Understanding the spatio-temporal patterns of taxi operations is crucial for urban transport planning and management. A recent study by researchers from Shandong University of Science and Technology and the Chinese Academy of Sciences introduces a novel model, the Taxi Mobility Chain (TMC), to analyze the continuous movement patterns of taxis in Xiamen, China. By comparing traditional taxis and e-hailing cars, the study reveals significant differences in their operational behaviors, providing valuable insights for optimizing urban transportation systems.

A novel lensless imaging technique, utilizing a programmable Fresnel zone aperture, offers high-resolution, artifact-free imaging with compact, cost-effective hardware. This method enhances resolution by 2.5 times and improves the signal-to-noise ratio by 3 dB compared to traditional approaches. The system's adaptability between static and dynamic modes enables diverse applications, from biomedical imaging to virtual reality and human-computer interaction, paving the way for miniaturized and multifunctional imaging platforms.

Research team developed PathActMarker, a software package for pathway activity inference and analysis. It preprocesses gene expression data, converts it into pathway activity matrices using eight tools, and ranks pathways to identify biomarkers.

Research team introduced CHASER, an incentive mechanism for blockchain-based EMCS systems. It ensures budget balance, truthfulness, rationality, and high social welfare. Simulations show a 42% increase in social welfare and high task completion rates.

This study focused on exploring the distribution patterns, driving factors and microbial interaction patterns of methane-metabolizing microorganisms along the sedimentation gradient in the Yellow River estuary.

Researchers have demonstrated a novel approach to generating terahertz (THz) radiation using N-polar AlGaN/GaN high-electron-mobility transistors (HEMTs) under asymmetric boundary conditions. This groundbreaking work offers new insights into the design of high-power on-chip THz sources, essential for next-generation technologies such as 6G communications, non-invasive imaging, and biological monitoring.

Research team presented FairAT, a novel fair adversarial training algorithm. It improves robustness of hard classes by augmenting hard examples dynamically, outperforming state-of-the-art methods in robustness and fairness.