ASSET

ASSET是由谢菲尔德大学和Facebook AI Research等机构合作创建的一个用于评估英语句子简化模型的新数据集。该数据集包含23,590个人工简化的句子，每个原始句子对应10个简化版本，涵盖多种重写转换。ASSET旨在通过提供多样的简化实例，更好地理解和评估简化模型的能力。该数据集适用于非母语英语使用者、阅读障碍者等，有助于提高语言处理任务的效率和准确性。

ASSET is a novel dataset for evaluating English sentence simplification models, co-developed by the University of Sheffield, Facebook AI Research and other institutions. It includes 23,590 manually simplified sentences, with 10 simplified variants corresponding to each original sentence, covering a wide range of rewriting transformations. ASSET is designed to enable better comprehension and rigorous evaluation of the capabilities of sentence simplification models by providing diverse simplification instances. This dataset is applicable to non-native English speakers, people with dyslexia and other similar groups, and helps improve the efficiency and accuracy of language processing tasks.