yale-nlp/LimitGen
收藏Hugging Face2025-07-03 更新2025-08-09 收录
下载链接:
https://hf-mirror.com/datasets/yale-nlp/LimitGen
下载链接
链接失效反馈官方服务:
资源简介:
LimitGen Benchmark是一个用于评估大型语言模型在辅助同行评审,尤其是在识别论文局限性方面的综合基准数据集。它包括两个子集:LimitGen-Syn,一个包含11种常见问题子类型的合成数据集;以及LimitGen-Human,一个包含ICLR 2025论文提交和人类编写的评审评论的数据集。
LimitGen Benchmark is a comprehensive benchmark for evaluating large language models capability to assist in peer review, particularly in identifying paper limitations. It consists of two subsets: LimitGen-Syn, a synthetic dataset with 11 common issue subtypes; and LimitGen-Human, a dataset containing ICLR 2025 paper submissions and human-written review comments.
提供机构:
yale-nlp



