The increasing pervasiveness of software-intensive systems requires involving domain experts more directly in technological development. Visual models, expressed in semi-formal notations, can act as shared artefacts that support communication and collaboration between developers and domain experts. However, modelling with semi-formal notations can be challenging for novice modellers. This study presents an AI-infused, web-based modelling tool designed to support users in formalising domain knowledge without requiring advanced modelling skills. The tool features a block-based, domain-specific language that automatically transforms user-generated structures into semi-formal diagrams. AI-based functionalities include a diagram reader, contextual hints, natural-language instructions, and interaction logging. We evaluated the tool through a Wizard of Oz experiment with agronomists in digital agriculture, where participants completed an exploratory modelling task while interacting with AI assistance. Results reveal three key design implications: (i) adaptive AI support accommodating diverse modelling strategies, (ii) concise, actionable guidance delivered at moments of difficulty, and (iii) practice-oriented assistance that preserves user agency and supports learning-by-doing.

The notion of explainable user interfaces (XUI) has been proposed as a way to address the growing complexity and unpredictability of many UIs. A key question is whether this aspirational goal can be achieved with reasonable amounts of development effort. This paper takes React as its focus as it is one the most widely used web development frameworks. It shows through practical demonstration that the existing event mechanisms within React can be repurposed with the addition of a small amount of annotation to implement one of the proposed XUI techniques.

This paper summarises the third edition of the SYNERGY workshop on Designing and Building Hybrid Human-AI Systems, held at AVI 2026 in Venice, Italy. The workshop addresses the critical challenge of designing AI systems that genuinely augment human capabilities through meaningful combination, moving beyond approaches where humans function as mere cogs in the machine. Directly aligned with AVI 2026's theme of "Interactive Creativity: Agencies, Interfaces, and Ethics", the workshop brings together researchers and practitioners to advance both theoretical frameworks and practical implementations that preserve human agency while leveraging AI's computational power.

This workshop explores the evolving landscape of Human-AI collaboration, focusing on Advanced Visual Interfaces and Artificial Intelligence to enhance human cognition. We explore synergistic models of collaboration that merge human insights with AI capabilities, addressing ethical dimensions and practical AI applications. Our goal is to foster rich interdisciplinary dialogue and challenge existing paradigms of human-machine interaction. We aim to redefine interaction paradigms and establish new benchmarks for intelligent systems, ensuring AI not only supports but significantly augments human decision-making processes.

This work explores the potential of Visual Programming Languages (VPLs) and no-code platforms to foster participation among users with limited computing experience in the design of ML-based systems. Conducting expert-based testing of the PyFlowML prototype, our preliminary research focuses on developing trustworthy ML-based prototypes that provide Explainable AI (XAI) techniques. Our evaluation employs heuristic methods and direct interactions with experts using the prototype. Utilizing cognitive walkthroughs with a think-aloud protocol, along with quantitative assessments such as task completion time and the System Usability Scale (SUS), our findings highlight the need of streamlining ML processes to enhance broader participation. These insights lay the groundwork for future research aimed at making the design of ML-based systems more accessible and collaborative through VPL-based tools.

This paper introduces FlowPilot, a novel flow-based visual programming language designed to enhance Computational Thinking (CT) education. FlowPilot leverages generative AI to create a dynamic, browser-based environment where users can construct programs using natural language descriptions. By integrating AI-driven block generation with a flow-based visual interface, FlowPilot supports key CT pillars such as abstraction and decomposition. This approach offers a unique platform for learners to explore programming concepts at various levels of complexity, fostering a deeper understanding of computational processes.

In the era of ubiquitous Artificial Intelligence (AI), Machine Learning (ML) systems have emerged as pivotal in advancing societal progress. Yet, their complexity often requires the expertise of professionals, presenting an entry barrier for less-experienced computing practitioners. In the realm of Human-Centered AI (HCAI), ongoing research is developing new approaches to engage users via intuitive interfaces in the design of ML-based systems. For example, Visual Programming Languages (VPLs) and no-code platforms offer user-friendly experiences by avoiding complex computational operations. However, literature on VPL-based tools for novices reveals a gap: many tools offer limited interpretability approaches or are overly complex for easy customization. To address this gap, we developed the PyFlowML prototype, an open-source, flow-based programming tool with a focus on Explainable AI (XAI). This study aims to set a benchmark in the field by comparing PyFlowML with KNIME, a similar tool known for its usability among domain specialists. Our results show that domain specialists perceive flow-based language tools as valuable. This study highlights the potential of VPL-based tools to make the design of trustworthy ML-based systems more inclusive, thereby contributing to the advancement of AI democratization.

The ongoing digital revolution in the healthcare sector, emphasized by bodies like the US Food and Drug Administration (FDA), is paving the way for a shift towards person-centric healthcare models. These models consider individual needs, turning patients from passive recipients to active participants. A key factor in this shift is Artificial Intelligence (AI), which has the capacity to revolutionize healthcare delivery due to its ability to personalize it. With the rise of software in healthcare and the proliferation of the Internet of Things (IoT), a surge of digital data is being produced. This data, alongside improvements in AI's explainability, is facilitating the spread of person-centric healthcare models, aiming at improving health management and patient experience. This paper outlines a human-centered methodology for the development of an AI-as-a-service platform with the goal of broadening access to personalized healthcare. This approach places humans at its core, aiming to augment, not replace, human capabilities and integrate in current processes. The primary research question guiding this study is: "How can Human-Centered AI principles be considered when designing an AI-as-a-service platform that democratizes access to personalized healthcare?" This informed both our research direction and investigation. Our approach involves a design fiction methodology, engaging clinicians from different domains to gather their perspectives on how AI can meet their needs by envisioning potential future scenarios and addressing possible ethical and social challenges. Additionally, we incorporate Meta-Design principles, investigating opportunities for users to modify the AI system based on their experiences. This promotes a platform that evolves with the user and considers many different perspectives.

