Classifier Detected Datasets (PI-CD)
收藏arXiv2025-09-30 收录
下载链接:
https://github.com/tyliupku/nli-debiasing-datasets
下载链接
链接失效反馈官方服务:
资源简介:
该数据集的概要显示,在假设句子上训练的神经分类器能够识别SNLI测试集中的“困难”实例。这项任务的目的是进行自然语言推理。
The summary of this dataset indicates that neural classifiers trained on hypothesis sentences can identify the "hard" instances in the SNLI test set. The purpose of this task is to perform natural language inference.
提供机构:
Original authors



