Language change is influenced by many factors, but often starts from synchronic variation, where multiple linguistic patterns or forms coexist, or where different speech communities use language in increasingly different ways. Besides regional or economic reasons, communities may form and segregate based on political alignment. The latter, referred to as political polarization, is of growing societal concern across the world. Here we map and quantify linguistic divergence across the partisan left-right divide in the United States, using social media data. We develop a general methodology to delineate (social) media users by their political preference, based on which (potentially biased) news media accounts they do and do not follow on a given platform. Our data consists of 1.5M short posts by 10k users (about 20M words) from the social media platform Twitter (now "X"). Delineating this sample involved mining the platform for the lists of followers (n=422M) of 72 large news media accounts. We quantify divergence in topics of conversation and word frequencies, messaging sentiment, and lexical semantics of words and emoji. We find signs of linguistic divergence across all these aspects, especially in topics and themes of conversation, in line with previous research. While US American English remains largely intelligible within its large speech community, our findings point at areas where miscommunication may eventually arise given ongoing polarization and therefore potential linguistic divergence. Our methodology - combining data mining, lexicostatistics, machine learning, large language models and a systematic human annotation approach - is largely language and platform agnostic. In other words, while we focus here on US political divides and US English, the same approach is applicable to other countries, languages, and social media platforms.

Graph Neural Networks (GNNs) have attracted much attention due to their ability in learning representations from graph-structured data. Despite the successful applications of GNNs in many domains, the optimization of GNNs is less well studied, and the performance on node classification heavily suffers from the long-tailed node degree distribution. This paper focuses on improving the performance of GNNs via normalization. In detail, by studying the long-tailed distribution of node degrees in the graph, we propose a novel normalization method for GNNs, which is termed ResNorm (\textbf{Res}haping the long-tailed distribution into a normal-like distribution via \textbf{norm}alization). The $scale$ operation of ResNorm reshapes the node-wise standard deviation (NStd) distribution so as to improve the accuracy of tail nodes (\textit{i}.\textit{e}., low-degree nodes). We provide a theoretical interpretation and empirical evidence for understanding the mechanism of the above $scale$. In addition to the long-tailed distribution issue, over-smoothing is also a fundamental issue plaguing the community. To this end, we analyze the behavior of the standard shift and prove that the standard shift serves as a preconditioner on the weight matrix, increasing the risk of over-smoothing. With the over-smoothing issue in mind, we design a $shift$ operation for ResNorm that simulates the degree-specific parameter strategy in a low-cost manner. Extensive experiments have validated the effectiveness of ResNorm on several node classification benchmark datasets.

Douglas Adams created the most famous ebook reader – The Hitchhiker's Guide to the Galaxy – almost 30 years before the first Kindle was released, but he didn't restrict his ideas to his science fiction. In the late 1990s, at least a decade before Amazon's e-reader first came on to the market in 2007, the author and humorist made a series of notes uncannily predicting the rise of electronic books. But Adams, who died in 2001, did not live to see his musings, spread over three A4 pages, become reality. He wrote: "Lots of resistance to the idea of ebooks from the public. Particularly all those people who 10 years ago said they couldn't see any point typing on a computer. "I believe this resistance will gradually disappear as the electronic book itself improves and becomes smaller, lighter, simpler, cheaper, in other words more like a book." Adams's notes are presented in their original handwritten form in a new book, 42: The Wildly Improbable Ideas of Douglas Adams. Featuring unseen material from Adams's personal archive, including notes, letters, speeches, fanmail and unused sections of his most famous work, The Hitchhiker's Guide, it has been put together by Kevin Jon Davies, who first met Adams in 1978 to interview him for a fanzine. Davies gained access to Adams's archived material held at St John's College, Cambridge to assemble a suitably eclectic insight into the writer's thoughts, processes and ideas. He describes Adams as "a man fascinated by technology" and both an "advocate for conservation and a forward thinking innovator". "His ideas in his writing, articles and speeches were often arguably ahead of their time," says Davies. "The three pages of notes which are Douglas's thoughts on the future possibilities of electronic books and publishing date from the late 1990s, and the musings are well ahead of Kindles and other ebooks.

