Long-separated research has been conducted on two highly correlated tracks: traffic and incidents. Traffic track witnesses complicating deep learning models, e.g., to push the prediction a few percent more accurate, and the incident track only studies the incidents alone, e.g., to infer the incident risk. We, for the first time, spatiotemporally aligned the two tracks in a large-scale region (16,972 traffic nodes) from year 2022 to 2024: our TraffiDent dataset includes traffic, i.e., time-series indexes on traffic flow, lane occupancy, and average vehicle speed, and incident, whose records are spatiotemporally aligned with traffic data, with seven different incident classes. Additionally, each node includes detailed physical and policylevel meta-attributes of lanes. Previous datasets typically contain only traffic or incident data in isolation, limiting research to general forecasting tasks.

Accident anticipation aims to predict potential collisions in an online manner, enabling timely alerts to enhance road safety. Existing methods typically predict frame-level risk scores as indicators of hazard. However, these approaches rely on ambiguous binary supervision--labeling all frames in accident videos as positive--despite the fact that risk varies continuously over time, leading to unreliable learning and false alarms. To address this, we propose a novel paradigm that shifts the prediction target from current-frame risk scoring to directly estimating accident scores at multiple future time steps (e.g., 0.1s-2.0s

Synaptic plasticity is widely considered to be crucial to the brain's ability to learn throughout life. Decades of theoretical work have therefore been invested in deriving and designing biologically plausible learning rules capable of granting various memory abilities to neural networks. Most of these theoretical approaches optimize directly for a desired memory function; but this procedure can lead to complex, finely-tuned rules, rendering them brittle to perturbations and difficult to implement in practice. Instead, we build on recent work that automatically discovers large numbers of candidate plasticity rules operating in recurrent spiking neural networks. Surprisingly, despite the fact that these rules are selected solely to achieve network stabilization, we observe across a range of network models -feedforward, recurrent; rate and spiking-that almost all these rules endow the network with simple forms of memory such as familiarity detection - seemingly by accident.

A year ago, Air India flight 171 crashed less than a minute after taking off from Ahmedabad airport in the western Indian state of Gujarat, en route for London. The official investigation that followed has sparked intense controversy, in India and beyond, with some questioning its integrity amid claims of conflicts of interest. It is not the first time such an investigation has proved contentious. So is it time for a different approach when investigating air crashes? It was a hot and dry afternoon on 12 June last year, when Flight 171 left the terminal at Sardar Vallabhbhai Patel Airport in Ahmedabad. Settling into their seats for the nine-and-a-half-hour journey to London were 230 passengers, 53 of them British citizens. Looking after them were 10 cabin crew.

The measure is part of a European Union-led strategy to eliminate all drunk-driving-related deaths and injuries by 2050. As of July 1, all vehicles sold within the European Union must include a standard, preinstalled interface that allows a breathalyzer lock to be added to the ignition system. This measure is part of a larger strategy promoted by the EU to reduce drunk-driving-related deaths and injuries by at least 50 percent by 2030. The requirement falls under the Vision Zero program, launched by European authorities more than five years ago, which aims to eliminate alcohol-related traffic fatalities entirely--or get as close to zero as possible--by 2050. The measure also aligns with the timetable established in the EU's General Safety Regulation, which sets specific deadlines for manufacturers to incorporate various safety features into vehicle designs, starting at the factory.

Nine people have died in an explosion at a coal mine in Colombia in the latest fatal accident to hit the country's mining sector. Emergency workers said they had rescued six miners from the shafts in Sutatausa, north of the capital, Bogotá. Colombia's national mining agency said a build-up of gases was thought to have caused the explosion at 16:00 (21:00 GMT) on Monday. It also published a list of recommendations it said it had made to the mine's operators after an inspection less than a month ago, in which it had warned of a potentially dangerous gas build-up. Many mines in Colombia are operated informally and without proper safety standards.

A blast at a fireworks factory in China's Hunan province has killed 21 people and left 61 wounded, according to state media. The explosion at the Changsha Liuyang Huasheng Fireworks plant happened at around 16:40 local time (08:40 GMT) on Monday, in the city of Liuyang, leading rescuers to evacuate everyone within a 3km (1.9mi) radius of the plant. Authorities deployed nearly 500 personnel to conduct search and rescue operations and treat the injured, while robots were used to help find those trapped within the building. Police, who are investigating the cause of the blast, have taken control measures against the person in charge of the fireworks company, Chinese state media reported. Authorities said that two gunpowder warehouses within the factory area posed a high risk amid rescue efforts, state media reported.

An autonomous robot from the company Honor ran a half marathon in 50:26, beating the human record by 7 minutes. A humanoid robot from the Honor remote-controlled team crosses the finish line during the E-Town Humanoid Robot Half Marathon in Beijing on April 19, 2026. Over the weekend in China, a humanoid robot shattered world half-marathon record--the human record--by seven minutes. The star performer was a robot developed by the Chinese company Honor (the smartphone maker), which finished the 13.1-mile race in 50 minutes, 26 seconds. The human record, set by Ugandan Olympic medalist Jacob Kiplimo, is 57 minutes, 20 seconds.

This article was featured in the One Story to Read Today newsletter. T he smell was strange . The concrete wall was too close. One of my kids was standing on the sidewalk next to our car--not crying, just confused. The seat belt had held. The crumple zone had crumpled.

While recent developments in autonomous vehicle (AV) technology highlight substantial progress, we lack tools for rigorous and scalable testing. Real-world testing, the de facto evaluation environment, places the public in danger, and, due to the rare nature of accidents, will require billions of miles in order to statistically validate performance claims. We implement a simulation framework that can test an entire modern autonomous driving system, including, in particular, systems that employ deep-learning perception and control algorithms. Using adaptive importance-sampling methods to accelerate rare-event probability evaluation, we estimate the probability of an accident under a base distribution governing standard traffic behavior. We demonstrate our framework on a highway scenario, accelerating system evaluation by 2-20 times over naive Monte Carlo sampling methods and 10-300P times (where P is the number of processors) over real-world testing.