vnu-llm2023-ftdata/1k_test_finetune_crawl-quyche

这是一个由问答题对组成的越南语数据集，内容基于河内国立大学的招生计划、培训规定等官方文件和实际情况。数据集采用JSON格式，包含问题的唯一标识符、问题内容、答案、法律依据引用、是否含多意图、是否缺乏上下文、推理级别、主题和问题类型等信息。该数据集旨在评估语言模型在教育培训领域的表现，以及模型对法规的依从性和多步推理能力。

This is a Vietnamese dataset composed of question-answer pairs based on the enrollment plans, training regulations of Hanoi National University, and real-life situations. The dataset is in JSON format, including fields such as unique question identifier, question content, answer, legal reference, multi-intent flag, insufficient context flag, reasoning level, topic, and question type. The dataset aims to evaluate the performance of language models in the educational training field, as well as their compliance with regulations and multi-step reasoning ability.