SEARCH · Searcharxiv

Search Searcharxiv

Search indexed arXiv papers on artificial intelligence, large language models, computer vision and robotics. Read source abstracts and follow links to arXiv.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 235 records · Page 13Linked to original sources

arXiv2026

Challenging Techno-Solutionism: The Role of ICT Innovation and the Value of Technological Growth

Innovation in Information and Communication Technology (ICT) has become one of the key economic drivers of our technology-dependent world. Digital devices/systems have become so pervasive that it is hard to imagine new technology developments that are not totally or partially influenced by ICT innovations. Furthermore, the pace of innovation in ICT sector over the last few decades has been unprecedented in human history. In this paper, we argue that the ICT innovation paradigm has crucially shaped collective expectations and imagination about what technology more broadly can actually deliver, particularly for a more sustainable and equitable world. These expectations have often crystalised into a widespread acceptance, among general public and policy makers, of techno-solutionism. We emphasise the role of electronic microchips in deriving relentless innovation and its impacts. We identify the many impacts of this innovation cycle into three different categories: embodied, induced and implied impact.

cs.CY↗

arXiv2026

Cognitive Chain-of-Thought (CoCoT): Structured Multimodal Reasoning about Social Situations

Chain-of-Thought (CoT) prompting helps models think step by step. But naive CoT breaks down in visually grounded social tasks, where models must perceive, understand, and judge all at once; bridging perception with norm-grounded reasoning. Recent work has introduced structured reasoning for multi-turn agent planning and visual QA, decomposing tasks into sequential sub-goals. To extend this to single-shot multimodal social reasoning, we introduce Cognitive Chain-of-Thought (CoCoT), a reasoning framework that structures vision-language-model (VLM) reasoning through three cognitively inspired stages: Perception (extract grounded facts), Situation (infer situations), and Norm (applying social norms). Evaluation across multiple distinct tasks such as multimodal intent disambiguation, multimodal theory of mind, social commonsense reasoning, and safety instruction following, shows consistent improvements (5.9% to 4.6% on average). We further explore the utility of CoCoT for improving models' reasoning through training and show that supervised fine-tuning on CoCoT-structured traces yields 5-6% improvements without explicit CoCoT prompting at inference, demonstrating that models internalize the structured reasoning pattern rather than merely following instructions. We show that structuring model reasoning through cognitively grounded stages enhances interpretability and social alignment, laying the groundwork for more reliable multimodal systems.

cs.CL↗

arXiv2026

Why are all LLMs Obsessed with Japanese Culture? On the Hidden Cultural and Regional Biases of LLMs

LLMs have limitations when it comes to cultural coverage and competence, and in some cases, show specific cultural biases. Although prior studies have examined the cultural capabilities of LLMs, none have specifically investigated their regional preferences in generic culture-related questions. In this work, we propose a new dataset based on a comprehensive taxonomy of Culture-Related Open Questions (CROQ), with questions available in 24 languages. We evaluate LLMs by prompting them to answer questions from CROQ and provide a sample location. The results show that, contrary to previous cultural bias work, LLMs show a clear tendency towards countries such as Japan in their answers. Moreover, our results show that when prompting in languages such as English or other high-resource ones, LLMs tend to provide more diverse outputs. Low-resource languages, on the other hand, show more inclinations towards answering questions highlighting countries for which the input language is an official language. Finally, we also investigate at which point of LLM training this cultural bias emerges, with our results suggesting that the first clear signs appear after supervised fine-tuning, and not during pre-training. Dataset available at https://huggingface.co/datasets/HiTZ/CROQ

cs.CL↗

arXiv2026

What's on Your Mind? Exploring Privacy of Mental Health Apps

Therapy and life-coaching apps have grown rapidly in number, variety, and popularity. At the same time, their users often share highly sensitive and personal information, including mental health issues, trauma experiences, fantasies, desires, and relationship difficulties. This prompts the need to examine privacy practices across the ecosystem. In this paper, we present a comprehensive analysis of a corpus of 25 popular Android mental health and life-coaching apps, such as Replika and Headspace. It builds on static analysis and dynamic network traffic analysis, coupled with identifying gaps between each app's observed behavior and its privacy policies. Our analysis highlights serious concerns and substantial transparency gaps. First, every app in our corpus embeds at least one tracker SDK not named in its privacy policy, and 85\% of the apps we instrument fail to disclose at least half of the trackers detected in their APKs. Second, more than half of the apps declare several dangerous permissions without corresponding privacy-policy disclosures, including camera or microphone permissions. Third, nearly half the apps disclose third-party AI processing (e.g., via OpenAI, Anthropic, and Groq) in their privacy policies, while several use only generic language (e.g., "AI services"), failing to identify which company receives user data. Overall, our results show that current disclosure practices fall short of the transparency needed for meaningful informed consent. We argue for a significantly updated regulatory framework for therapy apps that more closely aligns with the professional and ethical standards that bind licensed human therapists.

