PyQTax
收藏arXiv2024-06-16 更新2024-08-06 收录
下载链接：
http://arxiv.org/abs/2406.10803v1
下载链接
链接失效反馈官方服务：
资源简介：
PyQTax是由摩根大通人工智能研究团队创建的数据集，包含116,671个问题-表格-答案-Python四元组，覆盖多种分类和表格格式。该数据集通过合作游戏'HiddenTables'生成，旨在通过代码生成能力提高语言模型在表格问题回答任务中的表现，同时确保数据安全和降低生成成本。PyQTax不仅支持学术研究，还为语言模型在实际应用中的部署提供了更安全、高效的选择。
PyQTax is a dataset created by the AI Research Team at JPMorgan Chase. It comprises 116,671 question-table-answer-Python quadruples, covering diverse classification and tabular formats. Generated via the collaborative game "HiddenTables", this dataset aims to enhance the performance of language models in tabular question answering tasks through code generation capabilities, while ensuring data security and reducing generation costs. PyQTax not only supports academic research but also provides a safer and more efficient option for the deployment of language models in real-world applications.
提供机构：
摩根大通人工智能研究
创建时间：
2024-06-16
