Search Results (612)

The Q-Learning Relational Game (QLRG) framework provides a theoretically rigorous method for identifying minimal winning coalitions in online social networks (OSNs) under the restrictive assumption of global agent symmetry or uniform matroid structure. Real-world OSNs, however, exhibit significant asymmetry. This paper introduces the Cluster-Based Q-Learning Relational Game (C-QLRG), a practical extension that relaxes the global symmetry requirement by leveraging community structure. We partition the agent set into communities with bounded internal variation and represent the state solely by community membership counts of the seed set. Because the closure operator already captures all eventual influence spread, the problem reduces to a sequential seed selection task where the agent decides, at each step, from which community to add the next seed. We prove that the optimal Q-function of a suitably regularized reach-efficiency objective is Lipschitz continuous and derive a performance bound for the learned policy. The full algorithm is presented, and its complexity is analyzed. Empirical evaluations on a synthetic asymmetric network and Zachary’s Karate Club demonstrate that C-QLRG is highly sensitive to reward parameters, where default settings lead to premature stopping, but parameter tuning combined with a corrected minimality verification recovers high-efficiency coalitions by removing non-contributing agents. With tuned parameters, C-QLRG produces a near-winning coalition of size 11 and 99% reach on the synthetic network, surpassing the greedy baseline’s efficiency (size 12) despite a one-node coverage gap, while identifying the optimal winning coalition of size 1 on the Karate Club dataset, matching all baselines. The framework thus offers a principled trade-off between model fidelity and scalability, with the reward design choice being critical for practical deployment. Full article

This study investigates the influence of PLA flowability and talc content on the performance of compostable thermoplastic starch/poly(butylene succinate) (TPS/PBS)-based systems for rigid applications. Different PLA grades with varying melt flow index (PLA23, PLA8 and PLA70) and talc contents (0, 5 and 10 wt%) were incorporated. Twelve formulations were compounded by twin-screw extrusion and processed by injection moulding. FTIR confirmed the coexistence of TPS, PBS and PLA phases without evidence of chemical interactions. Morphological analysis showed that PLA flowability plays a key role in phase distribution, with higher-flow PLA promoting improved dispersion and interfacial adhesion, while talc addition (5 and 10 wt%) increased structural heterogeneity; at higher loadings, particularly, DSC analysis revealed that talc acted as a nucleating agent for the PBS phase, increasing crystallisation temperatures from approximately 73 °C to 81 °C depending on formulation. Mechanical results showed that Young’s modulus increased from approximately 1.4 GPa to 2.7 GPa with decreasing PLA flowability and increasing talc content. Formulations containing low-flow PLA reached tensile strengths close to 32 MPa, although elongation at break decreased to values near 2%. In contrast, high-flow PLA formulations exhibited a more balanced mechanical response, with elongation values up to approximately 8%, associated with improved phase dispersion. Hybrid PLA systems showed intermediate behaviour, reaching elongations up to 22% while maintaining modulus values around 1.8 GPa. Talc provided additional reinforcement but reduced deformation capacity. HDT values remained relatively constant, indicating limited improvement in thermomechanical resistance despite increased stiffness. These results demonstrate that the combined control of PLA molecular characteristics and talc content enables tuning of the mechanical and thermomechanical performance of TPS/PBS/PLA/talc systems for rigid packaging applications. Full article