cs.CR↗

arXiv2026

Impact of introducing "Informatics I" to the common university entrance examination in Japan: a longitudinal study on students' perceptions of their information-related knowledge and skills from 2006 to 2026

Despite the recent intensive development of secondary education curricula and assessments in informatics, the impact of assessments has not been well studied in this field. Since informatics education covers a diverse range of content, from computer science knowledge to ICT skills, careful consideration is needed to prevent assessments from distorting education. This study investigates the impact of introducing ``Informatics I'' into the Common Test for University Admissions in Japan, as an example of a large-scale, standardized, high-stakes assessment in 2025. As the data source for this analysis, this study uses a questionnaire that has been administered every year from 2006 to 2026 to all first-year students at the University of Tokyo. The questionnaire asks students for their self-perceptions of the information-related knowledge and skills they studied and acquired in high school. Using these data, we conduct a longitudinal study of the 2013 curriculum reform, the 2022 reform, and the introduction of the new entrance examination. We attempt to isolate the impact of the entrance examination through two comparisons: between the 2013 curriculum reform and the 2022 reform; and between direct-entry and gap-year students among those entering in 2025, who followed different curricula but took the new examination. We use the theoretical framework of the washback effect as a lens for interpreting these differences. We found that (1) the 2013 curriculum reform produced no discontinuity in students' perceptions, whereas (2) the introduction of the new entrance examination in 2025 produced a sharp change, particularly in the proportion of students reporting acquisition of computer-science topics; and (3) this change is too large to be interpreted as a gain in proficiency, and is better understood as a shift in students' criteria for judging acquisition.

cs.CY↗

arXiv2026

Energy and CO2 Footprint of Climate Model Intercomparison Projects

Earth System Models (ESMs) rely heavily on High-Performance Computing (HPC) resources to simulate global climate. As these models evolve, their computational demands continue to grow, driven by three factors: (1) finer spatial grid resolutions, (2) the integration of complex biogeochemical processes (e.g., atmospheric chemistry, interactive vegetation, land use, and ice sheets), and (3) larger climate ensembles to manage uncertainty. Historically, growth in peak computing performance (FLOP/s) has outpaced improvements in energy efficiency (FLOP/Watt), increasing total HPC power consumption. Despite the central role of Model Intercomparison Projects (MIPs) in climate research, quantifying their computational and environmental costs has received limited systematic attention. This paper examines the evolution of climate model carbon accounting from voluntary post-hoc estimation in the Coupled Model Intercomparison Project phase 6 (CMIP6) to standardized accounting under the newly established CMIP7 Task Team on Energy Consumption. Using high-resolution Destination Earth simulations on MareNostrum 5, we empirically evaluate how different accounting boundaries (operational, active-only, and embodied carbon) impact reported energy, carbon emissions, and financial costs. Finally, we outline key methodological considerations for standardizing energy and carbon accounting for Model Intercomparison Projects (MIPs).

cs.CY↗

arXiv2026

Generative AI Alignment with Hinduism's Theological Plurality and Sacred Representation

Generative AI systems are increasingly used to answer personal questions and mediate everyday practices, including religion. However, existing discussions around AI alignment and ethics have largely centered secular, Western, and Abrahamic assumptions about religion, offering limited attention to other faith-based traditions. In this paper, we examine how Hindu users engage with generative AI systems in relation to their religious knowledge, belief, and practice. Drawing on 15 semi-structured interviews with Bangladeshi Hindu participants, we analyze how users interpret AI-generated religious representations, scriptural explanations, devotional interactions, and synthetic religious media. We found that AI can be both accessible and ethically troubling. While AI supported scriptural inquiry, devotional visualization, and religious storytelling, our study also identified concerns about theological flattening, cultural misrepresentation, devotional manipulation, and the simulation of sacred presence and authority. We conclude by arguing that religious alignment in generative AI requires interpretive alignment: systems that disclose their limits, preserve plurality, and avoid simulating sacred authority and sycophantic personalization.

