Nondeterministic Polynomial-time Problem Challenge (NPPC)
收藏资源简介:
NPPC是由香港理工大学等机构提出的一个推理能力评估数据集,旨在构建一个不可被攻克、不可被破解、可自动验证的通用推理基准。该数据集包含25个著名的NP完全问题,能够生成任意数量和任意难度级别的实例,以适应不断增长的复杂性。这些问题被设计为能够反映现实世界问题,并为评估大型语言模型在解决复杂推理问题方面的性能提供了一种标准化的方法。
NPPC is a reasoning ability evaluation dataset proposed by institutions including the Hong Kong Polytechnic University, which aims to construct a universal reasoning benchmark that is unbreakable, uncrackable and automatically verifiable. This dataset includes 25 well-known NP-complete problems, and is capable of generating instances of arbitrary quantity and difficulty levels to accommodate increasing complexity. These problems are designed to reflect real-world problems, and provide a standardized framework for evaluating the performance of large language models (LLMs) in solving complex reasoning tasks.




