Poor quality seed represents a constraint on introducing favorable traits to farmers and consumers. The bean seed system in Uganda comprises three subsystems: informal (dominated by purchases from grain markets); semiformal Quality-Declared Seed (QDS) from farmer groups; and formal / certified seed. We don’t know how seed quality varies across these subsystems, nor at which step(s) problems may occur. We assess dimensions of seed quality for three biofortified bean varieties at different stages in the Ugandan bean seed system. Working with the regulator, we collected 419 samples from different seed system stages (foundation, production, aggregation, distribution). Regulatory audits were augmented with DNA fingerprinting (DArTag genotyping) and assays for estimating iron and zinc concentration. We compared results from raw samples to samples that had been sorted using visual cues (color, shape, size) designed to mimic farmer seed processing behavior. Results indicate that seed quality loss begins immediately and persists systemwide, with QDS performing the best of the three subsystems. The modest improvement in genetic purity (+3%) achieved by visual sorting confirms that it is inadequate as a substitute for a functioning quality control system. Most biofortified seed samples fell far below the iron and zinc breeding targets (only 25% and 0.3% succeeded, respectively), a result that is independent of genetic purity, revealing a fundamental weakness in the breeding and/or seed multiplication processes. The study recommends overhauling the certified seed distribution through more effective regulation, quality assurance/quality control; further expansion of QDS and reappraising bean biofortification breeding approaches.

In civil engineering, estimating the probability of rare events is essential for ensuring the safety and reliability of structural systems. Scenarios such as failures in reinforced concrete elements, progressive collapse, and performance-based design require accurate evaluation of low-probability events. Traditional methods like Monte Carlo Simulation (MCS), the First-Order Reliability Method (FORM), and the Second-Order Reliability Method (SORM) are widely used due to their theoretical foundation and practical simplicity. However, MCS becomes computationally prohibitive for rare events, while FORM and SORM may lose accuracy in nonlinear or non-Gaussian problems, common in structural analysis. Subset Simulation (SuS) overcomes these limitations by decomposing the rare event into a sequence of intermediate events, sampled efficiently through Markov Chain Monte Carlo (MCMC) techniques. To reduce computational time, SuS can be parallelized by distributing the sampling across multiple processors, this strategy enables faster analysis without compromising accuracy, making it suitable for large-scale models in civil engineering. This work explores the use of MCS, FORM, SORM, and SuS within the Julia programming environment, which offers high-performance computing capabilities. The methods are evaluated based on accuracy, computational cost, and robustness across a range of examples, including benchmark problems and structural models of reinforced concrete. Results indicate that SuS delivers accurate estimates with significantly fewer samples. When parallelized, it becomes a computationally efficient tool for reliability analysis in civil engineering, outperforming traditional methods in complex scenarios.

Abstract We present a detailed temporal and spectral analysis of the blazar S5 1044+71 using multiwavelength data obtained from the Fermi-LAT and Swift-XRT/UVOT telescopes. Applying the Bayesian block algorithm to the 3-day binned γ -ray light curve, we identify pronounced variability, including four major outbursts marked by significant flux enhancements. The highest flux recorded was (1.1 ± 0.2) × 10 −6 photons cm −2 s −1 on 57868.5 MJD. Each outburst comprises multiple components, and light-curve profile analysis indicates predominantly symmetric temporal structures. The shortest variability timescale of 4.5 hr constrains the emission region to be located within 0.03 pc of the central engine, likely near the broad-line region (BLR). Additionally, two highest-energy photons were detected with energies of 46.4 GeV (on 57739.6 MJD) and 42.5 GeV (on 59161.9 MJD), observed outside the peak flaring activity. The fractional variability shows an overall increasing trend from UV/optical to γ -ray bands, with a noticeable dip in the X-ray range, consistent with the shape of the broadband spectral energy distribution (SED). The flux distributions during flares exhibit lognormal or double-lognormal behavior, suggesting multiplicative variability processes and evolving emission zones. Cross-correlation analysis reveals a strong positive correlation between the γ -ray and X-ray bands, with X-rays lagging by 42.5 days. Broadband SED modeling across different flux states supports a one-zone leptonic scenario, with γ -ray emission produced via external Compton scattering of IR and BLR photons. High flux states show harder electron spectra, elevated break energies, and reduced magnetic fields—features consistent with efficient particle acceleration and Compton dominance.