cs.AI↗

arXiv2026

It Takes Three to Converse: Empirical Observations on How the Developer, the Convener and the Participant Shaped 119 Polis Conversations

Polis is a popular democratic innovation tool that allows asynchronous citizen engagement through atomic statements: short statements that together describe a complex question, inviting the citizen to vote Agree or Disagree on each. This paper uses 119 conversations with 100 or more participants and an extensive data export, drawn from a wider set of 271 collected processes. The paper asks what determines the output of such a process. Three parties shape the result. The developer of the platform has made important design choices that restrict the outcome: the number of groups the platform is able to report (restricted to 2--5) and which statements are prioritized. The convener defines the assignment: the initial statements that set the tone, the policy that accepts or rejects new statements and who can be invited. Finally, the participant works within these boundaries. With access to less than half of the generated statements, they end up responding to more statements when their conversation seems to have an achievable number of statements to complete, than when they are presented with more statements. Due to choices such as warm path clustering, the exported resulting clustering cannot be reproduced based on the voting data. Conveners may want to re-analyse their own conversations once the process is closed, to consider the data in its entirety, and make their own analysis priorities explicit.

cs.CY↗

arXiv2026

A Guided Inquiry Approach to Students Co-Designing Generative AI Course Policies

As generative AI (GenAI) use among students increases, educators face growing questions about how to support learning while addressing ethical and institutional concerns. This exploratory study examines a guided inquiry activity in which students co-designed a GenAI course policy. Students first developed individual policy proposals focused on appropriate and ethical use of GenAI, then collaboratively refined them by incorporating diverse stakeholder perspectives. The following research questions guided the study: 1) what practical factors do students prioritize in their GenAI use policies, and how do they justify these choices? and 2) how do participants reflect on the policy design process? Participants first completed readings, then used GenAI to brainstorm initial policy ideas. Next, they articulated their own perspectives through a written assignment and a course policy they designed individually. Finally, they incorporated diverse stakeholder perspectives by collaborating with peers to develop a collective policy. Analysis of student artifacts and group discussions showed that participants prioritized training for students and instructors, standardized procedures for disclosing AI use, and stronger institutional support. Participants also wanted greater involvement in GenAI-related decision-making. They described the policy design process as a way to engage with multiple perspectives and the inherent trade-offs involved in governing AI use. This study offers pedagogical insights into how policy co-design activities can surface student values, concerns, and sensemaking about GenAI in educational contexts.

cs.CY↗

arXiv2026

Stress-testing university AI governance: A prospective method for locating policy breakpoints

Universities are producing AI principles and use policies faster than they are building decision pathways for unfamiliar forms of AI agency. This study develops Institutional AI Governance Stress Testing (IAGST), a prospective documentary method for locating where publicly documented governance ceases to yield an accountable response. IAGST adapts established policy stress-testing and wind-tunneling logic. Its originality lies in combining controlled capability escalation, a frozen documentary corpus, a six-dimensional governance response chain, non-compensatory decision rules, and case-level breakpoint diagnosis. The method was demonstrated using 133 substantive public documents from five Western Australian universities and 15 quality-screened scenarios, resulting in 75 university-scenario encounters. Six cases were resolved, 14 were resolved through structured discretion, and 55 were indeterminate. Governed pathways fell from 16 of 25 augmentation cases to four delegation cases and none at autonomous substitution. The dominant weakness was not the complete absence of responsible roles: all 50 authority-gap cases named a role at only a generic level but lacked sufficient decision criteria or process. The findings show how universities can move beyond policy inventories and principal statements by testing whether authority, procedures, safeguards, and reviews remain connected as AI capabilities evolve. IAGST is a reproducible diagnostic for policy learning, not a ranking or measure of implementation.

cs.CY↗

arXiv2026

Classification Drives Geographic Bias in Street Scene Segmentation

Previous studies showed that image datasets lacking geographic diversity can lead to biased performance in models trained on them. While earlier work studied general-purpose image datasets (e.g., ImageNet) and simple tasks like image recognition, we investigated geo-biases in real-world driving datasets on a more complex task: instance segmentation. We examined if instance segmentation models trained on European driving scenes (Eurocentric models) are geo-biased. Consistent with previous work, we found that Eurocentric models were geo-biased. Interestingly, we found that geo-biases came from classification errors rather than localization errors, with classification errors alone contributing 10-90% of the geo-biases in segmentation and 19-88% of the geo-biases in detection. This showed that while classification is geo-biased, localization (including detection and segmentation) is geographically robust. Our findings show that in region-specific models (e.g., Eurocentric models), geo-biases from classification errors can be significantly mitigated by using coarser classes (e.g., grouping car, bus, and truck as 4-wheeler).

