Estimating soil nutrient components is a key analytical methodology for creating fertilization maps for agricultural fields. To develop a compact laser-induced breakdown spectroscopy (LIBS) system for fertilizer map creation, this study examined the sensitivity enhancement of phosphorus (P) and potassium (K) in soil by adding magnetic field assistance using Halbach-arrayed neodymium magnets. The LIBS signal intensity was compared with the plant-available content determined by chemical analysis. The addition of a magnetic field ( 0.5 T) increased the spectral intensity by 1.3 times, owing to densification of the laser-induced plasma. Linearity in the calibration curve of P at 213.6 nm in soil was maintained up to 0.5 wt
This paper presents a multi-stage framework for detecting reclaimed slurs in multilingual social media discourse. It addresses the challenge of identifying reclamatory versus non-reclamatory usage of LGBTQ+-related slurs across English, Spanish, and Italian tweets. The framework handles three intertwined methodological challenges like data scarcity, class imbalance, and cross-linguistic variation in sentiment expression. It integrates data-driven model selection via cross-validation, semantic-preserving augmentation through back-translation, inductive transfer learning with dynamic epoch-level undersampling, and domain-specific knowledge injection via masked language modeling. Eight multilingual embedding models were evaluated systematically, with XLM-RoBERTa selected as the foundation model based on macro-averaged F1 score. Data augmentation via GPT-4o-mini back-translation to alternate languages effectively tripled the training corpus while preserving semantic content and class distribution ratios. The framework produces four final runs for the evaluation purposes where RUN 1 is inductive transfer learning with augmentation and undersampling, RUN 2 with masked language modeling pre-training, RUN 3 and RUN 4 are previous predictions refined via language-specific decision thresholds optimized via ROC analysis. Language-specific threshold refinement reveals that optimal decision boundaries vary significantly across languages. This reflects distributional differences in model confidence scores and linguistic variation in reclamatory language usage. The threshold-based optimization yields 2-5
Abstract Large Forbush decreases (FDs) in the Galactic cosmic ray flux were observed by the paired neutron monitor (SNM) and the vertical channel of the muon detector (SMDV) at the Antarctic Syowa Station in May and October, 2024. The maximum count rate depressions in SNM and SMDV are $$\sim$$ ∼ 10% and $$\sim$$ ∼ 5% in the May FD, respectively. In the October FD, the observed count rates decreased in two steps following two successive interplanetary shocks. The maximum count rate depression recorded by SNM (SMDV) is $$\sim$$ ∼ 7% ( $$\sim$$ ∼ 3%) in the first step, while it is $$\sim$$ ∼ 8% ( $$\sim$$ ∼ 5%) in the second step. It is demonstrated that the SNM count rate fraction relative to the total count rate in SNM and SMDV ( $$F_n$$ F n ) is a good real-time indicator of the temporal variation of the cosmic ray rigidity spectrum ( $$\Delta \Gamma (P,t)$$ Δ Γ ( P , t ) ), which is an important parameter for identifying the physical processes responsible for FD effects. It is verified that the variation of $$F_n$$ F n is consistent with $$\Delta \Gamma (P,t)$$ Δ Γ ( P , t ) calculated from the Global Fitting Analysis (GFA) at the Syowa Station. $$F_n$$ F n also shows a significant increase of $$\Delta \Gamma (P,t)$$ Δ Γ ( P , t ) during the ground-level enhancement (GLE) due to solar energetic particles on May 11 superposed on the FD recovery phase, implying that the total (Galactic + solar) cosmic ray spectrum is softened due to the contribution from GLE intensity steeply increasing with decreasing rigidity. This implies that $$F_n$$ F n can be also a good indicator of small GLEs which are sometimes difficult to identify in the count rate variation when masked by an FD and the associated diurnal anisotropy. Graphic Abstract
Unmanned aerial vehicles serving as aerial base stations can rapidly restore connectivity after disasters, yet abrupt changes in user mobility and traffic demands shift the quality of service trade-offs and induce strong non-stationarity. Deep reinforcement learning policies suffer from plasticity loss under such shifts, as representation collapse and neuron dormancy impair adaptation. We propose plasticity enhanced multi-agent mixture of experts (PE-MAMoE), a centralized training with decentralized execution framework built on multi-agent proximal policy optimization. PE-MAMoE equips each UAV with a sparsely gated mixture of experts actor whose router selects a single specialist per step. A non-parametric Phase Controller injects brief, expert-only stochastic perturbations after phase switches, resets the action log-standard-deviation, anneals entropy and learning rate, and schedules the router temperature, all to re-plasticize the policy without destabilizing safe behaviors. We derive a dynamic regret bound showing the tracking error scales with both environment variation and cumulative noise energy. In a phase-driven simulator with mobile users and 3GPP-style channels, PE-MAMoE improves normalized interquartile mean return by 26.3% over the best baseline, increases served-user capacity by 12.8%, and reduces collisions by approximately 75%. Diagnostics confirm persistently higher expert feature rank and periodic dormant-neuron recovery at regime switches.