Globally, Trichogramma Westwood (Hymenoptera: Trichogrammatidae) is known as the most effective biological control agent due to its ability to parasitize insect pest eggs. However, identifying an appropriate host is vital for Trichogramma to prosper. Therefore, this study delves into the complex role of semiochemicals in shaping the host-seeking behavior of Trichogramma parasitoids, with a particular focus on their responses to both plant-derived and host-derived cues. The mechanism of semiochemical reception in Trichogramma wasps relies on a highly specialized, sensitive olfactory and gustatory system to locate host eggs and mates. Semiochemicals, which mediate ecological interactions, have been identified as pivotal in influencing the parasitic efficiency of Trichogramma species. Trichogramma’s host-seeking behavior is influenced not solely by ovipositional cues but also by the intrinsic physical attributes of Lepidopteran hosts, such as the scales on the wings and abdomen, which emit semiochemicals capable of eliciting positive chemotactic responses, thereby guiding parasitoids toward optimal sites for oviposition. Furthermore, the interplay between insect-derived and plant-derived chemical cues exhibits a synergistic effect, collectively enhancing the chemotactic attraction of Trichogramma, thereby fine-tuning its host-seeking behavior with greater precision and specificity. This study further underscores Trichogramma’s innate behavioral ability to discriminate between host eggs of varying developmental stages, facilitating the precise identification and selection of the most suitable host for parasitization. Age and experience both make Trichogramma more selective of hosts, but younger parasitoids may take a broader approach to host selection due to their greater life expectancy. Furthermore, the removal of these cues affects their host localization and learning abilities. Associative learning enables Trichogramma to exhibit flexible behaviors, providing them with a selective advantage; allows them to explore various hosts; and reduces environmental uncertainty. Plant structure, host density, and host age are the key factors that significantly influence the foraging and parasitism of Trichogramma. The searching speed of this parasitoid is significantly influenced by temperature. Heat stress increases VOC emissions in plants such as potato via stomatal opening, reducing herbivore attraction and enhancing parasitoid recruitment. Furthermore, air pollution, including CO₂, O₃, and NOx, impairs parasitoid efficiency by disrupting volatile-mediated host location and reducing biological control performance. Trichogramma wasps are generally effective biological control agents, but their success depends on the species used, target pest, crop, release density, and field conditions. Overall, species such as T. ostriniae, T. japonicum, and T. leucaniae show the strongest performance in several crops by increasing parasitism, reducing pest damage, and improving yield. This study highlights the successful integration of semiochemical cues in pest management programs and the effective utilization of Trichogramma in conjunction with entomopathogenic bacteria to control Lepidopteran pests. This approach contributes to the development of more effective pest management strategies, thereby promoting agricultural sustainability. Full article

Reasoning-intensive multimodal retrieval suffers from a counter-intuitive bottleneck: on MM-BRIGHT multimodal-to-text (Query+Image → Documents), the strongest dense multimodal encoder reaches only 27.6 nDCG@10 and the rest of the dense vision–language retrievers cluster between 10.0 and 23.0. The visual signal, encoded as a dense vector, adds noise rather than evidence; even augmenting strong text retrievers with raw image captions degrades performance by up to 12.0 points. We propose VISA, a Visual Symbolic Agent that re-casts multimodal-to-text as text retrieval over three parallel streams. A Vision LLM is dispatched in three roles via separate prompts: a zero-shot router that classifies the query image into up to three parser types from a fixed taxonomy of nine (chart, circuit, equation, screenshot, code, figure, diagram, map, photograph); typed parsers that extract structured text per type; and a holistic captioner. The agent constructs three text streams (raw query, query ⊕ symbolic, query ⊕ caption), scores each with a single frozen 4B-parameter retrieval LLM, and fuses the per-document scores via Reciprocal Rank Fusion or a confidence-weighted linear combination. The whole agent contains no trainable parameters. The key novelty is a change of substrate: rather than projecting the query image into a dense multimodal vector that competes with text, VISA is, to our knowledge, the first retrieval system to convert the image into typed symbolic text and keep retrieval entirely text-side, so that a frozen text retriever can match the literal tokens (axis values, variable names, function signatures) that answering documents actually contain. Across all 29 MM-BRIGHT multimodal-to-text domains, VISA achieves 32.4 nDCG@10, an absolute improvement of +4.8 over the strongest dense multimodal encoder and substantially larger margins over the remaining six dense vision–language baselines. Per-domain analysis shows VISA maintains its margin across STEM and software domains where image content is structure-heavy. In practical terms, VISA is training-free and model-agnostic: it requires no fine-tuning, reuses any off-the-shelf vision LLM and text retriever, caches all per-image parsing so re-runs cost only three query encodes, and can therefore be dropped into an existing text-retrieval stack to add reasoning-intensive multimodal capability without building or training a multimodal encoder. Full article

