MGSM_en

評価スコアの再現性確保と SB Intuitions 修正版の公開用クローン

- ソース: [juletxara/mgsm on Hugging Face](https://huggingface.co/datasets/juletxara/mgsm)
- Subset: en のみを抽出

# MGSM

> Multilingual Grade School Math Benchmark (MGSM) is a benchmark of grade-school math problems,
> proposed in the paper Language models are multilingual chain-of-thought reasoners.

## Licensing Information

[Creative Commons Attribution Share Alike 4.0 International](https://choosealicense.com/licenses/cc-by-sa-4.0/)

## Citation Information

```
@article{cobbe2021gsm8k,
title={Training Verifiers to Solve Math Word Problems},
author={Cobbe, Karl and Kosaraju, Vineet and Bavarian, Mohammad and Chen, Mark and Jun, Heewoo and Kaiser, Lukasz and Plappert, Matthias and Tworek, Jerry and Hilton, Jacob and Nakano, Reiichiro and Hesse, Christopher and Schulman, John},
journal={arXiv preprint arXiv:2110.14168},
year={2021}
}
@misc{shi2022language,
title={Language Models are Multilingual Chain-of-Thought Reasoners},
author={Freda Shi and Mirac Suzgun and Markus Freitag and Xuezhi Wang and Suraj Srivats and Soroush Vosoughi and Hyung Won Chung and Yi Tay and Sebastian Ruder and Denny Zhou and Dipanjan Das and Jason Wei},
year={2022},
eprint={2210.03057},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
```

# Subsets

## default

- `question` (`str`): a string for the grade-school level math question
- `answer` (`str`): a string for the corresponding answer with chain-of-thought steps (train only)
- `answer_number` (`int`): the numeric solution to the question（答えは正の整数のみ）
- `equation_solution` (`str`): the equation solution to the question (train only)

本数据集旨在保障评估评分的可复现性，为SB Intuitions修正版的公开克隆数据集

- 数据集来源：Hugging Face上的juletxara/mgsm数据集（https://huggingface.co/datasets/juletxara/mgsm）
- 子集设置：仅提取英文数据

# MGSM
> 多语言小学数学基准测试（Multilingual Grade School Math Benchmark, MGSM）是一个包含小学数学题目的基准数据集，由论文《Language models are multilingual chain-of-thought reasoners》提出。

## 许可证信息
采用知识共享署名-相同方式共享4.0国际许可协议（Creative Commons Attribution Share Alike 4.0 International，https://choosealicense.com/licenses/cc-by-sa-4.0/）

## 引用信息

@article{cobbe2021gsm8k,
title={Training Verifiers to Solve Math Word Problems},
author={Cobbe, Karl and Kosaraju, Vineet and Bavarian, Mohammad and Chen, Mark and Jun, Heewoo and Kaiser, Lukasz and Plappert, Matthias and Tworek, Jerry and Hilton, Jacob and Nakano, Reiichiro and Hesse, Christopher and Schulman, John},
journal={arXiv preprint arXiv:2110.14168},
year={2021}
}
@misc{shi2022language,
title={Language Models are Multilingual Chain-of-Thought Reasoners},
author={Freda Shi and Mirac Suzgun and Markus Freitag and Xuezhi Wang and Suraj Srivats and Soroush Vosoughi and Hyung Won Chung and Yi Tay and Sebastian Ruder and Denny Zhou and Dipanjan Das and Jason Wei},
year={2022},
eprint={2210.03057},
archivePrefix={arXiv},
primaryClass={cs.CL}
}

# 子集

## 默认子集
- `question`（字符串类型）：小学阶段数学题目的文本表述
- `answer`（字符串类型）：对应题目的带思维链（Chain-of-Thought）步骤的解答（仅训练集包含该字段）
- `answer_number`（整数类型）：题目的数值解（注：答案仅为正整数）
- `equation_solution`（字符串类型）：题目的方程式解答（仅训练集包含该字段）