The rapid proliferation of agentic AI systems has been dominated by Large Language Models (LLMs), but their substantial operational costs and high latency present significant barriers to widespread adoption. In response, the research community has increasingly turned to Small Language Models (SLMs), which offer a compelling combination of efficiency, task-specificity, and cost-effectiveness. This paper introduces the Adaptive Agentic Meta-Controller (AAMC), a deep reinforcement learning (RL) framework designed for intelligent SLM/LLM orchestration. The AAMC transforms the model selection problem into a principled, multi-objective optimization task, learning a dynamic policy that routes queries to the most appropriate model—preferring SLMs for routine tasks and escalating to LLMs only when necessary. Our framework features a Task Complexity Estimator (TCE) and an RL-based Router (RLR) that collaboratively balance the trade-offs between performance, cost, and latency. We conduct extensive experiments in a high-fidelity simulation environment, demonstrating that the AAMC achieves a task success rate comparable to an LLM-only approach while reducing operational costs by over 70% and significantly improving inference latency. We further introduce a comprehensive set of experiments on robustness, scalability, and fairness, including new ablation studies on the impact of the TCE and the sensitivity to user preferences, alongside a detailed complexity analysis and a discussion of real-world deployment. We further introduce a comprehensive set of experiments on robustness, scalability, and fairness, alongside a detailed complexity analysis and a real-world enterprise deployment case study. We also release code and experiments to support reproducibility.

The combination of Federated Learning (FL), Multimodal Large Language Models ( MLLMs ), and edge-cloud computing enables distributed and real-time data processing while preserving privacy across edge devices and cloud infrastructure. However, the deployment of MLLMs in FL environments with resource-constrained edge devices presents significant challenges, including resource management, communication overhead, and non-IID data. To address these challenges, we propose a novel hybrid framework wherein MLLMs are deployed on edge devices equipped with sufficient resources and battery life, while the majority of training occurs in the cloud. To identify suitable edge devices for deployment, we employ Particle Swarm Optimization ( PSO ), and Ant Colony Optimization ( ACO ) is utilized to optimize the transmission of model updates between edge and cloud nodes. This proposed swarm intelligence-based framework aims to enhance the efficiency of MLLM training by conducting extensive training in the cloud and fine-tuning at the edge, thereby reducing energy consumption and communication costs. The framework is particularly applicable in Unmanned Vehicle System (UVS)-based edge-cloud computing environments, where efficient resource management and real-time decision-making are critical for autonomous operations. The autonomous nature of UVS requires efficient resource management and a reliable communication system. The proposed swarm intelligence-based framework aims to satisfy these requirements through optimized edge device selection, while minimizing communication overhead. Our experimental results show that the proposed method significantly improves system performance, achieving an accuracy of 92%, reducing communication cost by 30%, and enhancing client participation compared to traditional FL methods. These results make the proposed approach highly suitable for large-scale edge-cloud computing systems.

Cardiopulmonary resuscitation (CPR) is a critical, life-saving intervention aimed at restoring blood circulation and breathing in individuals experiencing cardiac arrest or respiratory failure. Accurate and real-time analysis of biomedical signals during CPR is essential for monitoring and decision-making, from the pre-hospital stage to the intensive care unit (ICU). However, CPR signals are often corrupted by noise and artifacts, making precise interpretation challenging. Traditional denoising methods, such as filters, struggle to adapt to the varying and complex noise patterns present in CPR signals. Given the high-stakes nature of CPR, where rapid and accurate responses can determine survival, there is a pressing need for more robust and adaptive denoising techniques. In this context, an unsupervised machine learning (ML) methodology is particularly valuable, as it removes the dependence on labeled data, which can be scarce or impractical in emergency scenarios. This paper introduces a novel unsupervised ML approach for denoising CPR signals using a multi-modality framework, which leverages multiple signal sources to enhance the denoising process. The proposed approach not only improves noise reduction and signal fidelity but also preserves critical inter-signal correlations (0.9993) which is crucial for downstream tasks. Furthermore, it outperforms existing methods in an unsupervised context in terms of signal-to-noise ratio (SNR) and peak signal-to-noise ratio (PSNR), making it highly effective for real-time applications. The integration of multi-modality further enhances the system's adaptability to various biomedical signals beyond CPR, improving both automated CPR systems and clinical decision-making. • Unsupervised multi-modal ML approach for robust CPR signal denoising. • Independently processes signals while preserving inter-signal correlations. • Outperform existing methods without labels, enhance reliability in practice. • Improves adaptability and explainability for real-time healthcare use cases. • Adaptable for various biomedical signals beyond CPR in clinical settings.

The proliferation of the Internet of Things (IoT) has led to an explosion of data generated by interconnected devices, presenting both opportunities and challenges for intelligent decision-making in complex environments. Traditional Reinforcement Learning (RL) approaches often struggle to fully harness this data due to their limited ability to process and interpret the intricate patterns and dependencies inherent in IoT applications. This paper introduces a novel framework that integrates transformer architectures with Proximal Policy Optimization (PPO) to address these challenges. By leveraging the self-attention mechanism of transformers, our approach enhances RL agents’ capacity for understanding and acting within dynamic IoT environments, leading to improved decision-making processes. We demonstrate the effectiveness of our method across various IoT scenarios, from smart home automation to industrial control systems, showing marked improvements in decisionmaking efficiency and adaptability. Our contributions include a detailed exploration of the transformer’s role in processing heterogeneous IoT data, a comprehensive evaluation of the framework’s performance in diverse environments, and a benchmark against traditional RL methods. The results indicate significant advancements in enabling RL agents to navigate the complexities of IoT ecosystems, highlighting the potential of our approach to revolutionize intelligent automation and decision-making in the IoT landscape.