The proliferation of Unmanned Aerial Vehicles (UAVs) in various applications has created a pressing need for robust and efficient communication systems. Fifth-generation (5G) networks can support UAV connectivity through high bandwidth and low-latency communication; however, rapid three-dimensional UAV mobility creates handover-management challenges that can increase signalling overhead, service interruption, and Quality of Service (QoS) degradation. This paper presents an integrated framework that combines LSTM-based multi-UAV trajectory prediction with proactive handover optimization using an Advantage Actor–Critic (A2C) Deep Reinforcement Learning (DRL) agent. The LSTM predictor is evaluated on a real-world UAV trajectory dataset and reports a root mean square error (RMSE) of 4.37 m over a 5 s prediction horizon after conversion to a local East–North–Up coordinate frame. A lightweight simulation-level coordination mechanism is included to reduce simultaneous target-cell contention among multiple UAVs; it is not claimed as a new standardized 3GPP signalling procedure. Handover performance is evaluated by replaying 180 held-out flight trajectories in a controlled 5G simulation across ten independent random seeds. Under these stated assumptions, the proposed framework achieves a handover success rate of $94.2\pm 0.8$%, an average SINR of $15.8\pm 0.2$ dB, a handover delay of $45.2\pm 1.1$ ms, and a handover frequency of $0.85\pm 0.05$ HOs/min, outperforming the tuned 3GPP A3, reactive SINR, and CASH baselines in the reported simulation results (Wilcoxon signed-rank test, $p<0.01$, Bonferroni-corrected). The experimental setup is described in detail to support methodological transparency and facilitate future replication, but the handover results should be interpreted as simulation-based evidence rather than live-network validation. Full article

Surface modification of zinc oxide nanoparticles (ZnO NPs) with organosilane capping agents represents an effective strategy to control their physicochemical and biological properties. In this work, we report for the first time the use of halogenosilanes, namely (3-chloropropyl)trimethoxysilane (CPTMS), (3-bromopropyl)trimethoxysilane (BPTMS) and (3-iodopropyl)trimethoxysilane (IPTMS), for the surface functionalization of ZnO NPs obtained by chemical precipitation. Structural and morphological characterization (PXRD, TEM, SEM-EDX and FTIR) confirmed successful surface modification and revealed a significant particle size reduction from ~31 nm for unmodified ZnO to ~8 nm for BPTMS-modified ZnO (ZnO_b). The biological evaluation showed that halogenosilane-modified ZnO NPs exhibit enhanced cytotoxic activity against prostate cancer cell lines (PC3 and 22Rv1), with ZnO_b displaying the highest activity, likely associated with improved cellular uptake and increased reactive oxygen species (ROS) generation. In contrast, antimicrobial assays revealed only moderate bactericidal effects against Escherichia coli and Staphylococcus aureus at relatively high concentrations (≥1250 µg mL⁻¹), while no significant activity was observed against Pseudomonas aeruginosa, Burkholderia contaminans or Candida spp, within the tested range. These findings suggest that halogenosilane functionalization modulates the biological profile of ZnO nanoparticles by enhancing anticancer effects while also influencing microbiocidal activity, highlighting the role of surface chemistry in tuning biological selectivity. The present study supports the concept that rational surface engineering of ZnO-based nanoplatforms can be exploited to favor tumor-targeted activity over broad-spectrum antimicrobial effects, providing new perspectives for the design of application-oriented nanomaterials. Full article

