遇见数据集

DataRaceBench V1.4.1 and DataRaceBench-ML V0.1

收藏
arXiv2023-08-17 更新2024-06-21 收录
官方服务:

资源简介:

DataRaceBench V1.4.1和DataRaceBench-ML V0.1是由爱荷华州立大学和劳伦斯利弗莫尔国家实验室的研究人员创建的数据集,旨在评估数据竞争检测工具。DataRaceBench包含181个微基准测试,而DataRaceBench-ML V0.1则专注于机器学习方法,特别是大型语言模型(LLMs)的微调。数据集通过详细的标签和代码示例,支持机器学习模型在数据竞争检测中的应用,解决了并发编程中数据竞争检测的挑战。

DataRaceBench V1.4.1 and DataRaceBench-ML V0.1 are datasets developed by researchers from Iowa State University and Lawrence Livermore National Laboratory, designed to evaluate data race detection tools. DataRaceBench includes 181 microbenchmarks, while DataRaceBench-ML V0.1 focuses on machine learning approaches, particularly the fine-tuning of large language models (LLMs). These datasets provide detailed labels and code examples to support the application of machine learning models in data race detection, addressing the challenges of data race detection in concurrent programming.

提供机构:
爱荷华州立大学
创建时间:
2023-08-17
二维码
社区交流群
二维码
科研交流群
商业服务