S2ORC

包含 8110 万篇英语学术论文的大型语料库，涵盖多个学科。丰富的元数据、论文摘要、已解决的参考书目以及 810 万篇开放获取论文的结构化全文。使用自动检测的引文、图形和表格的内联提及注释的全文，每个都链接到其相应的论文对象。将来自数百家学术出版商和数字档案馆的论文汇总到一个统一的来源中，并创建迄今为止最大的公开可用的机器可读学术文本集合。

A large corpus containing 81.1 million English academic papers spanning multiple disciplines. It includes rich metadata, paper abstracts, resolved bibliographic references, as well as structured full texts for 8.1 million open access papers. Full texts are annotated with inline mentions of automatically detected citations, figures and tables, each linked to its corresponding paper object. It aggregates papers from hundreds of academic publishers and digital archives into a unified repository, and creates the largest publicly available machine-readable academic text collection to date.

搜集汇总

背景与挑战

背景概述

S2ORC是一个大型英语学术论文语料库，包含8110万篇论文，涵盖多学科，提供丰富的元数据和结构化全文，由华盛顿大学·艾伦人工智能研究所发布。