This paper proposes a time-domain parallel method that integrates the hybrid implicit-explicit finite-difference time-domain (HIE-FDTD) method with time-domain physical optics (TDPO), implemented using the message passing interface (MPI) library. Compared to conventional FDTD/TDPO hybrid approaches, the proposed method offers two main advantages, First, the HIE-FDTD algorithm, as a weakly conditionally stable implicit scheme, enables the use of significantly larger time steps by relaxing the Courant–Friedrichs–Lewy (CFL) constraint, and a sparse sample strategy combined with a sliding-window mechanism is employed in the near-to-far-field extrapolation process, to support long-time transient simulations. Second, that the method supports distributed parallelization across multiple processors, substantially reducing computational time. Several numerical examples are presented to validate the accuracy and efficiency of the proposed approach. The simulation results show good agreement with those obtained by traditional methods, and the parallel implementation consistently achieves approximately 80% efficiency across varying processor counts.

Data synthesised and published as response ratios in ecology ( , or ratio of means, ) remain isolated from broad secondary analyses because they cannot be converted to other effect size metrics. Here I address this lack of data interoperability by developing a conversion to the widely used Hedges' (standardised mean difference, ). This conversion is practical and near exact-as long as assumptions of homogeneity of variances are met, Hedges' correction is used to adjust for small-sample bias, and only additive and not multiplicative ecological processes are converted. I then generalise this conversion with abstract algebra to develop additional opportunities to reuse effect sizes-first by stating the response ratio as a geometric construction of Pythagorean means, and then as a proportional compass-and-straightedge construction of the response ratio. Constructability is a new pathway of interoperability for effect sizes, and without collecting new data, allows for the response ratio and to be repurposed into relative change datatypes such as the arithmetic, harmonic, geometric, quadratic and logarithmic means. Much of what has been synthesised in ecology is only available as response ratios, and I hope these conversions increase their value post-publication and facilitate reuse for bolder, more comprehensive meta-analyses.

The Quantum Internet is key for distributed quantum computing, by interconnecting multiple quantum processors into a virtual quantum computation system. This allows for scaling the number of qubits by overcoming the inherent limitations of noisy intermediate-scale quantum devices. Thus, the Quantum Internet is the foundation for large-scale, fault-tolerant quantum computation. Among the distributed architectures, quantum data centres emerge as the most viable in the medium term, since they integrate multiple quantum processors within a localized network infrastructure, by allowing modular design of quantum networking. We analyse the physical and topological constraints of quantum data centres, by emphasizing the role of entanglement orchestrators in dynamically reconfiguring network topologies through local operations. We examine the major hardware challenge of quantum transduction, essential for interfacing heterogeneous quantum systems. Furthermore, we explore how interconnecting multiple Quantum Data Centres could enable large-scale quantum networks. We discuss the topological constraints of such a scaling and identify open challenges, including entanglement routing and synchronization. The analysis positions quantum data centres as both a practical implementation platform and strategic framework for the future quantum internet. This article is part of the discussion meeting issue 'Bits, neurons and qubits for sustainable AI'.

Fully Homomorphic Encryption (FHE) is a promising privacy-preserving technique that has drawn increasing attention from academia and industry. It allows computation directly on encrypted data without decryption. However, FHE incurs intensive computations. Chiplet-based designs integrate multiple processors, delivering high performance and thereby are embraced by computation-intensive FHE tasks. Despite the chiplet-based system with various processors, it is designed for unencrypted applications, falling short in handling FHE with unique ciphertext manipulations. One common approach to make it capable of FHE is developing a new FHE accelerator. However, this approach overlooks existing abundant resources already in the system and introduces a large area overhead. In this paper, we propose Chiplever, a framework that empowers a non-FHE-tailored system to efficiently support FHE tasks via a hardware extension. Chiplever aims to leverage the existing resources already in the room for FHE tasks. To achieve this, (1) Chiplever introduces a hardware extension with an FHE unit providing efficient function support for FHE operators. (2) Chiplever proposes an FHE coordinator in the extension, which enables direct ciphertext transfer between the newly introduced extension and existing chiplets, achieving efficient integration of the extension. (3) Chiplever lowers the high-level homomorphic operations to primitive operators that can be matched by existing chiplets and constructs a fine-grained computation graph. Based on this, Chiplever employs a task scheduling algorithm, which partitions the FHE task across the extension and existing chiplets to exploit the parallelism between them and reduce the ciphertext communication overheads. With these hardware and software optimizations, Chiplever achieves efficient FHE acceleration. Compared with prior FHE ASICs, Chiplever achieves 9.6× 15.9× speedup and 6.2× 67.4× throughput improvement on TFHE, while consuming only 18.8% 35.6% of the area overhead of dedicated FHE ASICs.

