zenless-lab/jamp
收藏Hugging Face2024-12-18 更新2024-12-21 收录
下载链接：
https://hf-mirror.com/datasets/zenless-lab/jamp
下载链接
链接失效反馈官方服务：
资源简介：
Jamp是一个用于评估语言模型泛化能力的日语时间推理基准数据集。该数据集由模板、测试数据和训练数据组成，特别关注时间格式和时间跨度的处理。数据集由tomo-vv创建，使用日语，遵循CC BY-SA 4.0许可证。
Jamp is a Japanese temporal inference benchmark dataset designed to evaluate the generalization capacity of language models. The dataset includes templates, test data, and training data, and is split based on tense fragment, time format, or time span. Created by tomo-vv, the dataset is in Japanese and licensed under CC BY-SA 4.0.
提供机构：
zenless-lab
