Generative AI offers scalable engagement with digital cultural heritage, yet current workflows rely on complex prompts, creating an interaction bottleneck for non-expert audiences. To address this gap, we introduce MoodCurator, a web-based design probe enabling low-friction, implicit affective steering for curatorial storytelling. The system replaces prompt engineering with a three-channel loop: audiences (1) select a colour palette to suggest narrative tone; (2) curate artworks to ground the narrative in visual content; and (3) choose an interpretive voice to constrain rhetorical stance. A mixed-methods study (N = 64) demonstrated encouraging patterns of perceived usability and user agency. A follow-up think-aloud study (N = 10) surfaced recurring cases of interpretive mismatch in this English-language deployment. Contributions include: (1) a legible steering paradigm for AI-supported cultural interpretation; (2) exploratory empirical evidence on usability, agency, and narrative resonance; and (3) preliminary design implications for more explicit and contestable framing controls in future systems.

Rhythm and articulation are essential for expressive guitar performance. Existing tools provide basic beat cues, whereas beginners often struggle to align with these cues when playing complex techniques, such as strumming and muting. Informed by a formative study with five instructors and grounded in embodied learning theories, we present FretFlow, a haptic vest-based tool that simulates common instructional practices to guide learners through physical interactions like tapping. The key to FretFlow is its design space that maps rhythmic and articulation patterns in various playing techniques to distinct haptic patterns, enabling authoring of haptic scores. FretFlow further dynamically adapts haptic intensity based on learners’ real-time performance accuracy, accompanied by multimodal guidance across haptic, visual, and audio channels. We iteratively refined haptic designs across two rounds with 46 participants, followed by a two-week user study with 20 beginners. Results show that FretFlow improves learners’ rhythmic accuracy and expressive performance.

Current data protection legal frameworks, including the GDPR, classify “special” categories of personal data that are deemed deserving of higher protection due to their impact on fundamental rights. Yet, these legal abstractions fail to capture how individuals themselves judge sensitivity in everyday digital contexts. This disconnect may undermine intelligibility and erode trust in data protection as a legal institution. Despite its centrality to privacy protection, limited empirical work has systematically compared public sensitivity judgments against the special categories of protected data under Article 9 GDPR. We address this gap through a mixed-methods design that integrates nine semi-structured interviews with a game-like survey deployed on public arcade machines. This approach generated 2,935 responses from 224 participants enabling in-situ analysis of everyday judgments. By operationalising an ontology capturing who collects data, what data are collected, and for what purpose, we systematically compared responses across demographic groups. Contrary to literature assumptions that health and financial data are primary markers of data sensitivity, our findings demonstrate that expressive content, messages, photos, and social ties elicited the strongest resistance to sharing by citizens. Acceptance was shaped decisively by purpose. Citizens tolerated safety and functionality, whilst advertising and vague claims of “research” were rejected. Attitudes varied systematically, with women disproportionately resistant to sharing expressive content, and higher education and digital literacy predicting greater caution. This study demonstrates that data sensitivity cannot be reduced to fixed legal categories. Rather, it is socially situated and purpose-dependent. Our findings provide empirical foundations for reimagining consent flows, privacy defaults, and transparency mechanisms that align with everyday logics. This can enable the development of systems that people can genuinely understand, trust, and consent to.

The United Nations’ Sustainable Development Goals (UN SDGs) prioritise inclusive and fair employment. However, AI-powered recruitment tools—particularly Large Language Models (LLMs)—raise concerns about potential demographic bias. This paper presents a controlled synthetic dataset and methodology to measure how sensitive attributes (e.g., race, gender, age) influence candidate rankings and pairwise comparisons in LLM-based hiring pipelines. Specifically, we generated a balanced dataset of 1,000 synthetic candidate profiles (each including a cover letter) and evaluated it using 28 frontier LLMs, including proprietary (e.g., OpenAI GPT, Gemini, Grok, Claude) and opensource (e.g., Llama, GigaChat) models. Synthetic data eliminates real-world demographic/occupational confounders, ensuring observed disparities reflect only LLMs’ intrinsic behaviour. Results show professional attributes (e.g., skills, experience) are primary ranking drivers, with 76%–80% statistically significant; however, 8%–9% of demographic attributes exhibit persistent, significant biases across multiple LLMs.We develop a “bias map” quantifying LLM performance, emphasising that mitigating even minor biases in automated hiring is critical to avoid perpetuating employment inequities and uphold the UN SDGs’ inclusive vision.

