SCIR-HI/PseudoMD-1M

---
license: apache-2.0
task_categories:
- translation
- text2text-generation
language:
- en
tags:
- chemistry
- biology
- medical
size_categories:
- 1M<n<10M
---

Pre-training dataset used in paper "[From Artificially Real to Real: Leveraging Pseudo Data from Large Language Models for Low-Resource Molecule Discovery](https://arxiv.org/abs/2309.05203)" (AAAI 2024)

PseudoMD-1M dataset is the first artificially-real dataset for cross-modal molecule discovery, which consists of 1,020,139 pseudo molecule-description pairs. Every molecule is represented using its Canonical SMILES notation, sourced from PubChem via the PUG View API. On average, each description within PseudoMD-1M contains 5.11 sentences, 106.47 words, and 165.07 tokens.

### Citation
If you found the dataset useful, please cite:
```bibtex
@article{chen2023artificially,
title={From Artificially Real to Real: Leveraging Pseudo Data from Large Language Models for Low-Resource Molecule Discovery},
author={Chen, Yuhan and Xi, Nuwa and Du, Yanrui and Wang, Haochun and Jianyu, Chen and Zhao, Sendong and Qin, Bing},
journal={arXiv preprint arXiv:2309.05203},
year={2023}
}
```

原始信息汇总

数据集概述

基本信息

• 许可证：Apache 2.0
• 任务类别：
  • 翻译
  • 文本生成
• 语言：英语
• 标签：
  • 化学
  • 生物学
  • 医学
• 数据集大小：1M<n<10M

详细描述

• 数据集名称：PseudoMD-1M
• 数据集用途：用于跨模态分子发现的伪数据集
• 数据集组成：包含1,020,139个伪分子-描述对
• 分子表示：使用Canonical SMILES表示法，数据来源为PubChem通过PUG View API
• 描述特征：每个描述平均包含5.11个句子，106.47个单词，165.07个词

引用信息

• 论文标题：From Artificially Real to Real: Leveraging Pseudo Data from Large Language Models for Low-Resource Molecule Discovery
• 作者：Chen, Yuhan 等
• 发表期刊：arXiv preprint arXiv:2309.05203
• 发表年份：2023