Recent studies on transformer-based language models show that they can answer questions by reasoning over knowledge provided as part of the context (i.e., in-context reasoning). However, since the available knowledge is often not filtered for a particular question, in-context reasoning can be sensitive to distractor facts, additional content that is irrelevant to a question but that may be relevant for a different question (i.e., not necessarily random noise). In these situations, the model fails todistinguish the necessary knowledge to answer the question, leading to spurious reasoning and degraded performance. This reasoning failure contrasts with the model's apparent ability to distinguish its contextual knowledge from all the knowledge it has memorized during pre-training. Following this observation, we propose teaching the model to reason more robustly by folding the provided contextual knowledge into the model's parameters before presenting it with a question. Our method, RECKONING, is a bi-level learning algorithm that teaches language models to reason by updating their parametric knowledge through back-propagation, allowing them to answer questions using the updated parameters.

Recent studies on transformer-based language models show that they can answer questions by reasoning over knowledge provided as part of the context (i.e., in-context reasoning). However, since the available knowledge is often not filtered for a particular question, in-context reasoning can be sensitive to distractor facts, additional content that is irrelevant to a question but that may be relevant for a different question (i.e., not necessarily random noise). In these situations, the model fails todistinguish the necessary knowledge to answer the question, leading to spurious reasoning and degraded performance. This reasoning failure contrasts with the model's apparent ability to distinguish its contextual knowledge from all the knowledge it has memorized during pre-training. Following this observation, we propose teaching the model to reason more robustly by folding the provided contextual knowledge into the model's parameters before presenting it with a question.

Mental models are an important concept in UX and product design, but they need to be more readily embraced by the AI community. At one level, mental models often don't appear because they are routine patterns of our assumptions about an AI system. This is something we discussed at length in the process of putting together the latest volume of the Thoughtworks Technology Radar, a biannual report based on our experiences working with clients all over the world. For instance, we called out complacency with AI generated code and replacing pair programming with generative AI as two practices we believe practitioners must avoid as the popularity of AI coding assistants continues to grow. Both emerge from poor mental models that fail to acknowledge how this technology actually works and its limitations.

Automated driving systems face challenges in GPS-denied situations. To address this issue, kinematic dead reckoning is implemented using measurements from the steering angle, steering rate, yaw rate, and wheel speed sensors onboard the vehicle. However, dead reckoning methods suffer from drift. This paper provides an arc-length-based map matching method that uses a digital 2D map of the scenario in order to correct drift in the dead reckoning estimate. The kinematic model's prediction is used to introduce a temporal notion to the spatial information available in the map data. Results show reliable improvement in drift for all GPS-denied scenarios tested in this study. This innovative approach ensures that automated vehicles can maintain continuous and reliable navigation, significantly enhancing their safety and operational reliability in environments where GPS signals are compromised or unavailable.

Swarm-Merging, SwarMer, is a decentralized framework to localize Flying Light Specks (FLSs) to render 2D and 3D shapes. An FLS is a miniature sized drone equipped with one or more light sources to generate different colors and textures with adjustable brightness. It is battery powered, network enabled with storage and processing capability to implement a decentralized algorithm such as SwarMer. An FLS is unable to render a shape by itself. SwarMer uses the inter-FLS relationship effect of its organizational framework to compensate for the simplicity of each individual FLS, enabling a swarm of cooperating FLSs to render complex shapes. SwarMer is resilient to both FLSs failing and FLSs leaving to charge their battery. It is fast, highly accurate, and scales to remain effective when a shape consists of a large number of FLSs.

Fox News correspondent Matt Finn has the latest on the impact of AI technology that some say could outpace humans on'Special Report.' Former Google CEO Eric Schmidt said the tech industry will face a "reckoning" over artificial intelligence, comparing the potential dangers of the technology to the risks associated with social media when the platforms were first rolled out years ago. "What happened with social media is we, including myself, just offered social media because we had a simple model of how humans would use social media. But, instead, look at how social media was used to interfere in elections, to cause harm. People have died over social media," Schmidt told ABC News on Sunday.

If you've spent any time on Twitter lately, you may have seen a viral black-and-white image depicting Jar Jar Binks at the Nuremberg Trials, or a courtroom sketch of Snoop Dogg being sued by Snoopy. These surreal creations are the products of Dall-E Mini, a popular web app that creates images on demand. Type in a prompt, and it will rapidly produce a handful of cartoon images depicting whatever you've asked for. More than 200,000 people are now using Dall-E Mini every day, its creator says--a number that is only growing. A Twitter account called "Weird Dall-E Generations," created in February, has more than 890,000 followers at the time of publication.

If you've spent any time on Twitter lately, you may have seen a viral black-and-white image depicting Jar Jar Binks at the Nuremberg Trials, or a courtroom sketch of Snoop Dogg being sued by Snoopy. These surreal creations are the products of Dall-E Mini, a popular web app that creates images on demand. Type in a prompt, and it will rapidly produce a handful of cartoon images depicting whatever you've asked for. More than 200,000 people are now using Dall-E Mini every day, its creator says--a number that is only growing. A Twitter account called "Weird Dall-E Generations," created in February, has more than 890,000 followers at the time of publication.

Artist Stephanie Lepp hosts Reckonings, a narrative podcast that explores how people shift their political worldviews, transcend extremism, and make other kinds of transformative change. Recently, she has been experimenting with a maligned technology, deepfakes, to create Deep Reckonings, a series of synthetic videos that imagine controversial public figures having a reckoning; in the deepfake footage, Alex Jones, Brett Kavanaugh, and Mark Zuckerberg reflect on the damage they have inflicted on society. I spoke to Stephanie about the impact of her project, and the positive potential of deepfake technology. What inspired you to create Deep Reckonings? We think of social change as requiring large numbers of people pushing for change. But there's also the question of, what are the fewest number of people it would take to create broad-based social change?

This article is part of the Policing and Technology Project, a collaboration between Future Tense and the Tech, Law, & Security Program at American University Washington College of Law that examines the relationship between law enforcement, police reform, and technology. On Sept. 18 at noon Eastern, Future Tense will host "Power, Policing, and Tech," an online event about the role of technology in law enforcement reform. Public scrutiny around data-driven technologies in the criminal justice system has been on a steady rise over the past few years, but with the recent widespread Black Lives Matter mobilization, it has reached a crescendo. Alongside a broader reckoning with the harms of the criminal justice system, technologies like facial recognition and predictive policing have been called out as racist systems that need to be dismantled. After being an early adopter of predictive policing, the Santa Cruz, California, became the first city in the United States to ban its use.