The April 16 attack was claimed by Tehreek-e-Taliban Pakistan (TTP), commonly known as the Pakistan Taliban, and its fighters then opened fire from all sides of the Sarki Piyala police checkpost in Hangu district, about 30km (18 miles) from the border with Afghanistan and a hotbed of deadly attacks by armed groups, mainly the TTP. "There was a great deal of fear because we were in the middle of a battle. At that moment, we did not even know whether we would make it back home alive," Khan told Al Jazeera. To defend against such attacks, the police in Hangu came up with an unlikely, low-cost solution - fishing nets strung several metres overhead between the checkpost's rooftops and perimeter walls, forming a kind of aerial ceiling. Simple in construction and resembling what fishermen use for their work, the netting was first installed two months ago, according to the local police.

Understanding and attributing mental states, known as Theory of Mind (ToM), emerges as a fundamental capability for human social reasoning. While Large Language Models (LLMs) appear to possess certain ToM abilities, the mechanisms underlying these capabilities remain elusive. In this study, we discover that it is possible to linearly decode the belief status from the perspectives of various agents through neural activations of language models, indicating the existence of internal representations of self and others' beliefs. By manipulating these representations, we observe dramatic changes in the models' ToM performance, underscoring their pivotal role in the social reasoning process. Additionally, our findings extend to diverse social reasoning tasks that involve different causal inference patterns, suggesting the potential generalizability of these representations.

The Defense Advanced Research Projects Agency (DARPA) has made a system that can catch drones mid-flight. Instead of risking damage when drones need to land in battlefields or on US Navy Ships, the DARPA SideArm capture system can retrieve drones up to 1100 pounds (500 kg) in weight. The system can fit in a shipping container and can be set up and operated by two to four people, enabling the SideArm to be portable. In December 2016, the system was tested with a 400-pound (181 kg) Lockheed Martin Fury Unmanned Aerial System (UAS) drone. Aurora Flight Sciences, who tested the SideArm, accelerated the drone to speeds that it would fly at using an external catapult.