In this research, we investigate the influence of utilizing artificial intelligence (AI)-generated photographs on malicious bots that engage in disinformation, fraud, reputation manipulation, and other types of malicious activity on social networks. Our research aims to compare the performance metrics of social bots that employ AI photos with those that use other types of photographs. To accomplish this, we analyzed a dataset with 13 748 measurements of 11 423 bots from the VK social network and identified 73 cases where bots employed generative adversarial network (GAN)-photos and 84 cases where bots employed diffusion or transformers photos. We conducted a qualitative comparison of these bots using metrics such as price, survival rate, quality, speed, and human trust. Our study findings indicate that bots that use AI-photos exhibit less danger and lower levels of sophistication compared to other types: AI-enhanced bots are less expensive, less popular on exchange platforms, of inferior quality, less likely to be operated by humans, and, as a consequence, faster and more susceptible to being blocked by social networks. We also did not observe any significant difference between GAN-based and diffusion/transformers-based bots, indicating that diffusion/transformers models did not contribute to increased bot sophistication compared to GAN models. Our contributions include a proposed methodology for evaluating the impact of photos on bot sophistication, along with a publicly available dataset for other researchers to study and analyze bots. Our research findings suggest a contradiction to theoretical expectations: in practice, bots using AI-generated photos pose less danger.

Leveraging user behavior has become an increasingly valuable resource for modeling and personalizing systems based on the unique characteristics of each user. While recent studies have recently been conducted along these lines, there is still a lack of understanding of the relationship between students’ Myers-Briggs Type Indicators and their behavior within educational systems. Facing this problem, we conducted a long-term study (15 weeks) with 96 students, analyzing how their engagement metrics and communication frequency in a Moodle Learning Management System are related to their Myers-Briggs personality types (i.e., extroversion/introversion, sensing/intuition, thinking/feeling, and judging/perceiving). The primary findings indicate that i) participants identified as extroverted demonstrated heightened activity levels throughout more weeks of the course, and ii) participants characterized by judging and thinking traits engaged in a greater number of activities over the course duration. The results contribute to the field of educational technologies by providing valuable insights into the relationships between different characteristics associated with the Myers-Briggs Type Indicator and students’ behavior when using an educational system.

Gamification is commonly employed to support the formation of positive psychological states and learning outcomes, with one of the primary psychological factors chiefly relevant to learning being the flow state. However, the effects of gamification on students’ flow experience are still little known. Filling this gap, we conducted a between-subjects controlled experiment (N = 65) to analyze the effects of gamification on students’ flow experience. Using descriptive and inferential statistical techniques, we compared the flow experience between participants who used a gamified version of an educational system (experimental group) and a group that used the same system without gamification (control group). The main results indicate that the employed gamification design did not affect students’ flow experience. Our study contributes especially to educational technologies and gamification fields, demonstrating that gamification may not affect students’ flow experience.

The STEAM approach in education has been gaining increasing popularity over the last decade. This is due to its potential in enhancing students' learning, when teaching arts and scientific disciplines together. This paper introduces the PICA-PICA concept, where we aim to develop a smart customisable environment, combining, in a unique way, teaching programming in conjunction with the engineering of artworks. The PICA-PICA concept was implemented, used and tested in real-life, by upper primary school students in Japan, during a 4-day workshop. Initial results illustrated the quality of the solution proposed by PICA-PICA. We noted that the integration was perceived as smooth, and not contrived: all participants understood how to use the PICA-PICA environment to engineer programmable art objects. Furthermore, the PICA-PICA approach led to high motivation: children did not get bored and were fully engaged. Finally, the quality of their work as a learning outcome was high: by including a programming segment with the other expressive activities in the artwork, the children were able to design the electronics in a more concentrated and meaningful way than their curriculum-structured learning. This study also presents an innovative implementation of the STEAM approach using Micro:bits technology to create exciting artwork whilst using household recyclable items, which also teaches about sustainability. The involvement of parents and their interest in learning is another unique aspect of this study.

