RebeccaYU920/ifeval-pp
收藏Hugging Face2025-12-18 更新2025-12-20 收录
下载链接：
https://hf-mirror.com/datasets/RebeccaYU920/ifeval-pp
下载链接
链接失效反馈官方服务：
资源简介：
IFEval++是一个评估数据集，旨在评估大型语言模型在指令遵循方面的可靠性和鲁棒性。该数据集基于论文《Revisiting the Reliability of Language Models in Instruction-Following》提出，并扩展了原始的IFEval基准测试，包括系统的清理、增强和验证。IFEval++包含原始IFEval示例和增强示例，以进行更全面的评估。
IFEval++ is an evaluation dataset introduced in the paper "Revisiting the Reliability of Language Models in Instruction-Following". It is designed to assess instruction-following reliability and robustness in large language models. IFEval++ extends the original IFEval benchmark with systematic cleaning, augmentation, and validation. IFEval++ contains both the original IFEval examples and augmented examples for more comprehensive evaluation.
提供机构：
RebeccaYU920
