jminc/resume-matching-dataset-v1

该数据集是一组简历和职位描述的对，旨在用于训练和评估大型语言模型在简历与职位匹配任务上的表现。简历是通过GPT-4o模型合成的，而职位描述则包括真实世界的数据和合成数据。每个样本都包含一个标准化的评估提示、一份合成简历、一份真实或合成的职位描述以及由GPT生成的评估输出。

This dataset consists of pairs of resumes and job postings for training and evaluating large language models on resume-job matching tasks. The resumes are synthetically generated using GPT-4o, and the job postings include both real-world data and synthetic examples. Each sample comprises a standardized evaluation prompt, a synthetic resume, a real or synthetic job description, and a GPT-generated evaluation output.