Abstract
As organizations navigate through complex and collaborative digital environments, Generative AI (GenAI) emerges as a transformative force for Knowledge Management (KM) processes. This paper highlights how GenAI technologies impact collaborative KM processes across individual, intraorganizational, and inter-organizational levels within the evolving paradigm of Industry 5.0 (i5.0). Through a literature review, the study explores how GenAI augments human cognition, enhances knowledge creation and sharing, and fosters organizational adaptability and innovation. The findings highlight GenAI's potential as cognitive partner, streamlining information flows, and improving decision-making across collaborative networks. However, challenges such as over-reliance, ethical risks, and the decline of critical human skills are also discussed. Furthermore, the paper identifies the evolution and gaps in current literature on Collaborative Networks (CNs) regarding the integration of AI technologies. It contributes to the ongoing discussion towards a socio-technical transformation while also providing an overview for rethinking collaboration and social strategies in the GenAI era.

Abstract
Federated learning offers a powerful framework for machine learning in medical imaging by enabling collaborative model training across decentralised institutions while preserving patient privacy. However, practical deployment faces challenges arising from data scarcity and heterogeneity, which can substantially impair model performance. This review systematically examines the diverse forms of heterogeneity in federated medical imaging, categorising them into feature skew, label skew, quantity skew, and quality skew. It also distinguishes data scarcity and label scarcity as separate but related challenges. By analysing more than 130 recent publications, the review surveys state-of-the-art algorithms designed to mitigate these problems, classifying them by methodology-including model architecture optimisation, aggregation strategies, and personalisation approaches. This review offers researchers aiming to develop federated learning systems for medical image analysis a comprehensive overview of the major challenges posed by different skews and scarcities, along with the range of solutions proposed to address them. As federated learning moves from research settings to clinical practice, understanding and addressing data heterogeneity and scarcity will be essential to ensuring that collaborative learning systems perform reliably across the diversity of real-world medical imaging environments.

Abstract
To achieve carbon neutrality by 2050, commercial buildings have installed photovoltaic systems to reduce carbon emissions and operational costs. Nevertheless, PV generation does not always match the building’s energy demand profile, therefore storage systems are needed to store excess energy and supply it when necessary. This paper presents a Mixed Integer Linear Programming optimisation algorithm designed to schedule the operation of the electric storage system, aiming to minimise the building’s energy-related costs. An annual hourly simulation of the optimised system was performed to assess the cost reduction. To prevent excessive operation of the electric storage system, an approach to penalise low energy charging was studied, with results showing a significant increase in the system’s lifespan. Key words. MILP, optimisation, renewable energy, energy storage system, commercial building

Abstract
The integration of AI into organizational settings leads to a growing need for hybrid human-AI collaborative approaches, necessary due to the increasing autonomy, impact and responsibility AI-based solutions have. Moreover, to ensure a sustainable integration into existing processes, such approaches must be context-aware, transparent, and human-centric. In line with the Industry 5.0 paradigm, this paper presents a novel Multi-Agent System architecture that enables meaningful collaboration between human and artificial agents through a socio-technical design approach. The proposed architecture is grounded in a structured, real-time context stream derived from organizational data sources, which semantically describe human actors, processes, and industrial resources. Central to this system is a set of four core LLM-based agents, each responsible for orchestrating hybrid human-AI tasks along distinct dimensions of timing, role selection, resource allocation, and execution sequencing. To assess the feasibility and effectiveness of the architecture, we report on an early-stage validation conducted within a representative industrial use case in the automotive sector, focused on information retrieval. In this use case, the architecture was tasked with answering a set of representative, domain-specific questions by dynamically interacting with distributed industrial databases. Results demonstrate the architecture's ability to coordinate relevant human and artificial agents, retrieve semantically-relevant data, and present explainable outputs, showcasing its potential for supporting decision-making processes in hybrid collaborative networks.

Abstract
The automotive sector is undergoing continuous technological evolution driven by the demand for sustainable and safe vehicles. Among the main factors influencing safety, driver behaviour has been identified as a critical contributor to road crashes. This systematic review explores recent innovations in detecting risky driver behaviours, addressing six research questions: the most relevant datasets used for algorithm development and evaluation; system architectures and methodologies for anomaly detection; the most studied driver behaviours and related environmental, human, and mechanical factors; advances in machine learning, deep learning, and statistical methods; performance metrics and validation approaches; and the role of embedded technologies and sensors in practical applications. The review included 93 peer-reviewed articles published between 2020 and 2024, sourced from ACM, IEEE, ScienceDirect, and Scopus. Exclusion criteria were duplicates, non-open access, retracted works, and studies unrelated to outlier detection or driver behaviour. The Parsifal tool was used to support systematic data processing. Results highlight the most frequently used datasets, proposed models, and their performance in detecting driver behaviours, as well as the influence of contextual factors such as traffic rules, road conditions, and sensor limitations. Despite advances, real-world integration remains challenging, requiring further research and development. This review aims to guide researchers in understanding the current state of anomaly detection in driving contexts and to emphasize the need for broader collaboration to create effective, deployable solutions that enhance road safety worldwide.

Abstract
Modern data-intensive systems rely on in-memory caching to achieve high throughput and low latency. CacheLib, Meta's general-purpose caching engine, provides high performance and flexibility for building specialized caches for a variety of applications. However, despite its wide adoption in large-scale infrastructures, CacheLib's data management mechanisms exhibit inefficiencies in shared environments. Particularly, its static and uncoordinated memory allocation leads to fragmented resource usage, unfair memory distribution, and degraded performance across tenants and instances. We present Holpaca, a general-purpose caching middleware that enables holistic and adaptable orchestration of shared caching environments. Holpaca introduces a shim data layer co-located with each cache instance and a centralized orchestrator with system-wide visibility, enabling global memory management and per-tenant QoS policies. Using production traces from Twitter, results show that, by continuously readjusting memory allocations based on workload dynamics, Holpaca achieves up to 3 higher throughput in multi-tenant and 2.2× improvement in multi-instance settings over CacheLib's rigid built-in mechanisms.