To improve the path-following performance of an underactuated autonomous underwater vehicle (AUV) under varying path geometries and desired velocities, this study proposes a direct X-rudder control method based on Task-Informed Inductive-Bias Conservative Soft Actor–Critic (TIB-CSAC). The proposed method directly learns the X-rudder control policy from the path-following information of the current and subsequent path segments in a data-driven way, thereby avoiding the complex design and manual tuning of guidance laws and attitude controllers for rudder command generation. To support such two-segment policy learning, a task-informed inductive-bias encoder is proposed to construct structured and conditioned state representations, thereby improving sample efficiency and overall training quality. In addition, given the long-tail characteristics of task difficulty in agent training, a multi-head conservative value evaluation mechanism is incorporated to mitigate return drawdowns induced by challenging tasks in the tail stage of training and to enhance tail-stage convergence stability. The path-following performance is validated in three representative scenarios with different path pitch, path heading variations, and desired surge velocity conditions. The results show that, compared with the baseline soft actor–critic (SAC) method, TIB-CSAC improves multiple vertical and horizontal error metrics, including maximum absolute error, mean absolute error, tail error, and error threshold exceedance ratio. These results indicate that TIB-CSAC not only improves overall adherence to the reference path, but also more effectively suppresses extreme errors and tail errors, thereby demonstrating stronger path-following robustness and reliability. Full article

This paper proposes a reinforcement learning-assisted active disturbance rejection control (ADRC) framework for a nonlinear quadrotor unmanned aerial vehicle (UAV). Conventional ADRC controllers are designed for the quadrotor altitude and attitude channels. To evaluate robustness under disturbance-intensive conditions, a composite external disturbance is injected into the roll-channel dynamics. A Deep Deterministic Policy Gradient (DDPG)-based adaptive tuning mechanism is integrated into the roll-channel ADRC for the nonlinear state error feedback (NLSEF) gain adaptation, while fixed-parameter ADRC is retained for the remaining three channels. Without requiring system linearization and prior knowledge of disturbance models, the reinforcement learning agent learns an optimal gain adaptation policy directly through interaction with the nonlinear roll subsystem. Quantitative simulations demonstrate superior roll-axis disturbance rejection, leading to 90% faster settling time, the root mean square (RMS) control effort being reduced by 5.1%, and a 7.6% peak input suppression compared to conventional ADRC. The learning-based adaptation maintains comparable tracking accuracy across all channels while significantly improving transient recovery and control smoothness in the most disturbance-sensitive axis, validating selective reinforcement learning integration for robust nonlinear quadrotor flight control. Full article

Automating system-level nuclear thermal-hydraulic (T-H) model construction remains challenging because platform-specific API syntax, graph connectivity, parameter dependency ordering, and solver admissibility must be satisfied simultaneously. This study develops a closed-loop modeling framework on the SAFRI platform by combining supervised fine-tuning (SFT), a Plan-and-Act agent with retrieval-grounded parameter completion, and reinforcement learning based on group relative policy optimization (GRPO). The SFT stage uses a 6003-record domain corpus derived from expert-authored or expert-verified SAFRI modeling exemplars, while system-level generalization is evaluated on a held-out 50-case in-house evaluation set separated at the case-template level. At the component level, LoRA-adapted Qwen3-8B achieves 100% code accuracy, compared with 50% for zero-shot and 74% for one-shot prompting. At the system level, the SFT agent attains a 100% syntax success rate (SSR), 90% topology success rate (TSR), and 72.4% physical convergence rate (PCR), showing that local API correctness is insufficient for solver-valid model assembly. After GRPO training with schema, topology, physics, and sequence rewards, the full SAFRI-SFT-RL agent reaches a 100% SSR, 100% TSR, and 88.8% PCR on the in-house evaluation set, while an error self-healing loop resolves execution-time failures in an average of 2.3 corrective iterations. These results show that solver-grounded reinforcement learning is effective for closing the gap between syntactically correct script generation and physically convergent nuclear T-H model construction. Full article