Autonomous driving has been gaining a lot of attention in the field of transportation technology in recent years. The use of autonomous vehicles has the potential to reduce the number of road accidents caused by human error, improve traffic flow, increase fuel efficiency and save time for travelers. In federated learning systems, selecting trustworthy autonomous vehicles (AVs) to participate in training is critical for ensuring system performance and reliability. In this work, we propose a trust-aware approach to AV selection that incorporates the performance of each AV using the Local Interpretable Model-Agnostic Explanations (LIME) method and One-Shot Federated Learning. We modify the XAI LIME Deep Q-learning-based AV selection model to include the trust metric, resulting in the Trust-Aware XAI LIME Deep Q-learning-based AV selection model. Our experiments show that the trust-aware approach outperforms the standard approach in terms of both accuracy and reliability, demonstrating the effectiveness of incorporating trust metrics in AV selection.

The “black-box” nature of artificial intelligence (AI) models has been the source of many concerns in their use for critical applications. Explainable Artificial Intelligence (XAI) is a rapidly growing research field that aims to create machine learning models that can provide clear and interpretable explanations for their decisions and actions. In the field of cybersecurity, XAI has the potential to revolutionize the way we approach network and system security by enabling us to better understand the behavior of cyber threats and to design more effective defenses. In this survey, we review the state of the art in XAI for cybersecurity and explore the various approaches that have been proposed to address this important problem. The review follows a systematic classification of cybersecurity threats and issues in networks and digital systems. We discuss the challenges and limitations of current XAI methods in the context of cybersecurity and outline promising directions for future research.

Recently, the concept of autonomous driving became prevalent in the domain of intelligent transportation due to the promises of increased safety, traffic efficiency, fuel economy and reduced travel time. Numerous studies have been conducted in this area to help newcomer vehicles plan their trajectory and velocity. However, most of these proposals only consider trajectory planning using conjunction with a limited data set (i.e., metropolis areas, highways, and residential areas) or assume fully connected and automated vehicle environment. Moreover, these approaches are not explainable and lack trust regarding the contributions of the participating vehicles. To tackle these problems, we design an Explainable Artificial Intelligence (XAI) Federated Deep Reinforcement Learning model to improve the effectiveness and trustworthiness of the trajectory decisions for newcomer Autonomous Vehicles (AVs). When a newcomer AV seeks help for trajectory planning, the edge server launches a federated learning process to train the trajectory and velocity prediction model in a distributed collaborative fashion among participating AVs. One essential challenge in this approach is AVs selection, i.e., how to select the appropriate AVs that should participate in the federated learning process. For this purpose, XAI is first used to compute the contribution of each feature contributed by each vehicle to the overall solution. This helps us compute the trust value for each AV in the model. Then, a trust-based deep reinforcement learning model is put forward to make the selection decisions. Experiments using a real-life dataset show that our solution achieves better performance than benchmark solutions (i.e., Deep Q-Network (DQN), and Random Selection (RS)).

We argue in this paper that the role of the cloud should be reshaped from being a passive virtual market to become an active platform for monetizing data. The objective is to enable the cloud to be an active platform that can help data providers reach a wider set of data consumers. This will allow these consumers to be exposed to a larger variety of data that benefits data analytic applications. To achieve this vision, we propose a novel game theoretical model, which consists of a mix of cooperative and competitive strategies. The players of the game are the data providers, cloud platform, and cloud users. The strategies of the players are modeled using the two-sided market theory that takes into consideration the network effects (externalities) among the players. Simulations conducted using Amazon and google clustered data show that the proposed model improves the total surplus of involved parties in terms of cloud resources provision and monetary profits compared to the current merchant model.

Summary Cloud computing is undeniably becoming the main computing and storage platform for today's major workloads. From Internet of things and Industry 4.0 workloads to big data analytics and decision‐making jobs, cloud systems daily receive a massive number of tasks that need to be simultaneously and efficiently mapped onto the cloud resources. Therefore, deriving an appropriate task scheduling mechanism that can both minimize tasks' execution delay and cloud resources utilization is of prime importance. Recently, the concept of cloud automation has emerged to reduce the manual intervention and improve the resource management in large‐scale cloud computing workloads. In this article, we capitalize on this concept and propose four deep and reinforcement learning‐based scheduling approaches to automate the process of scheduling large‐scale workloads onto cloud computing resources, while reducing both the resource consumption and task waiting time. These approaches are: reinforcement learning (RL), deep Q networks, recurrent neural network long short‐term memory (RNN‐LSTM), and deep reinforcement learning combined with LSTM (DRL‐LSTM). Experiments conducted using real‐world datasets from Google Cloud Platform revealed that DRL‐LSTM outperforms the other three approaches. The experiments also showed that DRL‐LSTM minimizes the CPU usage cost up to 67% compared with the shortest job first (SJF), and up to 35% compared with both the round robin (RR) and improved particle swarm optimization (PSO) approaches. Moreover, our DRL‐LSTM solution decreases the RAM memory usage cost up to 72% compared with the SJF, up to 65% compared with the RR, and up to 31.25% compared with the improved PSO.

We present a formal framework that allows individual and group of agents to reason about their trust toward other agents. In particular, we propose a branching time temporal logic BT which includes operators that express concepts such as everyone trust, distributed trust and propagated trust. We analyze the satisfiability and model checking problems of this logic using a reduction technique.