Since the popularisation of Internet in 1990s, cyberspace has kept evolving. We have created various computer-mediated virtual environments including social networks, video conferencing, 3D worlds (e.g., VR Chat), augmented reality applications Pokemon Go), and Non-Fungible Token Games Upland). Such environments, albeit non-perpetual unconnected, bought us degrees digital transformation. The term `metaverse' been coined to further facilitate transformation every aspect our physical lives. At...

The metaverse, enormous virtual-physical cyberspace, has brought unprecedented opportunities for artists to blend every corner of our physical surroundings with digital creativity. This article conducts a comprehensive survey on computational arts, in which seven critical topics are relevant the describing novel artworks blended realities. first cover building elements e.g., virtual scenes and characters, auditory, textual elements. Next, several remarkable types creations expanded horizons...

The era of pervasive computing has resulted in countless devices that continuously monitor users and their environment, generating an abundance user behavioural data. Such data may support improving the quality service, but also lead to adverse usages such as surveillance advertisement. In parallel, Artificial Intelligence (AI) systems are being applied sensitive fields healthcare, justice, or human resources, raising multiple concerns on trustworthiness systems. Trust AI is thus...

This paper presents the development and evaluation of a Large Language Model (LLM), also known as foundation models, based multi-agent system framework for complex event processing (CEP) with focus on video query use cases. The primary goal is to create proof-of-concept (POC) that integrates state-of-the-art LLM orchestration frameworks publish/subscribe (pub/sub) tools address integration LLMs current CEP systems. Utilizing Autogen in conjunction Kafka message brokers, demonstrates an...

Artificial intelligence shows promise for solving many practical societal problems in areas such as healthcare and transportation. However, the current mechanisms AI model diffusion Github code repositories, academic project webpages, commercial marketplaces have some limitations; example, a lack of monetization methods, traceability, auditabilty. In this work, we sketch guidelines new method based on decentralized online marketplace. We consider technical, economic, regulatory aspects...

In Federated learning (FL) systems, a centralized entity (server), instead of access to the training data, has model parameter updates computed by each participant independently and based solely on their samples. Unfortunately, FL is susceptible poisoning attacks, in which malicious or malfunctioning entities share polluted that can compromise model's accuracy. this study, we propose FedClean, an mechanism robust attacks. The accuracy models trained with assistance FedClean close one where...

Secure and usable user authentication on mobile headsets is a challenging problem. The miniature-sized touchpad such devices becomes hurdle to interactions that impact usability. However, the most common methods, i.e., standard QWERTY virtual keyboard or mid-air inputs enter passwords are highly vulnerable shoulder surfing attacks. In this paper, we present PassWalk, keyboard-less system leveraging multi-modal headsets. PassWalk demonstrates feasibility of driven by user's gaze lateral...

Many factors affect speech intelligibility in face-to-face conversations. These lead conversation participants to speak louder and more distinctively, exposing the content potential eavesdroppers. To address these issues, we introduce Theophany, a privacy-preserving framework for augmenting speech. Theophany establishes ad-hoc social networks between exchange contextual information, improving real-time. At core of develop first privacy perception model that assesses risk based on its topic,...

In the modern era of mobile apps (the surveillance capitalism - as termed by Shoshana Zuboff) huge quantities data about consumers and their activities offer a wave opportunities for economic societal value creation. ln-app advertising multi-billion dollar industry, is an essential part current digital ecosystem driven free applications, where entities usually comprise consumer apps, clients (consumers), ad-networks, advertisers. Sensitive information often being sold downstream in this...

Service liability interconnections among globally networked IT- and IoT-driven service organizations create potential channels for cascading disruptions worth billions of dollars, due to modern cyber-crimes such as DDoS, APT, ransomware attacks. A natural question that arises in this context is: What is the likelihood a cyber-blackout? , where latter term defined probability all (or major subset of) chain become dysfunctional certain manner cyber-attack at some or points chain. The answer...

Smartphones are nowadays the dominant end-user device. As a result, they have become gateways to all users' communications, including sensitive personal data. In this paper, we present Aquilis, privacy-preserving system for mobile platforms following principles of contextual integrity define appropriateness an information flow. Aquilis takes form keyboard that reminds users potential privacy leakages through simple three-colour code. considers instantaneous risk related posting (Local...

This paper proposes the neural publish/subscribe paradigm, a novel approach to orchestrating AI workflows in large-scale distributed systems computing continuum. Traditional centralized broker methodologies are increasingly struggling with managing data surge resulting from proliferation of 5G systems, connected devices, and ultra-reliable applications. Moreover, advent AI-powered applications, particularly those leveraging advanced network architectures, necessitates new orchestrate...

Continual learning approaches help deep neural network models adapt and learn incrementally by trying to solve catastrophic forgetting. However, whether these existing approaches, applied traditionally image-based tasks, work with the same efficacy sequential time series data generated mobile or embedded sensing systems remains an unanswered question. To address this void, we conduct first comprehensive empirical study that quantifies performance of three predominant continual schemes...

As the use of Large Language Models (LLMs) becomes more widespread, understanding their self-evaluation confidence in generated responses increasingly important as it is integral to reliability output these models. We introduce concept Confidence-Probability Alignment, that connects an LLM's internal confidence, quantified by token probabilities, conveyed model's response when explicitly asked about its certainty. Using various datasets and prompting techniques encourage model introspection,...

The recent emergence of the small cloud (SC), both in concept and practice, has been driven mainly by issues related to service cost complexity commercial providers (e.g., Amazon) employing massive data centers. However, resource inelasticity problem faced SCs due their relatively scarce resources might lead a potential degradation customer QoS loss revenue. A proposed solution this recommends federated sharing between competing alleviate that arise. Based on idea, effort SC-Share,...