Engineers from Adamson University, the University of the Philippines Diliman, and the University of Santo Tomas have successfully converted a common kitchen scrap into a powerful tool for environmental cleanup. By transforming discarded saba banana peels into a specialized biochar, the team has created a low-cost, sustainable adsorbent capable of efficiently removing diesel oil from water. This approach not only addresses the significant challenge of hydrocarbon pollution from industrial activities and accidental spills but also provides a valuable new use for agricultural waste, aligning with the principles of a circular economy.

Foundation models are transforming financial engineering, with specialized financial foundation models emerging to tackle domain-specific needs. Categorized into three types, they excel at text processing, time-series analysis, and visual-text integration. This survey unpacks their progress, real-world uses, and key challenges, offering a vivid roadmap for AI-driven finance innovation.

From computer chips and cloud computing to artificial intelligence, Germany and the rest of Europe want to become less dependent on digital technologies from abroad by driving forward their own innovations. The Center for Advanced Security, Strategic and Integration Studies (CASSIS) at the University of Bonn has now prepared a policy brief, in cooperation with the Vodafone Institute for Society and Communications, Vodafone’s European think tank. The paper tackles the continent’s “digital trade deficit” and shows how hidden trade dependencies on the US and China are raising questions about the effectiveness of current European strategies. 

Grasping and transporting objects is one of the most critical tasks for robots in a variety of fields. This task requires accurate 3D measurement of the objects. However, transparent or specular objects make measurement challenging, reducing grasping success rate. To address this, researchers have developed HEAPGrasp, a new technique that performs 3D measurement of objects using only their silhouettes, thus avoiding dependence on their optical properties. This approach significantly improves grasping success rate of robots.