NextGenWhu/FinCDM-Fin-KQA

FinCDM-CPA-KQA是一个专门用于金融知识问答的数据集，源自研究论文From Scores to Skills: A Cognitive Diagnosis Framework for Evaluating Financial Large Language Models。该数据集旨在评估大型语言模型在金融知识推理、合规评估和基于知识的问答方面的能力。数据集以JSON格式提供，包含多项选择题，每个条目有id、query、answer、choices、gold和text等字段。使用案例包括评估大型语言模型、训练问答系统以及合规和审计任务。

FinCDM-CPA-KQA is a specialized dataset for financial knowledge-based question answering, derived from the research presented in From Scores to Skills: A Cognitive Diagnosis Framework for Evaluating Financial Large Language Models. This dataset is designed to evaluate large language models (LLMs) on their ability to perform financial knowledge reasoning, compliance assessment, and knowledge-based question answering. It serves as a robust benchmark for assessing how well models understand and apply financial domain knowledge, making it valuable for both research and practical applications in finance. The dataset is provided in JSON format and consists of multiple-choice questions tailored to financial knowledge. Each entry includes fields such as id, query, answer, choices, gold, and text. Use cases include benchmarking LLMs, training QA systems, and compliance and auditing tasks.