BigGSM++

该数据集是一个扩展的基准，它为像DeepSeek-R1这样的大型语言模型提供了更为全面的推理能力评估。此外，它还特别关注于优化推理大型语言模型的性能评估需求，并在文本模态实验中进行了相关任务的研究。

This dataset is an extended benchmark that provides more comprehensive evaluations of the reasoning capabilities for large language models (LLMs) such as DeepSeek-R1. Furthermore, it specifically addresses the performance evaluation requirements for optimizing reasoning-focused large language models, and conducts research on relevant tasks in text-modality experiments.