Arthropod natural enemies—encompassing predators and parasitoids—form the backbone of sustainable agriculture, delivering irreplaceable ecosystem services via biological pest suppression. Driven by global demand for eco-friendly alternatives to synthetic pesticides, research in this domain has grown sharply over the past decade. Here, we report a systematic bibliometric analysis of 6515 Web of Science Core Collection papers focused on arthropod natural enemies in biological control (2016–2025), with the goal of charting the field’s intellectual structure. Performance metrics confirmed an initial rapid increase from 2016 to 2019 followed by a plateau and a slight rise in 2025, with the US, China, and Brazil dominating output. Keyword co-occurrence networks pinpointed core themes, including conservation biological control, predatory mites, and integrated pest management (IPM). Temporal trends further revealed a pivot toward applied work on invasive pest systems. Co-citation analysis uncovered six foundational research clusters, while bibliographic coupling of 2021–2025 papers uncovered five active emerging subfields: landscape ecology and habitat manipulation, tri-trophic interaction mechanisms, high-impact invasive pest biocontrol, non-target risk assessment for introduced agents, and fall armyworm integrated management. We synthesize cross-cutting implications and outline future priorities—including AI-enabled rearing systems, functional biodiversity boosting, climate adaptation, and multifunctional landscape tuning. By consolidating historical progress and forward-looking directions, this framework empowers researchers, extension practitioners, and policymakers to scale sustainable pest management worldwide. Full article

Modelling constitutes a disciplined transformation process through which heterogeneous, unstructured evidence is translated into structured representations that support reasoning and decision-making. The integration of generative artificial intelligence into such processes introduces new possibilities for automation, yet risks undermining methodological rigour, traceability, and human accountability. This paper proposes a methodology-grounded multi-agent architecture for constructing structured business ecosystem maps from unstructured document collections. The architecture decomposes the modelling lifecycle into specialised agent functions covering boundary specification, source discovery, document analysis, semantic extraction, and controlled model editing, addressing four of the five methodology stages while leaving automated completeness verification outside the current scope. A central orchestrator coordinates agents while enforcing ontological constraints derived from a formal modelling methodology. All proposed modifications are staged for human review before execution, and each map element maintains explicit provenance links to source material. To evaluate the reliability and correctness of generative modelling pipelines, a hybrid evaluation framework integrates operational metrics, semantic assessment using an LLM-based judge, and human agreement validation. Empirical evaluation across 34 generative models and 4382 experimental runs characterises capabilities across modelling tasks. In a controlled single-document extraction task, text-based extraction achieves a mean semantic match score of 0.947, whereas interaction extraction scores 0.431 and visual diagram interpretation scores 0.470, identifying relational reasoning and multimodal interpretation as principal bottlenecks. Model performance varies across agent roles, with task-aligned model selection associated with larger performance changes than hyperparameter tuning; the architecture’s causal contribution is not isolated, and comparison against monolithic or ablated baselines remains future work. Full article

Machine learning (ML) methods for network anomaly detection are emerging as effective proactive strategies in threat hunting, substantially reducing the time required for threat detection and response. However, the challenges in training and maintaining ML models, coupled with frequent false positives, diminish their acceptance and trustworthiness. In response, Explainable AI (XAI) techniques have been introduced to enable cybersecurity operations teams to assess alerts generated by AI systems more confidently. Despite these advancements, XAI tools have encountered limited acceptance from incident responders and have struggled to meet the decision-making needs of both analysts and model maintainers. Large Language Models (LLMs) offer a unique approach to tackling these challenges. Through tuning, LLMs have the ability to discern patterns across vast amounts of information and meet varying functional requirements. In this research, we introduce the development of HuntGPT, a specialized intrusion detection dashboard created to implement a Random Forest classifier trained utilizing the KDD99 dataset. The tool incorporates XAI frameworks like SHAP and Lime, enhancing user-friendliness and intuitiveness of the model. When combined with a GPT-3.5 Turbo conversational agent, HuntGPT aims to deliver detected threats in an easily explainable format, emphasizing user understanding and offering a smooth interactive experience. We investigate the system’s comprehensive architecture and its diverse components, assess the prototype’s technical accuracy using the Certified Information Security Manager (CISM) Practice Exams, and analyze the quality of response readability across six unique metrics. Our results indicate that conversational agents, underpinned by LLM technology and integrated with XAI, can enable a robust mechanism for generating explainable and actionable AI solutions, especially within the realm of intrusion detection systems. Full article

