FinLLMs
收藏资源简介:
FinLLMs是由北京理工大学计算机科学与技术学院的研究团队开发的一个金融领域问答数据集,旨在通过大型语言模型生成金融问题解答数据。该数据集包含15,361个样本,涵盖文本和表格信息,用于训练和评估金融领域的数值推理模型。数据集的创建过程涉及收集常用金融公式,构建公式图,并利用GPT-3.5生成问答数据。FinLLMs数据集的应用领域主要是金融分析,旨在自动化复杂数值推理过程,提高决策质量。
FinLLMs is a financial domain question answering dataset developed by the research team from the School of Computer Science and Technology, Beijing Institute of Technology. It is designed to generate financial question-answering data through large language models. This dataset contains 15,361 samples covering both textual and tabular information, and is used for training and evaluating numerical reasoning models in the financial domain. The dataset creation process involves collecting commonly used financial formulas, constructing formula graphs, and generating question-answering data using GPT-3.5. The main application areas of FinLLMs are financial analysis, aiming to automate complex numerical reasoning processes and improve decision-making quality.




