yale-nlp/SciArena-with-paperbank
收藏Hugging Face2025-08-25 更新2025-09-13 收录
下载链接：
https://hf-mirror.com/datasets/yale-nlp/SciArena-with-paperbank
下载链接
链接失效反馈官方服务：
资源简介：
SciArena是一个开放的科学文献任务基础模型评估平台，通过社区投票的方式对模型进行比较，采用集体智能进行模型性能评估。数据集包含训练集和测试集，测试集为从训练集中随机抽取的2000个样本，作为元评估基准。
SciArena is an open collaborative platform for evaluating foundation models on scientific literature tasks, using community voting to compare models and leveraging collective intelligence for performance assessment. The dataset includes a training set and a test set, with the test set being a collection of 2,000 randomly sampled examples from the training set serving as a meta-evaluation benchmark.
提供机构：
yale-nlp