The increasing penetration of nonlinear loads aggravates grid harmonic distortion and imposes higher requirements on the control performance of the shunt active power filter (SAPF). To address the problems of fixed parameters in conventional PI controllers and the dependence of existing improved methods on accurate models with relatively high computational complexity, this paper proposes a TD3-based model-free adaptive control method for online coordinated tuning of dual-loop PI parameters in the SAPF system. The proposed method dynamically adjusts PI parameters through agent–environment interaction without requiring an accurate system model or a complex control structure. Simulation results show that the proposed strategy outperforms fixed-parameter PI control, PI-SMC, and DDPG methods in both steady-state and dynamic performance, and maintains good control performance under unseen composite disturbances such as capacitive inrush and load variation. RT-LAB-based hardware-in-the-loop (HIL) validation further demonstrates that the proposed method can achieve effective harmonic compensation and DC-link voltage regulation on a real-time simulation platform. Meanwhile, during online deployment, only Actor-network forward inference is required to update the PI parameters, indicating a low additional computational burden and engineering implementation potential for SAPF real-time control systems. Full article

The success of DNA amplification in Polymerase Chain Reaction (PCR) devices inherently depends on the rapid and absolute zero-overshoot temperature control of thermoelectric cooler (TEC) systems. In the literature, metaheuristic algorithms employed for proportional–integral–derivative (PID) tuning typically operate within unconstrained search spaces, relying exclusively on time-domain error metrics like ITAE. This conventional approach causes ‘gradient blindness’ and neglects frequency-domain robustness, resulting in excessive temperature overshoots that violate biological safety limits and lead to enzyme denaturation. To solve this problem, we propose a hybrid frequency-time domain optimization framework. Utilizing a first order plus dead-time (FOPDT) model for TEC dynamics, the PID search space is analytically restricted via Ziegler–Nichol’s stability boundaries. Furthermore, Phase Margin (PM ≥ 45°) and absolute zero-overshoot conditions are integrated into the objective function as a strict penalty mechanism. Evaluations conducted with five distinct metaheuristic algorithms (PSO, GWO, WOA, ABC, and ACO) prove that while traditional unconstrained methods yield overshoots up to 19.04%, the proposed architecture successfully confines all optimization agents to a globally stable region, enabling specific algorithms like ABC, PSO, and WOA to achieve exactly 0.00% overshoot. Validated across a realistic multi-step PCR cycle (95–55–75 °C), the developed robust controller settles into the denaturation phase with a 0.00 °C peak error, guaranteeing biological sample safety and delivering a reliable control framework for rapid-cycle PCR platforms. Full article

Deploying Large Language Models (LLMs) in collaborative multi-agent settings represents a promising frontier for complex AI problem-solving, yet the field lacks systematic mechanisms to manage the inherent coordination overhead and resource contention that arise at scale. Existing LLM-based Multi-Agent System (MAS) frameworks predominantly adopt sequential or loosely coupled execution models, which fail to exploit the parallelism potential of modern computing environments and limit overall system throughput. To bridge this gap, this paper presents DynTaskMAS, a framework that redefines task orchestration in LLM-based MASs through a dynamic task graph abstraction. Rather than treating tasks as static pipelines, DynTaskMAS continuously models task interdependencies at runtime, enabling opportunistic parallel execution while preserving logical correctness. The architecture integrates four synergistic components: a runtime task decomposition module that captures evolving dependencies among subtasks; a scheduling engine that dispatches ready tasks to available agents without centralized bottlenecks; a context propagation layer that maintains shared semantic state across concurrently executing agents; and a self-tuning workflow controller that adapts execution priorities based on observed system load. Together, these components address a core tension in LLM-based MAS design, balancing agent autonomy with coordinated efficiency. Evaluations across tasks of varying complexity confirm that DynTaskMAS delivers substantial gains in execution efficiency (21.3–33.0% reduction), resource utilization (from 65% to 88%), and agent scalability (3.47× throughput with 16 concurrent agents) compared to sequential baselines. This work offers a generalizable architectural blueprint for next-generation LLM-based Multi-Agent Systems operating under real-world dynamic and resource-constrained conditions. Full article