We analyzed the globular cluster M5 (NGC 5904) using 15 years of gamma-ray data from the Fermi Large Area Telescope (LAT). Using rotation ephemerides generated from Arecibo and FAST radio telescope observations, we searched for gamma-ray pulsations from the seven millisecond pulsars (MSPs) identified in M5. We detected no significant pulsations from any of the individual pulsars. Also, we searched for possible variations of the gamma-ray emission as a function of orbital phase for all the six MSPs in binary systems, but did not detect any significant modulations. The gamma-ray emission from the direction of M5 is well described by an exponentially cutoff power-law spectral model, although other models cannot be excluded. The phase-averaged emission is consistent with being steady on a time scale of a few months. We estimate the number of MSPs in M5 to be between 1 and 10, using the gamma-ray conversion efficiencies for well-characterized gamma-ray MSPs in the Third Fermi Large Area Telescope Catalog of Gamma-ray Pulsars, suggesting that the sample of known MSPs in M5 is (nearly) complete, even if it is not currently possible to rule out a diffuse component of the observed gamma rays from the cluster.

The hydrodynamic interactions among bacterial cell bodies, flagella, and surrounding boundaries are essential for understanding bacterial motility in complex environments. In this study, we demonstrate that each slender flagellum can be modeled as a series of spheres, and that the interactions between these spheres can be accurately characterized using a resistance matrix. This approach allows us to effectively and efficiently evaluate the propulsive effects of the flagella. Notably, our investigation into bacterial motility near a colloidal sphere reveals significant discrepancies between results derived from the twin multipole moment and those obtained through resistive force theory. Consequently, neglecting the hydrodynamic interactions among cell bodies, flagella, and colloidal spheres may lead to substantial inaccuracies. Our model simplifies bacteria into a series of spheres, making it well-suited for examining bacterial motility near spherical boundaries, as well as the nonlinear deformation dynamics of elastic flagella.

Talent search is a cornerstone of modern recruitment systems, yet existing approaches often struggle to capture nuanced job-specific preferences, model recruiter behavior at a fine-grained level, and mitigate noise from subjective human judgments. We present a novel framework that enhances talent search effectiveness and delivers substantial business value through two key innovations: (i) leveraging LLMs to extract fine-grained recruitment signals from job descriptions and historical hiring data, and (ii) employing a role-aware multi-gate MoE network to capture behavioral differences across recruiter roles. To further reduce noise, we introduce a multi-task learning module that jointly optimizes click-through rate (CTR), conversion rate (CVR), and resume matching relevance. Experiments on real-world recruitment data and online A/B testing show relative AUC gains of 1.70% (CTR) and 5.97% (CVR), and a 17.29% lift in click-through conversion rate. These improvements reduce dependence on external sourcing channels, enabling an estimated annual cost saving of millions of CNY.