In the U.S. historically, local newspapers drew their content largely from newswires like the Associated Press. Historians argue that newswires played a pivotal role in creating a national identity and shared understanding of the world, but there is no comprehensive archive of the content sent over newswires. We reconstruct such an archive by applying a customized deep learning pipeline to hundreds of terabytes of raw image scans from thousands of local newspapers. The resulting dataset contains 2.7 million unique public domain U.S. newswire articles, written between 1878 and 1977. Locations in these articles are georeferenced, topics are tagged using customized neural topic classification, named entities are recognized, and individuals are disambiguated to Wikipedia using a novel entity disambiguation model. To construct the Newswire dataset, we first recognize newspaper layouts and transcribe around 138 millions structured article texts from raw image scans. We then use a customized neural bi-encoder model to de-duplicate reproduced articles, in the presence of considerable abridgement and noise, quantifying how widely each article was reproduced. A text classifier is used to ensure that we only include newswire articles, which historically are in the public domain. The structured data that accompany the texts provide rich information about the who (disambiguated individuals), what (topics), and where (georeferencing) of the news that millions of Americans read over the course of a century. We also include Library of Congress metadata information about the newspapers that ran the articles on their front pages. The Newswire dataset is useful both for large language modeling - expanding training data beyond what is available from modern web texts - and for studying a diversity of questions in computational linguistics, social science, and the digital humanities.
Whose Language Counts as High Quality? Measuring Language Ideologies in Text Data Selection
University of Michigan（密歇根大学）；University of New Mexico（新墨西哥大学）；Emory University（埃默里大学）；University of Washington（华盛顿大学）；Allen Institute for AI（艾伦人工智能研究所）
Who did What: A Large-Scale Person-Centered Cloze Dataset
Toyota Technological Institute at Chicago（芝加哥丰田技术学院）
我们构建了一个新的“Who-did-What”数据集，该数据集包含了来自LDC英语Gigaword新闻专线语料库的20多万个填空（完形填空）多项选择题。WDW数据集具有各种新颖的功能。首先，与CNN和每日邮报数据集（Hermann et al.，2015）我们避免使用文章摘要来形成问题。相反，每个问题都是由两个独立的文章组成的-一篇文章作为要阅读的段落，另一篇单独的文章关于用于形成问题的相同事件。第二，我们避免匿名化-每个选择都是一个命名实体的人。第三，这些问题已经过过滤，去除了一小部分可以通过简单的基线轻松解决的问题，而剩下的84%可以由人类解决。我们报告标准系统的性能基准，并提出WDW数据集作为社区的挑战性任务。
Entry Separation using a Mixed Visual and Textual Language Model: Application to 19th century French Trade Directories
CRH-EHESS（CRH-法国社会科学高等研究院）；EPITA Research & Development Laboratory (LRDE)（EPITA研究与开发实验室（LRDE））；LASTIG；Univ. Gustave Eiffel（古斯塔夫·埃菲尔大学）；IGN-ENSG（法国国家地理学院-国立地理科学学院）
Approaches to Analysing Historical Newspapers Using LLMs
Institute of Contemporary History（当代史研究所）；Faculty of Arts, University of Ljubljana（卢布尔雅那大学文学院）；Faculty of Computer Science, University of Ljubljana（卢布尔雅那大学计算机科学学院）
Processing the structure of documents: Logical Layout Analysis of historical newspapers in French
Centre de Recherches Interdisciplinaires et Transculturelles (CRIT)（跨学科与跨文化研究中心（CRIT））；Université de Bourgogne Franche-Comté（勃艮第-弗朗什-孔泰大学）；Institut Universitaire de France (IUF)（法国大学研究院（IUF））
HistNERo: Historical Named Entity Recognition for the Romanian Language
Research Institute for Artificial Intelligence “Mihai Drăgănescu”, Romanian Academy（罗马尼亚科学院“米哈伊·德勒格内斯库”人工智能研究所）；National University of Science and Technology POLITEHNICA Bucharest（布加勒斯特理工国立科技大学）
