Professional Certification Benchmark Exam Dataset
收藏arXiv2023-05-07 更新2024-06-21 收录
下载链接：
https://github.com/reveondivad/certify
下载链接
链接失效反馈官方服务：
资源简介：
专业认证基准测试数据集是由PeopleTec公司创建的，包含1149个专业认证考试，总计6137个多项选择题。该数据集旨在测试大型语言模型如GPT-3和Turbo-GPT3.5在职业技能评估中的表现，特别强调职业准备而非学术成绩。数据集涵盖了从网络安全到项目管理等多个计算机相关领域，以及护理、持牌咨询、药学和教学等其他专业领域。通过这些考试，评估模型在无需微调或考试准备的情况下，能否达到专业雇员或求职者的能力水平。
The Professional Certification Benchmark Dataset was created by PeopleTec, Inc., comprising 1,149 professional certification exams and a total of 6,137 multiple-choice questions. This dataset is designed to test the performance of large language models (LLMs) such as GPT-3 and Turbo-GPT 3.5 in occupational skill assessment, with a particular focus on career readiness rather than academic performance. The dataset covers multiple computer-related fields ranging from cybersecurity to project management, as well as other professional domains including nursing, licensed counseling, pharmacy, and teaching. These exams are used to evaluate whether a model can achieve the competency levels of professional employees or job applicants without fine-tuning or exam preparation.
提供机构：
PeopleTec, Inc., Huntsville, AL, USA
创建时间：
2023-05-07
