UCF Professor of Optics and Photonics Guifang Li is leading a five-year, $19.2 million Army Research Office-funded program to advance the understanding and control of light traveling through complex environments. The multi-institutional team will combine optical sensing, computational imaging, inverse methods and artificial intelligence to explore ways to measure and correct distortions in light. The award will support the next chapter of UCF’s Townes Laser Institute while providing research and training opportunities for students across participating universities.

JMIR Publications invites submissions to a new theme issue on Advancing Meaningful Engagement With mHealth Interventions in its open access journal JMIR mHealth and uHealth.

To address the dual challenges of reducing fuel consumption and ensuring maritime safety under extreme weather, researchers from Dalian University of Technology, Nanyang Technological University, and Chongqing Jiaotong University have developed a novel ship weather routing framework based on Dual-Mode Deep Reinforcement Learning. Utilizing the D3QN algorithm, this framework effectively navigates both normal sea states and severe typhoon conditions, providing a robust route toward maritime safety security and decarbonization.

Insilico Medicine ("Insilico", HKEX: 3696), a clinical-stage drug discovery company powered by generative artificial intelligence (AI), today announced the nomination of ISM9546, an orally bioavailable, peripheral-restricted VAV1 molecular glue degrader for the potential treatment of autoimmune diseases including IBD (inflammatory bowel disease) and RA (rheumatoid arthritis). Empowered by Insilico’s proprietary Pharma.AI solution, ISM9546 not only marks the 35^th PCC nomination of the company since 2021, but also strikes a balance between efficacy and safety, with nanomolar activity validated across cell assays, and peripherally restricted trait to alleviate risks related to the CNS (central nervous system).

Researchers developed Found-RL, an open-source platform that integrates vision-language models into reinforcement learning for autonomous driving. By using an asynchronous batch inference framework, Found-RL allows large foundation models to provide semantic feedback without slowing down high-frequency simulation training. Experiments in CARLA show that lightweight reinforcement learning agents can achieve competitive driving performance while maintaining real-time policy inference.

Researchers developed COIN, a collaborative and interaction-aware reinforcement learning framework for multi-agent autonomous driving. COIN enables vehicles to learn both individual navigation and system-level cooperation, improving safety and efficiency in dense scenarios such as intersections, roundabouts, and bottlenecks. The framework was evaluated in simulation and further validated on a hybrid robotic platform.

Scientists at the University of the Witwatersrand (Wits) have developed a “light-powered brain” that can recognise hidden information in complex patterns of light, potentially paving the way for faster communications systems.