NuminaMath-TIR

# Dataset Card for NuminaMath CoT

## Dataset Description

- **Homepage:** https://projectnumina.ai
- **Repository:** https://github.com/project-numina/aimo-progress-prize
- **Paper:** https://github.com/project-numina/aimo-progress-prize/blob/main/report/numina_dataset.pdf
- **Leaderboard:**
- **Point of Contact:** [Jia Li]([e-mail])

### Dataset Summary

Tool-integrated reasoning (TIR) plays a crucial role in this competition. However, collecting and annotating such data is both costly and time-consuming. To address this, we selected approximately 70k problems from the NuminaMath-CoT dataset, focusing on those with numerical outputs, most of which are integers. We then utilized a pipeline leveraging GPT-4 to generate TORA-like reasoning paths, executing the code and producing results until the solution was complete. We filtered out solutions where the final answer did not match the reference and repeated this process three times to ensure accuracy and consistency. This iterative approach allowed us to generate high-quality TORA data efficiently.

### Licensing Information

The dataset is available under the [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0).

### Citation Information

```
@misc{numina_math_datasets,
author = {Jia LI, Edward Beeching, Lewis Tunstall, Ben Lipkin, Roman Soletskyi, Shengyi Costa Huang, Kashif Rasul, Longhui Yu, Albert Jiang, Ziju Shen, Zihan Qin, Bin Dong, Li Zhou, Yann Fleureau, Guillaume Lample, and Stanislas Polu},
title = {NuminaMath TIR},
year = {2024},
publisher = {Numina},
journal = {Hugging Face repository},
howpublished = {\url{[https://huggingface.co/AI-MO/NuminaMath-TIR](https://github.com/project-numina/aimo-progress-prize/blob/main/report/numina_dataset.pdf)}}
}
```

# NuminaMath CoT 数据集卡片
## 数据集说明
- **项目主页**：https://projectnumina.ai
- **代码仓库**：https://github.com/project-numina/aimo-progress-prize
- **相关论文**：https://github.com/project-numina/aimo-progress-prize/blob/main/report/numina_dataset.pdf
- **排行榜**：
- **联系人**：[Jia Li]([e-mail])

### 数据集概况
工具集成推理（Tool-integrated Reasoning, TIR）在本次赛事中扮演着至关重要的角色。然而此类数据的收集与标注工作往往成本高昂且耗时良久。为解决这一难题，我们从NuminaMath-CoT数据集中筛选出约7万个问题，重点关注输出为数值的题目，其中绝大多数结果为整数。随后我们借助基于GPT-4的流水线生成类TORA的推理路径，执行代码并生成中间结果，直至得到完整的解题方案。我们会过滤掉最终答案与参考结果不符的解题方案，并将该流程重复三次，以确保数据的准确性与一致性。这种迭代式的方法帮助我们高效生成了高质量的类TORA数据。

### 许可证信息
本数据集遵循[Apache许可证2.0版](https://www.apache.org/licenses/LICENSE-2.0)发布。

### 引用信息

@misc{numina_math_datasets,
author = {Jia LI, Edward Beeching, Lewis Tunstall, Ben Lipkin, Roman Soletskyi, Shengyi Costa Huang, Kashif Rasul, Longhui Yu, Albert Jiang, Ziju Shen, Zihan Qin, Bin Dong, Li Zhou, Yann Fleureau, Guillaume Lample, and Stanislas Polu},
title = {NuminaMath TIR},
year = {2024},
publisher = {Numina},
journal = {Hugging Face 仓库},
howpublished = {url{[https://huggingface.co/AI-MO/NuminaMath-TIR](https://github.com/project-numina/aimo-progress-prize/blob/main/report/numina_dataset.pdf)}}
}