Automotive System-on-Chips (SoCs) must meet stringent functional safety standards, such as ISO 26262 and IEC 61508, to ensure reliable operation under hardware faults. FPGA-based fault injection has emerged as a practical and cost-effective technique for functional safety verification. However, instrumentation-based methods face scalability challenges when applied to the high fault densities typical of automotive SoCs. To address these challenges, we propose a hybrid cascaded fault-injection controller architecture (HCCA-SAFE) that simultaneously reduces high-fanout global nets and eliminates long serial propagation paths. The architecture constrains enable-signal cluster width and distributes control across cascaded stages, improving timing results and routability under limited FPGA resources. The proposed architecture is evaluated on multiple open-source RISC-V processor cores. On openE902, HCCA-SAFE reduces net delay from 27.276 ns to 22.535 ns and achieves 32.2% and 63.8% lower net delay compared with the representative centralized and shift-chain approaches, respectively. On openE906, the proposed HCCA-SAFE limits the net delay to 12.959 ns and reduces the maximum control-signal fanout to 1763, respectively, compared with 25.825 ns and 40.442 ns in the conventional method. On openC906, the proposed design lowers the maximum control-signal fanout from 7725 to 570 and reduces the net delay to 7.506 ns. Furthermore, HCCA-SAFE produces results fully consistent with software-based RTL simulation, while delivering substantial performance gains. Speed-up factors of 127×, 206×, and 2123× are achieved on openE902, openE906, and openC906, respectively, with efficiency improvements scaling with processor complexity These results confirm that HCCA-SAFE delivers scalable, timing-robust fault-injection control suitable for large automotive SoCs.

We study asymptotic properties of supercritical Galton-Watson (GW) branching processes in the asymptotic where the mean of the offspring distribution approaches 1 from above. We show that the population-size distribution of the GW branching processes at a sufficiently large generation in this asymptotic can be approximated by a compound Poisson-gamma distribution. Numerical experiments revealed that the compound Poisson-gamma models were in good agreement with the corresponding GW models for sufficiently large generations under a reasonable parameter regime. Our results can be regarded as supporting the use of the compound Poisson-gamma model as a model for cascaded multiplication processes.

Abstract To support the development of the data processing pipeline and the scientific performance assessment for the Cool Planet Imaging Coronagraph (CPI-C) on the China Space Station Telescope (CSST), we have developed the end-to-end instrument simulation program, CPISM. This paper details the core modules of CPISM that simulate the CPI-C instrument, focusing on the simulation of the high-contrast imaging optical system and the visible-band science camera. We modeled key optical components, such as the transmission apodizing filter, the wavefront corrector, and the focal plane mask using the HCIPy package. A $10^{−8}$ contrast dark hole region, consistent with design specifications, was simulated using the Electric Field Conjugation (EFC) optimization method, and broadband observation effects were considered. For the science camera, which is an electron multiplying charge-coupled device (EMCCD), we established a detailed model encompassing photon collection, charge transfer, electron multiplication (EM), and readout processes, based on test data. This model simulates complex instrumental features including dark current, charge transfer efficiency, clock-induced charge, multiplication noise factor, and various readout effects like striping and drift. We also proposed and validated an improved statistical model for the EM process to enhance simulation efficiency. CPISM can generate simulated images containing rich instrumental details, closely similar to the expected real observational data, thus laying the foundation for the development and verification of CPI-C data processing algorithms and preparations for future scientific research.

The development of multifunctional artificial synapses capable of integrating molecular sensing and physical stimuli detection remains a challenge due to limited material systems that exhibit both multi-modal responsiveness and tunable synaptic characteristics. Herein, we report two-dimensional perovskite oxide La2Ti2O7 (LTO) nanosheets that are semiconducting and ion conductive, rendering them sensitive to humidity, NO2 gas, and light. A two-terminal device based on the LTO nanosheets exhibited tunable synaptic behaviors and input-dependent switching between excitatory and inhibitory postsynaptic currents. Importantly, a long-term inhibitory memory was achieved, resulting from the light-triggered reaction between environmental H2O and NO2 molecules, which deprived H2O from the LTO surface to bring LTO to a high resistance state. As a proof-of-concept, our two-terminal device was employed for evaluation and early warning of potential acid rain scenarios, demonstrating its advantage of reducing system complexity compared to a conventional system that requires multiple sensors and logic processors. Our work provides a new strategy of using multi-functional sensing materials for artificial synapses in responding to environmental chemical processes.

