PaRTE
收藏DataONE2023-05-22 更新2024-06-08 收录
下载链接：
https://search.dataone.org/view/sha256:7226410e29b96ea5876cc3f102eab1a80e63d1c534c1f584fb4c77e4b8db87bc
下载链接
链接失效反馈官方服务：
资源简介：
We present a test set of 1,126 sentence pairs to evaluate whether Recognizing Textual Entailment (RTE) models are robust to paraphrasing. We posit that if RTE models understand language, their predictions should be consistent across inputs that share the same meaning. We develop an evaluation RTE set consisting of paraphrased examples that we use to determine if models' predictions change when examples are paraphrased.
创建时间：
2023-11-08
