The learning rate is 2e-5 with 100% warm up and a batch size of 32.

A diversity of tasks use language models trained on semantic similarity data.

Scott Adams, the US cartoonist who wrote and illustrated the comic strip Dilbert, has died of cancer at the age of 68. His ex-wife Shelly Miles announced his death on Tuesday during a live stream of his podcast, Real Coffee with Scott Adams. The satirical cartoon strip - about a competent but frustrated engineer and his dysfunctional workplace environment - was first published in 1989, and went on to feature in more than 2,000 newspapers in 65 countries. The character also later appeared in books, an animated TV series and video game. But in 2023, his comic strip was cancelled by newspapers including the Washington Post after Adams was accused of making racist comments about black people.

Could Ukraine hold a presidential election right now? Will Europe use frozen Russian assets to fund war? How can Ukraine rebuild China ties? 'Ukraine is running out of men, money and time' Each week, Myroshnyk Vassyl Savych heads north to deliver his newspaper to border communities exposed to Russian fire and disinformation. Editor-in-Chief Myroshnyk Vassyl Savych gets ready to deliver his weekly newspaper, Zorya Visnyk (The Dawn Bulletin), from his office in Zolochiv, in Ukraine's Kharkiv region, to front-line villages in November 2025 [Louis Lemaire/Al Jazeera] Editor-in-Chief Myroshnyk Vassyl Savych gets ready to deliver his weekly newspaper, Zorya Visnyk (The Dawn Bulletin), from his office in Zolochiv, in Ukraine's Kharkiv region, to front-line villages in November 2025 [Louis Lemaire/Al Jazeera] It's a cold, foggy morning in early November, and Myroshnyk Vassyl Savych is driving north on a narrow road in eastern Ukraine towards the Russian border. He's headed to villages where, owing to increasing exposure to Russian fire, only a fraction of residents remain. The war has cut them off from regular services. They no longer receive mail, and Russian transmitters often overpower or interfere with their Ukrainian mobile-phone signals. Before large-scale signal jamming was introduced to counter drones, Russian television and radio channels were accessible on televisions and radios in border-area communities. In his trunk are bundles of Zorya Visnyk ( The Dawn Bulletin), a local newspaper that Vassyl edits and delivers to front-line communities in Ukraine's Kharkiv region.

A diversity of tasks use language models trained on semantic similarity data. While there are a variety of datasets that capture semantic similarity, they are either constructed from modern web data or are relatively small datasets created in the past decade by human annotators.

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. news wire 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.

Micron to invest $9.6 billion in western Japan plant, report says Signage at the Micron Technology booth at the China International Import Expo in Shanghai is seen on Nov. 6. Micron Technology will spend ¥1.5 trillion ($9.6 billion) to build a plant in western Japan to make memory chips for artificial intelligence applications, Nikkei newspaper reported. The move comes as Micron looks to diversify advanced chip production outside of Taiwan, Nikkei said, citing people familiar with the matter. The new factory will manufacture high-bandwidth memory (HBM) chips, a key component for working with AI processors such as those made by Nvidia, according to the report. Micron will build the facility within the compound of its Hiroshima plant, starting in May, with plans to launch HBM shipments around 2028, Nikkei said. The Ministry of Economy, Trade and Industry will subsidize up to ¥500 billion of the costs for the project, the newspaper said.

AI is rapidly transforming journalism, but the extent of its use in published newspaper articles remains unclear. We address this gap by auditing a large-scale dataset of 186K articles from online editions of 1.5K American newspapers published in the summer of 2025. Using Pangram, a state-of-the-art AI detector, we discover that approximately 9% of newly-published articles are either partially or fully AI-generated. This AI use is unevenly distributed, appearing more frequently in smaller, local outlets, in specific topics such as weather and technology, and within certain ownership groups. We also analyze 45K opinion pieces from Washington Post, New York Times, and Wall Street Journal, finding that they are 6.4 times more likely to contain AI-generated content than news articles from the same publications, with many AI-flagged op-eds authored by prominent public figures. Despite this prevalence, we find that AI use is rarely disclosed: a manual audit of 100 AI-flagged articles found only five disclosures of AI use. Overall, our audit highlights the immediate need for greater transparency and updated editorial standards regarding the use of AI in journalism to maintain public trust.

Existing full text datasets of U.S. public domain newspapers do not recognize the often complex layouts of newspaper scans, and as a result the digitized content scrambles texts from articles, headlines, captions, advertisements, and other layout regions. OCR quality can also be low. This study develops a novel, deep learning pipeline for extracting full article texts from newspaper images and applies it to the nearly 20 million scans in Library of Congress's public domain Chronicling America collection. The pipeline includes layout detection, legibility classification, custom OCR, and association of article texts spanning multiple bounding boxes. To achieve high scalability, it is built with efficient architectures designed for mobile phones. The resulting American Stories dataset provides high quality data that could be used for pre-training a large language model to achieve better understanding of historical English and historical world knowledge. The dataset could also be added to the external database of a retrieval-augmented language model to make historical information - ranging from interpretations of political events to minutiae about the lives of people's ancestors - more widely accessible. Furthermore, structured article texts facilitate using transformer-based methods for popular social science applications like topic classification, detection of reproduced content, and news story clustering. Finally, American Stories provides a massive silver quality dataset for innovating multimodal layout analysis models and other multimodal applications.

High quality summarization data remains scarce in under-represented languages. However, historical newspapers, made available through recent digitization efforts, offer an abundant source of untapped, naturally annotated data. In this work, we present a novel method for collecting naturally occurring summaries via Front-Page Teasers, where editors summarize full length articles. We show that this phenomenon is common across seven diverse languages and supports multi-document summarization. To scale data collection, we develop an automatic process, suited to varying linguistic resource levels. Finally, we apply this process to a Hebrew newspaper title, producing HEBTEASESUM, the first dedicated multi-document summarization dataset in Hebrew.