Mobile Cloud Offloading (MCO) addresses resource limitations of mobile devices by migrat- ing compute-intensive tasks to powerful remote servers. This paper presents a predictive MCO framework that incorporates a machine learning-based decision engine to intelligently select the optimal execution environment among local, edge, and cloud resources. The framework is evalu- ated using diverse workloads including matrix multiplication and image processing. Experimental results demonstrate that the ML-driven predictive approach consistently achieves lower execu- tion latency compared to static and reactive offloading strategies, validating the effectiveness of context-aware, proactive decision-making in mobile computing environments.

Is field space infinite? If not, it either loops back on itself or ends altogether. Periodic boundary conditions are of course familiar, but field space endpoints—which appear in real-world systems—are far less explored. In this paper we argue that boundaries in field space are generic, radiatively stable structures that allow for new physics at very low scales not ruled out by experiment. Such boundaries are delocalized in field space from the vacuum, so they can only be accessed by coherent fields or high multiplicity processes, both of which are weakly constrained observationally. Low multiplicity interactions do not detect the boundary and instead perceive a “mirage cutoff” that is parametrically higher than the true cutoff of the theory. Hence, field space boundaries are deformations of the standard model that are Lorentz invariant, local, unitary at low energies, and experimentally unconstrained. We comment on the possibility of field space boundaries on the long-range force carriers and the Higgs, as well as possible implications for the hierarchy problem.

This paper investigates a cell-free massive multiple-input multiple-output enabled multi-access edge computing (termed CF-MEC) system, where multiple users are served by multiple central processing units (CPUs) and their connected access points (APs), both of which are equipped with computation resources. For this system, a dynamic user-centric task offloading scheme is designed to provide seamless and efficient computation services for users. Based on this scheme, the joint optimization of user-centric AP clustering, edge server selection, communication and computation resources is formulated as a long-term problem to minimize the average energy consumption. The formulated problem is complicated non-convex due to the highly coupled time-varying discrete and continuous variables, resulting in high complexity and non-real-time to obtain the optimal solution. To tackle this challenging problem, we propose a multi-layer hierarchical multi-agent deep reinforcement learning (ML-HMADRL) based resource allocation algorithm. Specifically, the proposed algorithm incorporates a hierarchical structure with high, middle, and low-level agents that iteratively train the actor-critic networks to obtain discrete and continuous variables of the formulated problem. To further enhance the training effectiveness by leveraging the CF-MEC system, we design distinct actor-critic networks for the agents at different levels to facilitate centralized training and distributed execution. Simulation results validate the training stability of the proposed algorithm at each level, and demonstrate the superiority of the proposed algorithm over benchmark schemes in terms of the average energy consumption, providing a stable distributed framework for practical implementation in dynamic environments.

The entomophagous Bracon hebetor is a larval endoparasitoid with ontogenetic development strictly dependent on the thermal regime. Investigations carried out under controlled conditions have demonstrated that the optimal thermal range for the development of immature stages is 25-27°C, in which the life cycle is completed in 12-15 days. At temperatures of 15-17°C, development is significantly slowed down, extending the ontogenetic cycle to approximately 32 days. At extreme thermal values of 10°C and 32°C, no adult emergence was recorded. The hatching, larval development and pupation stages showed significant variations depending on the thermal regime, highlighting the thermal susceptibility of the entomophagous. The results obtained demonstrate that the range of 25-27°C constitutes the optimal thermal range for the efficient reproduction and multiplication of the B. hebetor species in a biotechnological context.

Students frequently struggle while in solving the problem in the percentage, beside the common use of percentages in educational and daily lives. The descriptive analysis of the randomly selected answer-sheets of taken from the four students of grade IX. The errors in solving a verbal problem on percentage calculation. In the analysis of the solution given by the students’ shows that many students demonstrated the transformation errors in misunderstandings regarding the percentage and its relationships in understanding of arithmetic aptitudes. Students perceived the concept of percentage increase in additive and then multiplicative processes. Highlighting the errors occur when calculating original values from the increased amount. In this research paper, researcher examines the understanding of students’ problems related to percentage through the lenses of Newman's Error Analysis (NEA) framework. This research paper contributes in enhancing mathematics education by exploring the conceptual challenges in resolving percentage problems. It also recommend the effective instructional approach has explicitly addressed for the ontological transition from additive to multiplicative reasoning. Further this research address the cognitive effect in the mind of students that contradicts existing mental models and focused in restructuring the precise conceptual frameworks.

