tasksource/PLANE-ood
收藏数据集概述
基本信息
- 许可证: cc-by-2.0
- 任务类别: 文本分类
- 语言: 英语
- 数据集大小: 100K<n<1M
数据集结构
-
特征:
seq: 字符串类型,测试序列label: 字符串类型,标签(1: 蕴含,0: 非蕴含)Adj_Class: 字符串类型,序列形容词的类别Adj: 字符串类型,形容词(I: 内含的,S: 次类的,O: 意向的)Nn: 字符串类型,名词Hy: 字符串类型,名词的超类
-
分割:
train: 300132个样本,26047744字节test: 10080个样本,874524字节
-
下载大小: 4721262字节
-
数据集大小: 26922268字节
数据集用途
PLANE (phrase-level adjective-noun entailment) 是一个基准测试,用于测试模型在细粒度组合推理上的表现。该数据集包含五个采样分割,用于Bertolini et al., 22中的监督实验。
引用信息
若使用PLANE数据集,请引用COLING 2022的主要论文:
@inproceedings{bertolini-etal-2022-testing, title = "Testing Large Language Models on Compositionality and Inference with Phrase-Level Adjective-Noun Entailment", author = "Bertolini, Lorenzo and Weeds, Julie and Weir, David", booktitle = "Proceedings of the 29th International Conference on Computational Linguistics", month = oct, year = "2022", address = "Gyeongju, Republic of Korea", publisher = "International Committee on Computational Linguistics", url = "https://aclanthology.org/2022.coling-1.359", pages = "4084--4100", }



