ykwon-hf/reasonIF
收藏Hugging Face2025-10-20 更新2025-10-25 收录
下载链接：
https://hf-mirror.com/datasets/ykwon-hf/reasonIF
下载链接
链接失效反馈官方服务：
资源简介：
ReasonIF是一个评估大型推理模型在推理过程中遵循用户指令的系统化基准。它包括多语言推理、格式化和长度控制等多个方面。该数据集用于检测模型在遵循推理指令方面的性能，特别是在任务难度增加时模型的表现。
ReasonIF is a systematic benchmark for evaluating how large reasoning models adhere to user instructions throughout their reasoning process. It spans multilingual reasoning, formatting, and length control. The dataset is used to assess the performance of models in following reasoning instructions, especially when task difficulty increases.
提供机构：
ykwon-hf
