The brain solves the credit assignment problem remarkably well. For credit to be assigned across neural networks they must, in principle, wait for specific neural computations to finish. How the brain deals with this inherent locking problem has remained unclear. Deep learning methods suffer from similar locking constraints both on the forward and feedback phase. Recently, decoupled neural interfaces (DNIs) were introduced as a solution to the forward and feedback locking problems in deep networks.

SWE-Bench (pronounced "swee bench") launched in November 2024 as a way to evaluate an AI model's coding skill. It has since quickly become one of the most popular tests in AI. A SWE-Bench score has become a mainstay of major model releases from OpenAI, Anthropic, and Google--and outside of foundation models, the fine-tuners at AI firms are in constant competition to see who can rise above the pack. Despite all the fervor, this isn't exactly a truthful assessment of which model is "better." Entrants have begun to game the system--which is pushing many others to wonder whether there's a better way to actually measure AI achievement.

Designing concept cars seems kind of like being back in grade school, when kids are encouraged to dream up things like a bedroom with a bouncy-house floor or a spaceship with an ice cream machine on board. At least concept cars have a chance of making it to production at some point, even if that timeline is a long way off. At Kia's EV Day in Barcelona, Spain in March, the brand unveiled a new modular electric van it's calling the Platform Beyond Vehicle (PBV). The PV5 is the first in the automaker's plan, with four variants: Cargo, Passenger, Crew, and a Wheelchair Access Vehicle option. The designers pushed that a little further with the PV5 WKNDR concept, an EV made for camping and overlanding.

If you're a smartphone owner--and chances are that's everyone reading this--you've probably encountered an eerie, but all too common scenario: One day you're talking about a random topic while your phone is next to you and the following day you notice ads start popping up related to that same topic. How do these ads know what you were talking about? Your smartphone may be the culprit. Every smartphone has its built-in microphone constantly turned on in order for the virtual assistant to hear your voice commands. So, could it be that these devices are also secretly eavesdropping on your conversations in order to serve you ads? Here's everything you need to know, plus a simple test to find out.

If you use Google Chrome on your mobile phone, or if you have a modern Android phone, then you've probably stumbled across the Discover feed at some point. The Discover feed is available on Chrome's mobile New Tab page, in the Google app, and on the home screen (by swiping right). Soon, it'll also be available on desktop PCs. Google Discover is a personalized recommendation engine that shows you articles from around the web that Google thinks you'd be interested in. The recommendations are based on various factors like your location, your browsing history, your opted-in interests, and more.

If you're a smartphone owner--and chances are that's everyone reading this--you've probably encountered an eerie, but all too common scenario: One day you're talking about a random topic while your phone is next to you and the following day you notice ads start popping up related to that same topic. How do these ads know what you were talking about? Your smartphone may be the culprit. Every smartphone has its built-in microphone constantly turned on in order for the virtual assistant to hear your voice commands. So, could it be that these devices are also secretly eavesdropping on your conversations in order to serve you ads? Here's everything you need to know, plus a simple test to find out.

Out-of-Distribution (OOD) detection is a critical task in machine learning, particularly in safety-sensitive applications where model failures can have serious consequences. However, current OOD detection methods often suffer from restrictive distributional assumptions, limited scalability, and a lack of interpretability. To address these challenges, we propose STOOD-X, a two-stage methodology that combines a Statistical nonparametric Test for OOD Detection with eXplainability enhancements. In the first stage, STOOD-X uses feature-space distances and a Wilcoxon-Mann-Whitney test to identify OOD samples without assuming a specific feature distribution. In the second stage, it generates user-friendly, concept-based visual explanations that reveal the features driving each decision, aligning with the BLUE XAI paradigm. Through extensive experiments on benchmark datasets and multiple architectures, STOOD-X achieves competitive performance against state-of-the-art post hoc OOD detectors, particularly in high-dimensional and complex settings. In addition, its explainability framework enables human oversight, bias detection, and model debugging, fostering trust and collaboration between humans and AI systems. The STOOD-X methodology therefore offers a robust, explainable, and scalable solution for real-world OOD detection tasks.

Since the release of ChatGPT in late 2022, generative artificial intelligence has trickled down from adults in their offices to university students in campus libraries to teenagers in high school hallways. Now it's reaching the youngest among us, and parents and teachers are grappling with the most responsible way to introduce their under-13s to a new technology that may fundamentally reshape the future. Though the terms of service for ChatGPT, Google's Gemini and other AI models specify that the tools are only meant for those over 13, parents and teachers are taking the matter of AI education into their own hands. Inspired by a story we published on parents who are teaching their children to use AI to set them up for success in school and at work, we asked Guardian readers how and why – or why not – others are doing the same. Though our original story only concerned parents, we have also included teachers in the responses published below, as preparing children for future studies and jobs is one of educators' responsibilities as well.

We study a general clustering setting in which we have n elements to be clustered, and we aim to perform as few queries as possible to an oracle that returns a noisy sample of the weighted similarity between two elements. Our setting encompasses many application domains in which the similarity function is costly to compute and inherently noisy. We introduce two novel formulations of online learning problems rooted in the paradigm of Pure Exploration in Combinatorial Multi-Armed Bandits (PE-CMAB): fixed confidence and fixed budget settings. For both settings, we design algorithms that combine a sampling strategy with a classic approximation algorithm for correlation clustering and study their theoretical guarantees. Our results are the first examples of polynomial-time algorithms that work for the case of PE-CMAB in which the underlying offline optimization problem is NP-hard.

In neuroscience, recurrent neural networks (RNNs) are modeled as continuous-time dynamical systems to more accurately reflect the dynamics inherent in biological circuits. However, convolutional neural networks (CNNs) remain the preferred architecture in vision neuroscience due to their ability to efficiently process visual information, which comes at the cost of the biological realism provided by RNNs. To address this, we introduce a hybrid architecture that integrates the continuoustime recurrent dynamics of RNNs with the spatial processing capabilities of CNNs. Our models preserve the dynamical characteristics typical of RNNs while having comparable performance with their conventional CNN counterparts on benchmarks like ImageNet. Compared to conventional CNNs, our models demonstrate increased robustness to noise due to noise-suppressing mechanisms inherent in recurrent dynamical systems. Analyzing our architecture as a dynamical system is computationally expensive, so we develop a toolkit consisting of iterative methods specifically tailored for convolutional structures. We also train multi-area RNNs using our architecture as the front-end to perform complex cognitive tasks previously impossible to learn or achievable only with oversimplified stimulus representations. In monkey neural recordings, our models capture time-dependent variations in neural activity in higher-order visual areas. Together, these contributions represent a comprehensive foundation to unify the advances of CNNs and dynamical RNNs in vision neuroscience.