rubricreward/R3-Dataset-20K
收藏Hugging Face2025-06-21 更新2025-07-05 收录
下载链接:
https://hf-mirror.com/datasets/rubricreward/R3-Dataset-20K
下载链接
链接失效反馈官方服务:
资源简介:
R3-Dataset-20k是一个包含20,000个例子的数据集,用于训练R3系列的鲁棒评分无关奖励模型。数据集从超过一百万个公开数据集中精选而来,涵盖了一般聊天、推理和分类任务,并为每个例子添加了评分量表生成和解释轨迹。
R3-Dataset-20k is a dataset containing 20,000 examples curated for training the R3 series of robust rubric-agnostic reward models. The dataset is selected from over one million publicly available datasets, covering general chat, reasoning, and classification tasks, and each example is enriched with rubric generation and explanation traces.
提供机构:
rubricreward