The major labor union representing performers and broadcasters may branch out its historic strike for better working conditions to include the performers who bring video games to life. On Friday, more than a month after its current strike began, the SAG-AFTRA union's national board announced that it was seeking a strike vote against major video game companies like Activision and Disney Character Voices International. After more than a year, according to the union, its negotiations for better pay and protections against artificial intelligence in gaming have reached an impasse. While video game performers are part of SAG-AFTRA, their contracts are separate from the theatrical, TV, and streaming contracts that other members are currently striking over. "Once again we are facing employer greed and disrespect," wrote SAG-AFTRA's president, Fran Drescher.

Recent years have witnessed the wide adoption of large language models (LLM) in different fields, especially natural language processing and computer vision. Such a trend can also be observed in recommender systems (RS). However, most of related work treat LLM as a component of the conventional recommendation pipeline (e.g., as a feature extractor) which may not be able to fully leverage the generative power of LLM. Instead of separating the recommendation process into multiple stages such as score computation and re-ranking, this process can be simplified to one stage with LLM: directly generating recommendations from the complete pool of items. This survey reviews the progress, methods and future directions of LLM-based generative recommendation by examining three questions: 1) What generative recommendation is, 2) Why RS should advance to generative recommendation, and 3) How to implement LLM-based generative recommendation for various RS tasks. We hope that the survey can provide the context and guidance needed to explore this interesting and emerging topic.

This study investigates and suggests typologies for examining Artificial Intelligence (AI) within the domains of journalism and mass communication research. We aim to elucidate the seven distinct subfields of AI, which encompass machine learning, natural language processing (NLP), speech recognition, expert systems, planning, scheduling, optimization, robotics, and computer vision, through the provision of concrete examples and practical applications. The primary objective is to devise a structured framework that can help AI researchers in the field of journalism. By comprehending the operational principles of each subfield, scholars can enhance their ability to focus on a specific facet when analyzing a particular research topic.

With the rapid evolution of large language models (LLMs), new and hard-to-predict harmful capabilities are emerging. This requires developers to be able to identify risks through the evaluation of "dangerous capabilities" in order to responsibly deploy LLMs. In this work, we collect the first open-source dataset to evaluate safeguards in LLMs, and deploy safer open-source LLMs at a low cost. Our dataset is curated and filtered to consist only of instructions that responsible language models should not follow. We annotate and assess the responses of six popular LLMs to these instructions. Based on our annotation, we proceed to train several BERT-like classifiers, and find that these small classifiers can achieve results that are comparable with GPT-4 on automatic safety evaluation. Warning: this paper contains example data that may be offensive, harmful, or biased.

ChatGPT has proven it can help students with their homework, but now it is helping teachers create those very courses, a computer science professor told Fox News. Christians are responding to a fake Bible passage reportedly generated by ChatGPT that said Jesus accepts trans-identified individuals, stating "there is no man nor woman." "And a woman, whose heart was divided between spirit and body, came before him," the fake passage reads. "In quiet despair, she asked, 'Lord, I come to you estranged, for my spirit and body are not one. How shall I hope to enter the kingdom of God?'" "Jesus looked upon her with kindness, replying, 'my child, blessed are those who strive for unity within themselves, for they shall know the deepest truths of my Father's creation,'" the passage continued.

Would seminal Beatles classic Abbey Road have been so memorable if it had been called Everest - and featured George Harrison smoking a cigarette in front of a snow-covered volcano on the cover instead of the Fab Four crossing the street in London? That is one of several questions posed by digital experts today - who have re-imagined how some of the world's most iconic album covers might have looked if they had been released under their original working titles. The study, from digital agency WMG, has used image generation technology instead of the names by which they are now known the world over. An AI bot has predicted what iconic album covers might have looked like if world-famous artists including The Beatles and Queen had plumped for the original record names. Queen's studio album The Miracle was released in 1989 and was named after a song included on the album tracklist Using the working titles of some of music's most legendary albums, SEO and digital marketing experts WMG used AI tool Midjourney to visualise what their covers could have looked like.

The union, which represents 160,000 people in the entertainment industry, has a separate contract with major video game companies that was due to expire in November. It was extended a year, however, to allow talks to continue.