The exponential growth of digital data and computational requirements has significantly increased the demand for advanced computing paradigms capable of processing large-scale workloads efficiently. Parallel and distributed computing have emerged as key technologies that enable high-performance processing by utilizing multiple processors and interconnected computing nodes. These paradigms play an essential role in modern applications such as artificial intelligence, big data analytics, cloud computing, and scientific simulations. This paper provides a comprehensive overview of recent developments in parallel and distributed computing, focusing on heterogeneous computing architectures, distributed machine learning frameworks, cloud-native infrastructures, and energy-efficient computing techniques. Additionally, major technical challenges including communication overhead, resource management, and system scalability are discussed. Finally, potential future research directions are explored, including edge computing integration, intelligent scheduling mechanisms, and hybrid computing environments. The study highlights how emerging technologies continue to reshape large-scale computing systems and improve computational efficiency across various domains.

We consider an iterative branching process in which an abstract object can subdivide into other objects. The multiplication process may be varied by the occurrence of random "fatal" events in which some of the subsequent objects or states may fail. The process is also constrained to terminate upon reaching a given number of events or alternatively upon reaching a fixed number of iteration steps. A system of diophantine integer-variable equations capable of describing the aforementioned process is proposed. These equations can be applied prospectively to many branching phenomena of physical, biological and demographic nature. The equations, which we call systems of equations S, Q, U can be reformulated into three main classes based on the behavior of the sum of variables with respect to a fixed principal numerical parameter (TC= 'Total Cases'). These systems always admit solutions and these are sought for the three classes. The mathematical properties of the three systems are presented both analytically and graphically, and the software script for calculating numerical solutions is attached. In the case of high TC values, where direct calculation is not possible, special solutions are also sought for the steady state case and the "most probable" case, the latter using statistical mechanics methods. Solutions examples are given for a wide range of TC parameters. We also refer to real-world examples of applications ranging from prey/predator population dynamics to population mortality modeling and 2d lattice space tiling and also tree leaves branching alternatives. The main purpose of the study here proposed is to implement a mathematical frame that can provide tools to be used in the study of real-world applications.

Physical Review A central challenge in climate science and applied mathematics is developing data-driven models of multiscale systems that capture both stationary statistics and responses to external perturbations. Current neural climate emulators aim to resolve the atmosphere-ocean system in all its complexity but often struggle to reproduce forced responses, limiting their use in causal studies such as Green's function experiments. To explore the origin of these limitations, we first examine a simplified dynamical system that retains key features of climate variability. We interpret the results through linear response theory, providing a rigorous framework to evaluate neural models beyond stationary statistics and probe causal mechanisms. We argue that the ability of emulators of multiscale systems to reproduce perturbed statistics depends critically on (1) the choice of an appropriate coarse-grained representation and (2) careful parameterizations of unresolved processes. These insights highlight reduced-order models, tailored to specific goals, processes, and scales, as valuable alternatives to general-purpose emulators. We next consider a real-world application, developing a neural model to investigate the joint variability of the surface temperature field and radiative fluxes. The model infers a multiplicative noise process directly from data, largely reproduces the system's probability distribution, and enables causal studies through forced responses. We discuss its limitations and outline directions for future work. These results expose key challenges in data-driven modeling of multiscale physical systems and underscore the value of coarse-grained, stochastic approaches, with response theory as a principled framework to guide model design and enhance causal understanding.

The fractal dimension is the basic notion for describing structures that have a scaling symmetry. In finance, multi-fractality is one of the well known facts which characterized non-trivial properties of financial time series. The stock price (or index) fluctuations can be described in terms of long-range temporal correlations by a spectrum of the Holder exponents and a set of fractal dimensions. To forecast the market risk, assessing the stock price indices is the foundation. Multi-fractal has lots of advantages when explaining the volatility of the stock prices. The asset price returns are multi-period market depending on market scenarios which are the measure points. In this work, we use some tools of multi-fractal analysis to derive the worth growth rate of an investor’s portfolio for particular and general cases. For the particular case, we considered the situation when the mean interest rate of some stocks does not depend on other stocks in the market. That is, an investor has invested his money in a stock with a linear mean return. Under the general case, we considered a market comprising some units of assets in long position and a unit of the option in short position. Using Ito’s formula on the present value of the market, we derived the growth rate of investor’s portfolio. Our model equations, which are based on multiplicative processes, capture all the features of the returns. They are tested using data from Zenith Bank of Nigeria stock prices. From our graphs, the worth of investment grows as stock price increases and also decreases with stock price.