Blue mold caused by Penicillium italicum (P. italicum) is a major postharvest disease in citrus fruits. Lipoic acid (LA) is a potent antioxidant with biological activity that was evaluated for its inhibitory effects on P. italicum and citrus blue mold using in vitro and in vivo experiments. The results demonstrated that LA effectively suppressed the mycelial growth and spore germination of P. italicum. LA increased hydrogen peroxide levels, compromising cell membrane integrity and leading to enhanced membrane permeability, as indicated by the increased relative conductivity and decreased protein and total sugar contents in P. italicum mycelia. Furthermore, LA delayed disease progression in citrus fruits infected with P. italicum through increasing total phenol and flavonoid contents and enhancing the activities of phenylalanine ammonia lyase, polyphenol oxidase, superoxide dismutase, and peroxidase in citrus peel. Overall, LA exhibited strong antifungal activity against P. italicum and improved citrus fruit resistance to blue mold, highlighting its potential as a natural postharvest disease control agent.

The concept and connotation of the smart learning environment based on multi-space integration was firstly introduced. Also, a smart learning environment system model based on multi-space integration was demonstrated through taking the Central China Normal University as an example. Then, the evaluation indexes for smart learning environment in colleges and universities based on multi-space integration were proposed initially and verified by the Delphi method. According to the verification result, three first-level evaluation indexes were determined. Moreover, 26 second-level evaluation indexes were also determined on the basis of the first-level evaluation indexes. In the end, an evaluation index system for smart learning environment in colleges and universities based on multi-space integration was constructed. It can provide a reference for constructing and evaluating smart learning environment in colleges and universities, and effectively promote the reform and development of classroom teaching.

To classify sentences in cardiovascular German doctor's letters into eleven section categories, we used pattern-exploiting training, a prompt-based method for text classification in few-shot learning scenarios (20, 50 and 100 instances per class) using language models with various pre-training approaches evaluated on CARDIO:DE, a freely available German clinical routine corpus. Prompting improves results by 5-28% accuracy compared to traditional methods, reducing manual annotation efforts and computational costs in a clinical setting.