We present CINT (City Net-zero Tool): an analytical and predictive model designed to quantitatively assess carbon data in urban areas. We developed a workflow to collect existing data from city councils on carbon footprint, consumption and production, and tested the interoperability between urban public data and GIS data. We implemented the model using Kernel Density Estimation (KDE) to infer the carbon emissions related to individual buildings based on a station-based dataset. We present initial testing on the integration of data from OpenStreetMap with a vector model and initial testing on modelling data into graph networks to generate enquiries and inference on carbon data and urban scenarios. This method shows how we can integrate more datasets into our base model (graph-based geo-referenced map) to infer unknown information (for example the estimated NO2 emission per each building).

This contribution presents ModeLLer, a prototype of a web tool for system modelling based on a block-based visual editor. The aim of ModeLLer is to enable collaborative environments in requirements elicitation, allowing end-users to create UML class diagrams without any knowledge of the (semi-)formal UML notation.

Algorithms are more and more pervading our everyday life: from automatic checkouts in supermarkets and e-banking to booking a flight online. Understanding an algorithmic solution to a problem is a very relevant activity to improve end-users’ involvement. To this end, adopting a meta-design approach may help to support end-users to appropriate the design skills necessary for contributing to system design, in new and engaging modalities. By acquiring Computational Thinking (CT) skills (e.g., algorithmic thinking, abstraction), end-users will be able to understand and trust algorithms, while at the same time participate in the design and development of systems evolving in accordance with their needs. In this work, we focus on two different ways of improving CT skills: playfulness and collaboration. We introduce a game-based system, TAPASPlay, to foster CT skills and we report the results of an exploratory study with 18 users; our hypothesis is that learning CT through gameplay is effective and we tested it by involving participants in game sessions providing playful experience and collaborative learning.

In recent years, smart objects are increasingly pervading the environments we live in. For HCI researchers, an important challenge is how non-technical users can establish the behavior of such devices. This poster presents a new technology implementing a tangible-programming paradigm, which allows non-programmers to synchronize the behavior of ecologies of smart objects, thus determining the creation and customization of smart environments.

In this paper we describe how to use a Game-Based Learning (GBL) approach to help end-user programmers to develop Computational Thinking (CT) skills. CT represents a framework consisting of a set of skills such as: abstraction, decomposition and more in general problem solving expressed in a computational form (algorithmic thinking) that can be easily formalised into a computer program. Those skills are beneficial for end-user programmers in order to model a task for being tackled with computational tools, e.g. spreadsheets, domain specific and visual languages. TAPASPlay is an engaging GBL platform designed to help end-user to learn algorithmic thinking.

Treemaps have the desirable property of presenting overviews along with details of data and thus are of interest in visualizations of multi-attribute tabular data with attribute hierarchies. However, the original treemap algorithms and most subsequent variations are hampered in making parallel structures in a hierarchical data structure visually comparable. Structurally parallel elements are not aligned, making it difficult to compare them visually. We propose a method that allows for proportional and non-proportional subdivisions of subtrees while preserving visual alignment of parallel structures. We extend the framework so that other types of data visualizations can be placed within the graphical areas of a treemap to allow for the visual comparison of a broad collection of data types including temporal data.

Context: It is unclear that current approaches to evaluating or comparing competing software cost or effort models give a realistic picture of how they would perform in actual use. Specifically, we're concerned that the usual practice of using all data with some holdout strategy is at variance with the reality of a data set growing as projects complete.

Computational Thinking has recently returned into the limelight as an essential skill to have for both the general public and disciplines outside Computer Science. It encapsulates those thinking skills integral to solving complex problems using a computer, thus widely applicable in our technological society. Several public initiatives such as the Hour of Code successfully introduced it to millions of people of different ages and backgrounds, mostly using Blocks Programming Environments like Scratch that lower the barriers of programming and facilitate learning. In this paper we present our arguments for fostering Computational Thinking skills using a Blocks Programming Environment augmented with a Tangible User Interface, namely by exploiting objects whose interactions with the physical environment are mapped to digital actions performed on the system. Our demonstration includes a working prototype implementing our Tangible Blocks Programming Environment called TAPAS.

Nowadays we are continuously surrounded by new digital systems featuring easy to use and engaging interaction modalities, such as multi-touch, gesture, tangible, or voice interaction. The enriched and natural experience provided by this new paradigm - known as Natural User Interface - has promoted its adoption in many ubiquitous contexts due to its effectiveness in dealing with a wide audience of mostly inexperienced users; Pervasive Displays, namely variously-sized displays ecosystems supporting simultaneous interactions with public screens naturally use those new paradigms. Nonetheless, if we want these systems to stay out in public spaces for long periods of time, they must provide users with an easy way of being adapted to their heterogeneous usage contexts. In this paper, we propose an End- User Programming approach to this problem introducing TAPAS, a system that combines a tangible interaction with a puzzle metaphor, allowing users to create workflows on a Pervasive Display to satisfy their needs; we also carried out a preliminary evaluation of our system with second year university students and interaction designers, gathering useful feedback to improve TAPAS and employ it in many other domains.