A comprehensive survey published May 23 in Intelligent Computing, a Science Partner Journal maps out the role of large language models in task planning, underscoring the growing influence of artificial intelligence in complex decision-making tasks.

A team led by researchers at MUSC Hollings Cancer Center developed an artificial intelligence tool that uses natural language processing to identify the original cancer type of patients with brain metastases by analyzing their doctors’ clinical notes. The model outperformed standard diagnostic codes, accurately determining the primary cancer type in over 90% of cases, which can improve research efforts and enable more precise care. The lightweight, scalable tool could streamline cancer care and research by unlocking critical insights buried in electronic health records.

A newly developed white organic light-emitting diode operates at under 1.5 volts, report researchers from Institute of Japan. By using triplet–triplet annihilation to generate blue light at low voltage and adding in yellow and sky-blue dopants, the research team achieved efficient white emission. This breakthrough could contribute to reducing energy consumption in the future in state-of-the-art displays, including television backlights and lighting devices, advancing portable electronics and sustainable technologies.