A sign showing bear deterrent measures is displayed next to a wolf-shaped robotic device outside a convenience store in a rural area of Kiryu, Gunma Prefecture. Five people were hospitalized after a man accidentally set off an anti-bear spray in a post office in Nagoya's city center, reports and officials said, as Japan grapples with a sharp rise in maulings. A foreign national, who was not named, apologized and told police that he discharged the repellant unintentionally, the reports said. "Eight people felt sick, and of those five were sent to hospital. But there was no report of serious injuries or illness," fire department official Ryohei Asano said after the mishap on Wednesday. At least five people have been killed by bears in Japan since April 1 after a record 13 deaths last fiscal year.

With origins in the English countryside, The Cotswold Company is known for upscale furniture that evokes its bucolic backstory. It's a step-change for a business that's tracked the shift from mail-order catalogs and showrooms to websites. In its simplest form, AI means searching for items via chatbots, but agentic AI could one day see Cotswold's customers represented by bots picking and buying products like a personal shopper. "There is always first-mover advantage with these things," said Chief Executive Officer Ralph Tucker, who joined in 2020 and is AI-readying a business that did 80% of its £123 million ($163 million) in sales online in its latest fiscal year. "If our customers are there, we need to be there and we need to be brilliant at it." In a time of both misinformation and too much information, quality journalism is more crucial than ever.

Constitutional lawyer Bruce Fein says the US was founded on the principle that governments exist to protect inalienable rights. He argues expanded presidential powers and unchecked authority represents a step backwards for US democracy. How AI is being weaponised against India's Muslim women

The world's oceans are under heat stress, with average sea surface temperatures hitting 21 C, surpassing the record highs of 2023 and 2024. They're expected to rise further as El Niño, a natural climate pattern that warms the tropical Pacific for months, develops. How AI is being weaponised against India's Muslim women

Most existing multitask learning approaches are limited by their reliance on task-specific loss functions tailored to the scale and type of each outcome. When outcomes differ across tasks, these losses are generally not directly comparable, which makes it difficult to formulate a unified objective and may limit information sharing across tasks. We propose a multitask transformation framework in which task-specific responses may differ through unknown monotone transformations. Motivated by high-dimensional biological applications in which the predictor dimension may diverge with the sample size while only a common subset of predictors is informative, we consider shared sparsity across tasks. Under this framework, we estimate the target functions and identify important predictors by optimizing a smoothed rank-based criterion with a group-Lasso penalty, implemented through a multitask deep neural network with a shared first layer. We establish the nonasymptotic excess-risk bounds, and variable-selection consistency for the proposed estimator. Simulation studies show that the proposed method achieves competitive prediction and variable-selection performance compared with competing approaches. Analyses of gene-expression studies with continuous, binary, and mixed outcomes further illustrate that the proposed method improves prediction and identifies biologically meaningful shared predictors.

Knowing which training examples drive outputs is fundamental to auditing, correcting, and understanding language models, yet for modern LLMs this remains expensive, approximate, and largely post-hoc. Standard language models generate tokens through a dense network pathway, causing training data's influence to be distributed across parameters rather than organized along explicit, traceable components. We introduce a prototype language model architecture, Prototypes for Interpretable Sequence Modeling (PRISM), that forms each prediction via a sparse, non-negative mixture of learned prototypes, trained with clustering objectives that anchor each prototype to coherent neighborhoods of training examples. Across architectures from 130M to 1.6B parameters trained on up to 50B tokens, prototype language models either surpass or remain within 2.5 percentage points on average downstream accuracy of matched dense baselines. We show that sparse prototype structure localizes curvature in the loss landscape, yielding a more tractable Hessian and enabling training data attribution that is ~500x faster than post hoc baselines when consuming equivalent memory. Calibrating linear prototype controllers can improve downstream accuracy by roughly 3 points while tracing those corrections back to training neighborhoods, and targeted prototype suppression can remove model behaviors without finetuning or measurable loss in generation quality.

Traditional variational Kalman filtering with unknown noise statistics suffers from inconsistent process covariance estimation and slow convergence speed, limiting its practical utility. To address these issues, we introduce a surrogate variable representing the process-noise-free state, which enables explicit modeling and inference of process noise statistics. In addition, we reformulate the conventional coordinate ascent variation inference (CAVI) as a marginalized maximum a posteriori problem, followed by a single-step hyperparameter fitting. This reformulation obviates the need for multiple inner iterations inherent to CAVI and decouples the design of the covariance tracking filters. Consequently, this architecture permits the deployment of higher-order filters for covariance tracking and enables sliding-window hyperparameter estimation. Notably, when this window encompasses all historical data, the covariance tracking estimator intrinsically operates as a zero-phase filter. Numerical simulations validate the theoretical framework, demonstrating the enhanced convergence speed and superior estimation accuracy compared with existing methods.

'They played really well in the enemy's country.' Iran's football team has landed back in Tehran after their World Cup exit to hundreds of fans warmly welcoming them home. How AI is being weaponised against India's Muslim women

The US and Israel have signed a deal allocating land for a permanent US embassy in West Jerusalem, years after a temporary one was established during Trump's first term in office. The move is yet another blow to the hopes of a future Palestinian capital. How AI is being weaponised against India's Muslim women

Bolivia's President Paz declared a state of emergency after 50 days of protests against his policies. Workers are angry, accusing him of abandoning them with austerity cuts and privatisation. How AI is being weaponised against India's Muslim women