owner(s):
NVIDIA Corporation

## Dataset Creation Date:
05/05/2025

## License/Terms of Use:
Apache-2.0

## Intended Usage:
TechQA-RAG-Eval is particularly well-suited for:
Benchmarking RAG system performance on technical domain queries
Evaluating information retrieval systems in technical support contexts
Testing natural language understanding and generation for technical support applications

## Dataset Characterization
| Aspect | Details |
|-----------------------|----------------|
| Data Collection Method| Automated |
| Labeling Method | Not Applicable |

## Dataset Format
The dataset is composed of .txt and .json files.

## Dataset Quantification
| Metric | Value |
|--------------------|--------------|
| Record Count | 908 question/answer pairs |
| Feature Count | 5 |
| Features | ['id', 'question', 'answer', 'is_impossible', 'contexts'] |
| Data Storage Size | 46 MB (.zip)|

## Reference(s):
TechQA: https://github.com/ibm/techqa

## Ethical Considerations:
NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. When downloaded or used in accordance with our terms of service, developers should work with their internal model team to ensure this model meets requirements for the relevant industry and use case and addresses unforeseen product misuse.

Please report security vulnerabilities or NVIDIA AI Concerns [here](https://www.nvidia.com/en-us/support/submit-security-vulnerability/).

## 数据集描述
TechQA-RAG-Eval 是原版TechQA（[IBM官方GitHub页面](https://github.com/ibm/techqa)、[HuggingFace数据集页面](https://huggingface.co/datasets/PrimeQA/TechQA/tree/main)）的精简版本，专为检索增强生成（Retrieval-Augmented Generation，RAG）系统评估打造。该数据集包含源自IBM官方开发者论坛的技术支持问答对，其中可接受的标准答案均附带了技术文档参考链接。

本数据集可用于商业与非商业用途。

## 数据集所有者：英伟达公司（NVIDIA Corporation）

## 数据集创建日期：2025年5月5日（05/05/2025）

## 使用许可条款：Apache-2.0

## 预期用途
TechQA-RAG-Eval 特别适用于以下场景：
1. 针对技术领域查询的检索增强生成系统性能基准测试
2. 技术支持场景下的信息检索系统评估
3. 面向技术支持应用的自然语言理解与生成能力测试

## 数据集特征
| 评估维度 | 详情 |
|-----------------------|----------------|
| 数据采集方式 | 自动化采集 |
| 标注方式 | 不适用 |

## 数据集格式
本数据集由 .txt 与 .json 文件组成。

## 数据集量化统计
| 指标 | 数值 |
|--------------------|--------------|
| 样本数量 | 908条问答对 |
| 特征数量 | 5个 |
| 特征项 | ['id', 'question', 'answer', 'is_impossible', 'contexts'] |
| 数据存储大小 | 46 MB（压缩包格式） |

## 参考文献
TechQA：https://github.com/ibm/techqa

## 伦理考量
英伟达认为，可信人工智能是一项共同责任，我们已建立相关政策与实践规范，以支撑各类人工智能应用的开发。开发者在遵循本数据集服务条款的前提下下载或使用本数据集时，应与内部模型团队协作，确保该模型符合相关行业与应用场景的要求，并防范可能出现的产品误用情况。

请通过[此链接](https://www.nvidia.com/en-us/support/submit-security-vulnerability/)提交安全漏洞或英伟达人工智能相关问题。