EFFIBENCH-X
收藏资源简介:
EFFIBENCH-X是一个多语言基准数据集,旨在衡量大型语言模型生成的代码效率。该数据集支持Python、C++、Java、JavaScript、Ruby和Golang等多种编程语言,并包含来自各种平台的竞争性编程任务和人类专家解决方案作为效率基准。数据集涵盖了复杂的问题,需要高级算法和数据结构,以更好地评估LLM在挑战性场景下的效率。此外,EFFIBENCH-X还提供了一个全面的评估框架,确保LLM生成的代码效率的可靠测量。该数据集适用于研究LLM优化技术,以改善各种编程语言中的代码效率。
EFFIBENCH-X is a multilingual benchmark dataset designed to evaluate the code efficiency generated by Large Language Models (LLMs). This dataset supports multiple programming languages including Python, C++, Java, JavaScript, Ruby, and Golang, and incorporates competitive programming tasks from various platforms alongside human expert solutions as efficiency benchmarks. It covers complex problems that require advanced algorithms and data structures, enabling more robust assessment of LLM efficiency in challenging scenarios. Furthermore, EFFIBENCH-X provides a comprehensive evaluation framework to ensure reliable measurement of the code efficiency generated by LLMs. This dataset is applicable for researching LLM optimization techniques to improve code efficiency across various programming languages.




