ServiceNow-AI/mt-bench_audio
收藏Hugging Face2025-09-21 更新2025-09-13 收录
下载链接:
https://hf-mirror.com/datasets/ServiceNow-AI/mt-bench_audio
下载链接
链接失效反馈官方服务:
资源简介:
MT-Bench是一个用于评估大型语言模型作为评委的数据集,包含了问题ID、分类、对话轮次、参考答案、语言类型和音频文件等特征。数据集分为测试集,共有80个示例。数据集遵循Apache-2.0许可协议。
MT-Bench is a dataset for evaluating large language models as judges, containing features such as question ID, category, dialogue turns, reference answers, language type, and audio files. The dataset is split into a test set with a total of 80 examples. The dataset follows the Apache-2.0 license.
提供机构:
ServiceNow-AI



