ScisummNet
收藏资源简介:
ScisummNet是由耶鲁大学计算机科学系创建的一个大规模手动标注的科学论文摘要数据集,专注于计算语言学领域。该数据集包含1000篇最常被引用的ACL Anthology论文及其引用网络和专家编写的摘要。数据集的创建过程涉及专家阅读论文摘要和引用句,以快速创建全面的混合摘要。ScisummNet旨在解决科学论文摘要中的挑战,通过结合作者的原始亮点和研究社区的实际影响,提供一个新框架用于科学论文摘要研究。
ScisummNet is a large-scale manually annotated scientific paper summarization dataset developed by the Department of Computer Science at Yale University, focusing on the field of computational linguistics. This dataset comprises 1,000 most frequently cited papers from the ACL Anthology, along with their citation networks and expert-written summaries. The dataset construction process involves experts reading original paper abstracts and citation sentences to quickly create comprehensive hybrid summaries. ScisummNet aims to address the challenges in scientific paper summarization, providing a novel framework for scientific paper summarization research by combining authors' original highlights and the actual impacts of the research community.




