five

AfterQuery/FinanceQA

收藏
Hugging Face2025-02-21 更新2025-04-26 收录
下载链接:
https://hf-mirror.com/datasets/AfterQuery/FinanceQA
下载链接
链接失效反馈
官方服务:
资源简介:
FinanceQA是一个旨在评估LLM在模拟现实世界投资工作的复杂金融分析任务上表现的综合测试集。该数据集比现有的金融基准更具挑战性和实用性,专注于需要精确计算和专业知识判断的任务。数据集包含两种主要问题类别:战术性问题(基于财务文件,测试计算准确性、会计准则、假设制作和现实实践)和概念性问题(测试对金融关系、逻辑推导、行业估计和会计原则的理解)。数据集的字段包括:上下文(主要财务文件的相应部分)、问题(具体的金融分析任务或查询)、答案(正确的计算或响应)、思考链(得出正确答案的逻辑推理)、问题类型(基本、假设或概念)、公司(讨论的公司)、文件链接(上下文字段的来源链接)和文件名(来源文件的文件名)。

FinanceQA is a comprehensive testing suite designed to evaluate LLMs performance on complex financial analysis tasks that mirror real-world investment work. The dataset aims to be substantially more challenging and practical than existing financial benchmarks, focusing on tasks that require precise calculations and professional judgment. The dataset contains two main categories of questions: Tactical Questions (based on financial documents, testing calculation accuracy, accounting standards, assumption-making, and real-world practices) and Conceptual Questions (testing understanding of financial relationships, logical derivations, industry estimations, and accounting principles). The dataset includes fields such as context (relevant sections from primary financial documents), question (specific financial analysis task or query), answer (correct calculation or response), chain_of_thought (reasoning logic to arrive at the correct answer), question_type (categorized as basic, assumption, or conceptual), company (company in question), file_link (link to the source of the context field), and file_name (file name of the source of the context field).
提供机构:
AfterQuery
5,000+
优质数据集
54 个
任务类型
进入经典数据集
二维码
社区交流群

面向社区/商业的数据集话题

二维码
科研交流群

面向高校/科研机构的开源数据集话题

数据驱动未来

携手共赢发展

商业合作