Topic modeling is a well-established technique for exploring text corpora. Conventional topic models (e.g., LDA) represent topics as bags of words that often require "reading the tea leaves" to interpret; additionally, they offer users minimal semantic control over topics. To tackle these issues, we introduce TopicGPT, a prompt-based framework that uses large language models (LLMs) to uncover latent topics within a provided text collection. TopicGPT produces topics that align better with human categorizations compared to competing methods: for example, it achieves a harmonic mean purity of 0.74 against human-annotated Wikipedia topics compared to 0.64 for the strongest baseline. Its topics are also more interpretable, dispensing with ambiguous bags of words in favor of topics with natural language labels and associated free-form descriptions. Moreover, the framework is highly adaptable, allowing users to specify constraints and modify topics without the need for model retraining. TopicGPT can be further extended to hierarchical topical modeling, enabling users to explore topics at various levels of granularity. By streamlining access to high-quality and interpretable topics, TopicGPT represents a compelling, human-centered approach to topic modeling.

A Tesla owner charges his vehicle in April 2021 at a charging station in Topeka, Kan.. Tesla reported 273 crashes involving partially automated driving systems, according to statistics released by U.S. safety regulators on Wednesday. A Tesla owner charges his vehicle in April 2021 at a charging station in Topeka, Kan.. Tesla reported 273 crashes involving partially automated driving systems, according to statistics released by U.S. safety regulators on Wednesday. Automakers reported nearly 400 crashes of vehicles with partially automated driver-assist systems, including 273 involving Teslas, according to statistics released Wednesday by U.S. safety regulators. The National Highway Traffic Safety Administration cautioned against using the numbers to compare automakers, saying it didn't weight them by the number of vehicles from each manufacturer that use the systems, or how many miles those vehicles traveled. Automakers reported crashes from July of last year through May 15 under an order from the agency, which is examining such crashes broadly for the first time.

Tax laws, for example, don't currently take automated workers into account. While human employees contribute payroll and income taxes, an automated "employee" doesn't, Abbott noted. Governments could lose out on quite a bit of income tax as AI becomes more prevalent and possibly displaces more human workers. Granted, that argument only works if displaced employees don't find other jobs. Abbott predicted that that may happen as AI becomes smarter at a rate that outpaces people's ability to learn new skills or find job training.

Atima Lui was in primary school when she first learned that "nude" is not universal. Now 30, she still recalls playing with a white friend's makeup and struggling to find colours that complemented her rich skin tone. "I would try to put [her makeup] on and it would just make me look like a clown," says Lui, who is of Sudanese and African-American descent. "I think back to growing up and how my mother barely wore makeup. Now I know it's because makeup just wasn't made for her."

Kansas WebFile, the state's full service online tax filing portal, has a new feature. "We were the first state to implement a chatbot within a tax system, and we're the first state to offer a chatbot of any nature," Nolan Jones general manager of Kansas Information Consortium said. Agent Kay, the state's 24/7 official chatbot, can help you file your tax return online. "And it provides them a better experience with dealing with the government and some like taxes they don't want to be in there very long. They just want to get their filing done," Jones said.

SpaceX's CRS-15 Dragon cargo resupply ship has been attached to the International Space Station. the spacecraft rendezvoused with the orbiting outpost in the early-morning hours of July 2, 2018, and is expected to remain berthed for about a month. Capture took place at 6:54 a.m. EDT (10:54 GMT) by the 57.7-foot (17.6-meter) Canadian-built robotic Canadarm2, which was under the control of Expedition 56 NASA astronauts Ricky Arnold and Drew Feustel at the robotics work station in the station's cupola window. The vehicle was grappled while the station was flying 256 miles (412 kilometers) over Quebec City. "Looking forward to some really exciting weeks ahead as we unload the science and get started on some great experiments," Arnold said.

Artificial intelligence, robots, and automation are rising in importance in many areas. As noted in the recent book, "The Future of Work: Robots, AI, and Automation," there are exciting advances in finance, transportation, national defense, smart cities, and health care, among other areas. Businesses are developing solutions that improve the efficiency and effectiveness of their operations and using these tools to improve the way their firms function. Yet there also are concerns about the impact of these developments on jobs and personal privacy. A Pew Research Center national survey revealed considerable unease about emerging trends.

Robustness to adversarial perturbation has become an extremely important criterion for applications of machine learning in security-sensitive domains such as spam detection [25], fraud detection [6], criminal justice [3], malware detection [13], and financial markets [27]. Systematic methods for generating adversarial examples by small perturbations of original input data, also known as "attack," have been developed to operationalize this criterion and to drive the development of more robust learning systems [4, 26, 7]. Most of the work in this area has focused on differentiable models with continuous input spaces [26, 7, 14, 14]. In this setting, the proposed attack strategies add a gradient-based perturbation to the original input. It has been shown that such perturbations can result in a dramatic decrease in the predictive accuracy of the model. Thus this line of research has demonstrated the vulnerability of deep neural networks to adversarial examples in tasks like image classification and speech recognition. We focus instead on adversarial attacks on models with discrete input data, such as text data, where each feature of an input sample has a categorical domain. While gradient-based approaches are not directly applicable to this setting, variations of gradient-based approaches have been shown effective in differentiable models. For example, Li et al. [15] proposed to locate the top features with the largest gradient magnitude of their embedding, and Papernot et al. [20] proposed to modify randomly selected features of an input through perturbing each feature by signs of the gradient, and project them onto the closest vector in the embedding space.

On Wednesday the Department of Transportation announced the launch of a pilot program that will lead to new regulations. On Wednesday the Department of Transportation announced the launch of a pilot program that will lead to new regulations. Here's a hypothetical: How tolerant would you be of a drone flying over your head or zooming through your backyard, if it were carrying life-saving medicine to the scene of a hard-to-reach accident? The U.S. Department of Transportation plans to collect the answers to questions like this, and a slew of other data, in a new test project called the Integration Pilot Program. After combing through 149 applications from state, local and tribal governments seeking to partner with some of the world's leading technology companies, Transportation Secretary Elaine Chao announced the winners Wednesday.

These are the horrifying photos from the heyday of the Eugenics movement that the world wants to forget. Before the atrocities of Nazi Germany, eugenics - the system of measuring human traits, seeking out the desirable ones and cutting out the undesirable ones - was once practised the world over. In the decades following the 1859 publication of Charles Darwin's'On the Origin of Species', a veritable craze for eugenics spread through Britain, the United States and Europe. Bruno Beger was a German racial anthropologist who worked for the Ahnenerbe, a project in Nazi Germany to research the archaeological and cultural history of the Aryan race. In this image taken in 1938, Dr Beger is measuring a Tibetan woman's head to demonstrate what he believed were the'inferior' characteristics of her race Advocates of eugenics made significant advances during the early twentieth century - and claimed that'undesirable' genetic traits such as dwarfism, deafness and even minor defects like a cleft palate needed to be wiped out of the gene pool.