Technological advances in the smart home have created new opportunities for supporting digital citizens’ well-being and facilitating their empowerment but have enabled new types of complex online harms to develop. Recent statistics have indicated that ‘smart’ technology ownership increases yearly, driven by lower costs and increased accessibility. Research on smart homes has also grown, focusing on technology perspectives at the expense of a user-centric approach sensitive to the smart home’s harms, risks, and vulnerabilities. This scoping review addresses the information gap by underscoring the scope of literature that exists regarding complex online harms, vulnerabilities, and risks associated with smart home technologies and citizens’ agency. The goal is to understand the state of knowledge, gaps in the literature, and areas for future study. The importance and originality of this paper lie in its interdisciplinary review and approach. It is hoped that this research will contribute to a deeper understanding of complex online harms in the smart home. Three online databases were utilised to identify papers published between 2017 and 2022, from which we selected 235 publications written in English that addressed harms, risks, vulnerabilities, and agency in the smart home context. This allowed us to map contemporary literature to reveal significant gaps in our understanding of the complex online harms affecting smart home users and identify opportunities for further research. This review identified emerging themes of ‘risks’, ‘vulnerabilities’, and ‘harms’ in that order of frequency within the literature on smart homes. The usage of terms is skewed towards computing science and information security, which comprised the majority of the literature at 54.6%. Human–computer interaction papers contributed 24.4%, while social sciences accounted for 16.2%. Risks, harms and vulnerabilities within smart home ecosystems and IoTs are ongoing issues with complexities that necessitate research. Privacy, security, and well-being are key themes that embody the scope of complex harms affecting smart home devices in the broad literature. This review establishes disciplinary research gaps, especially in user-centred perspectives, due to a heavy technology focus in the existing literature. Therefore, further research is needed to address emergent risks, harms and vulnerabilities of smart home devices and understand how user agency and autonomy can complement the design, interface, and socio-technical aspects of smart home systems.

Abstract The continuous application of artificial intelligence (AI) technologies in online education has led to significant progress, especially in the field of Intelligent Tutoring Systems (ITS), online courses and learning management systems (LMS). An important research direction of the field is to provide students with customised learning trajectories via student modelling. Previous studies have shown that customisation of learning trajectories could effectively improve students’ learning experiences and outcomes. However, training an ITS that can customise students’ learning trajectories suffers from cold-start, time-consumption, human labour-intensity, and cost problems. One feasible approach is to simulate real students’ behaviour trajectories through algorithms, to generate data that could be used to train the ITS. Nonetheless, implementing high-accuracy student modelling methods that effectively address these issues remains an ongoing challenge. Traditional simulation methods, in particular, encounter difficulties in ensuring the quality and diversity of the generated data, thereby limiting their capacity to provide intelligent tutoring systems (ITS) with high-fidelity and diverse training data. We thus propose Sim-GAIL, a novel student modelling method based on generative adversarial imitation learning (GAIL). To the best of our knowledge, it is the first method using GAIL to address the challenge of lacking training data, resulting from the issues mentioned above. We analyse and compare the performance of Sim-GAIL with two traditional Reinforcement Learning-based and Imitation Learning-based methods using action distribution evaluation, cumulative reward evaluation, and offline-policy evaluation. The experiments demonstrate that our method outperforms traditional ones on most metrics. Moreover, we apply our method to a domain plagued by the cold-start problem, knowledge tracing (KT), and the results show that our novel method could effectively improve the KT model’s prediction accuracy in a cold-start scenario.

Medical visual question answering (Med-VQA) is to answer medical questions based on clinical images provided. This field is still in its infancy due to the complexity of the trio formed of questions, multimodal features and expert knowledge. In this paper, we tackle, a ’myth’ in the Natural Language Processing area - that unimodal bias is always considered undesirable in learning models. Additionally, we study the effect of integrating a novel dynamic attention mechanism into such models, inspired by a recent graph deep learning study.Unlike traditional attention, dynamic attention scores are conditioned on different query words in a question and thus enhance the representation learning ability of texts. We propose that some questions are answered more accurately with a reinforcement of question embedding after fusing multimodal features. Extensive experiments have been implemented on the VQA-RAD datasets and demonstrate that our proposed model, reinforCe unimOdal dynamiC Attention (COCA), outperforms the state-of-the-art methods overall and performs competitively at open-ended question answering.

Graph neural networks (GNNs) have attracted extensive interest in text classification tasks due to their expected superior performance in representation learning. However, most existing studies adopted the same semi-supervised learning setting as the vanilla Graph Convolution Network (GCN), which requires a large amount of labelled data during training and thus is less robust when dealing with large-scale graph data with fewer labels. Additionally, graph structure information is normally captured by direct information aggregation via network schema and is highly dependent on correct adjacency information. Therefore, any missing adjacency knowledge may hinder the performance. Addressing these problems, this paper thus proposes a novel method to learn a graph structure, NC-HGAT, by expanding a state-of-the-art self-supervised heterogeneous graph neural network model (HGAT) with simple neighbour contrastive learning. The new NC-HGAT considers the graph structure information from heterogeneous graphs with multilayer perceptrons (MLPs) and delivers consistent results, despite the corrupted neighbouring connections. Extensive experiments have been implemented on four benchmark short-text datasets. The results demonstrate that our proposed model NC-HGAT significantly outperforms state-of-the-art methods on three datasets and achieves competitive performance on the remaining dataset.

