This study examined the evolution of digital education policy in the United Kingdom (UK) from 2008 to 2024 based on a discourse analysis of 21 policy documents retrieved from the UK government’s official website, GOV.UK. 

The work introduces a large-language-model framework that predicts a student’s future performance and generates natural-language diagnoses from only a handful of activity records, thereby eliminating the data hunger and opacity that have long limited knowledge-tracing systems.

New evidence from the gravitational-wave event GW230814 shows that when two black holes merge, the horizon area of the resulting black hole is very likely greater than the sum of the horizon areas of the two progenitors. An independent analysis by the Purple Mountain Observatory team finds a significance level up to about 4.1σ, supporting the classic black-hole “area law” originally proposed by Stephen Hawking and further reinforcing general relativity in the highly dynamical, strong-field merger regime.

Qing Wei and colleagues from the College of Engineering, China Agricultural University, systematically elaborated on the innovative applications of neural networks in agricultural product drying, offering new insights to address industry pain points. The related article has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025620).

Researchers have developed a biomimetic mRNA delivery system that enhances the efficacy of PTEN mRNA in colorectal cancer immunotherapy. By using metal-ion coordination and cell-membrane coating, the system improves mRNA loading, cytosolic delivery, and stability, outperforming traditional lipid nanoparticles. The study also identifies a patient subgroup likely to benefit from PTEN-based treatment, provides an effective mRNA delivery system and conceptual proof for precise cancer immunotherapy.

A review by Professor Abdul Shukor JURAIMI’s team from Universiti Putra Malaysia points out that hyperspectral imaging technology boasts advantages of non-contact operation, high precision, and early detection. Compared with traditional manual visual inspection, it can complete detection within 10–30 days after rice sowing—a critical period when weeds are most competitive—with an identification accuracy generally exceeding 90%. For example, regarding Echinochloa crus-galli and weedy rice (Oryza sativa f. spontanea), the most common weeds in rice fields, researchers achieved identification accuracies of 100% and 92%, respectively, by analyzing spectral data with intelligent algorithms. This accurate identification lays the foundation for targeted weeding: combined with UAVs and prescription mapping technology, it enables site-specific herbicide application, reducing pesticide usage by up to 50%. This not only cuts costs but also alleviates environmental burdens. The relevant article has been published in Frontiers of Agricultural Science and Engineering (DOI: 10.15302/J-FASE-2025619).

Two novel 56π-electron fullerene derivatives, bearing bulky functional groups and named C₆₀-TFB and C₆₀-TFP, were developed and employed as electron transport layers in inverted perovskite solar cells (IPSCs). These new fullerene derivatives demonstrate excellent resistance to dimerization and effective defect passivation, which helps prevent the interdiffusion of iodide ions from the perovskite layer and silver atoms from the cathode. Consequently, the optimized IPSC achieved a high efficiency of 25.93% along with significantly enhanced stability.