XAI with natural language processing aims to produce human-readable explanations as evidence for AI decision-making, which addresses explainability and transparency. However, from an HCI perspective, the current approaches only focus on delivering a single explanation, which fails to account for the diversity of human thoughts and experiences in language. This paper thus addresses this gap, by proposing a generative XAI framework, INTERACTION (explain aNd predicT thEn queRy with contextuAl CondiTional varIational autO-eNcoder). Our novel framework presents explanation in two steps: (step one) Explanation and Label Prediction; and (step two) Diverse Evidence Generation. We conduct intensive experiments with the Transformer architecture on a benchmark dataset, e-SNLI [1]. Our method achieves competitive or better performance against state-of-the-art baseline models on explanation generation (up to 4.7% gain in BLEU) and prediction (up to 4.4% gain in accuracy) in step one; it can also generate multiple diverse explanations in step two.

Despite rapid advances of modern artificial intelligence (AI), there is a growing concern regarding its capacity to be explainable, transparent, and accountable. One crucial step towards such AI systems involves reliable and efficient uncertainty quantification methods. Existing approaches to uncertainty quantification in natural language processing (NLP) take a Bayesian Deep Learning approach. However, the latter is known to not be computationally efficient in testing time, thus hindering its applicability in real-life scenarios. This paper proposes a new focus on the efficiency of uncertainty quantification methods, evaluating them on four multi-label text classification tasks. Our novel methods of representing epistemic and aleatoric uncertainties enable efficient uncertainty quantification (around 13 to 45 times faster than existing approaches, depending on architecture) with posterior analysis in the (approximated) latent- and data space. We conduct extensive experiments and studies on diverse neural network architectures (LSTM, CNN and Transformer) to analyse their power. Our results prove the benefits of explicitly modelling uncertainty in neural networks.

Recently, methods enabling humans and Artificial Intelligent (AI) agents to collaborate towards improving the efficiency of Reinforcement Learning - also called Collaborative Reinforcement Learning (CRL) - have been receiving increasing attention. In this paper, we provide a long-term, in-depth survey, investigating human-AI collaborative methods based on both interactive reinforcement learning algorithms and human-AI collaborative frameworks, between 2011 and 2020. We elucidate and discuss synergistic analysis methods of both the growth of the field and the state-of-the-art; we suggest novel technical directions and new collaboration design ideas. Specifically, we provide a new CRL classification taxonomy, as a systematic modelling tool for selecting and improving new CRL designs. Furthermore, we propose generic CRL challenges providing the research community with a guide towards effective implementation of human-AI collaboration. The aim is to empower researchers to develop more efficient and natural human-AI collaborative methods that could utilise the different strengths of humans and AI.

As the largest distance learning university in the UK, the Open University has more than 250,000 students enrolled, making it also the largest academic institute in the UK. However, many students end up failing or withdrawing from online courses, which makes it extremely crucial to identify those “at risk” students and inject necessary interventions to prevent them from dropping out. This study thus aims at exploring an efficient predictive model, using both behavioural and demographical data extracted from the anonymised Open University Learning Analytics Dataset (OULAD). The predictive model was implemented through machine learning methods that included BART. The analytics indicates that the proposed model could predict the final result of the course at a finer granularity, i.e., classifying the students into Withdrawn, Fail, Pass, and Distinction, rather than only Completers and Non-completers (two categories) as proposed in existing studies. Our model’s prediction accuracy was at 80% or above for predicting which students would withdraw, fail and get a distinction. This information could be used to provide more accurate personalised interventions. Importantly, unlike existing similar studies, our model predicts the final result at the very beginning of a course, i.e., using the first assignment mark, among others, which could help reduce the dropout rate before it was too late.

Gamification has been widely employed in the educational domain over the past eight years when the term became a trend. However, the literature states that gamification still lacks formal definitions to support the design of gamified strategies. This paper aims to create a taxonomy for the game elements, based on gamification experts' opinions. After a brief review from existing work, we extract first the game elements from the current state of the art, and then evaluate them via a survey with 19 gamification and education experts. The resulting taxonomy taxonomy included the description of 21 game elements and their quantitative and qualitative evaluation by the experts. Overall, the proposed taxonomy was in general well accepted by most of the experts. They also suggested expanding it with the inclusion of Narrative and Storytelling game elements. Thus, the main contribution of this paper is proposing a new, confirmed taxonomy to standardise the terminology used to define the game elements as a mean to design and deploy gamification strategies in the educational domain.

Here we present a case study that analyzed the social interaction features in Topolor, an adaptive personalized social e-learning system. This paper focuses on the evaluation of the perceived usefulness and usability. The results show a considerably high satisfaction of the students. We discuss the evaluation results and outline the further improvement plan.

We introduce Topolor, a social personalized adaptive e-learning system aiming to improve social interaction in learning process, and apply classical adaptation based on user modeling. Here, we focus on the system architecture and preliminary evaluation.