cs.CV↗

arXiv2026

Thinking Inside the Box: Considerations for Putting Data Physicalization Workshops in a Box

Visualization researchers utilize workshops both for applied research and to engage different populations with visualization-based activities. While there are many benefits to running visualization workshops, their utility and impact rely on the presence of a researcher who has deep knowledge about visualization theory and practice. In this work, we introduce workshop-in-a-box as a design concept intended to challenge the researcher-centric approach to data physicalization workshops. Through a design study with a socially innovative organization, we deployed several data physicalization workshops that our collaborator ran instead of us. Based on this experience, along with two accompanying case studies that validate the concept, we present material and procedural considerations for how to put data physicalization workshops into a box and the implications it has for extending visualization research outside the bounds of academia.

cs.HC↗

arXiv2026

Integrating LLM and Diffusion-Based Agents for Social Simulation

Large language models (LLMs) offer strong semantic reasoning capabilities for user modeling, but applying LLM-based simulation to an entire social network is computationally expensive and often unreliable for users with sparse behavioral histories. Meanwhile, conventional information diffusion models efficiently exploit historical propagation patterns and social structures, but provide limited understanding of item content and user-item semantic compatibility. We propose HySID, a hybrid framework for individual-level information adoption prediction that combines semantic reasoning with structural diffusion. HySID first analyzes the historical user-relation graph to adaptively select a small set of structurally informative core users. It then applies LLM-based simulation to estimate the engagement of these users and converts the judgments into a diffusion-compatible seed. Finally, a plug-in diffusion backbone propagates this seed through historical interaction structures to rank potential adopters across the full user population. This design enables LLMs to focus on users for whom semantic reasoning is most informative while allowing neural diffusion models to generalize the evidence to users that are not explicitly simulated. Experiments on three real-world datasets from Weibo, Zhihu, and KuaiRand show that HySID consistently improves four diffusion backbones in Recall and NDCG, while outperforming full-population LLM simulation baselines. At the same time, selective simulation reduces LLM inference cost by approximately 83\% to 99.9\%, demonstrating that HySID provides an effective and computationally practical approach to scalable information adoption prediction.

cs.CY↗

arXiv2026

When Is an Agent Evaluation Over? Outcome Finality and Cross-Unit Separation

Agent evaluations commonly score the state observed when a run stops and count the run as one trial. Interpreting that score as a final result from a separate trial requires outcome finality and cross-unit separation. Outcome finality requires that later events cannot change the claimed result, while cross-unit separation requires that earlier runs cannot change the relevant conditions of later ones. The endpoint establishes neither condition by itself, and the two can hold independently. Waiting for a delayed outcome may settle the label even though its state remains available to another run. Isolation may prevent carryover even though the scored outcome remains unresolved. We develop a completion argument that identifies the evidence needed for each decision. A final success or failure label is justified only when every relevant effect is resolved or bounded tightly enough to fix the outcome. Any remaining uncertainty must be reported. First, in a controlled replay with fixed agent actions, we find that endpoint and terminal labels differ for every nonzero-delay operation and that a delayed write changes the next run's score under shared state but has no such effect after namespacing or verified reset. Second, in a review of ten public protocols, we find that reset or deliberate retention is documented explicitly more often than unfinished operations or evidence for separate scoring. Finally, we propose an open-effects record for operations and resources that may remain relevant after the endpoint, their status, and their possible effects on the scored outcome or another run.

cs.AI↗

arXiv2026

Measuring the Installed Base: Nordic Health Dataset Catalogues Against HealthDCAT-AP Release 7

The European Health Data Space requires member states to publish machine readable descriptions of the health datasets available for secondary use, and the European Commission publishes HealthDCAT-AP as the metadata profile those descriptions are meant to satisfy. The profile has been designed and validated against curated examples, never against the catalogues already live. We report that measurement for the Nordic region. On 25 August 2026 the 11 Nordic national catalogues harvested by the European data portal held 2,811 dataset descriptions carrying the EU health theme, and none satisfies all eight properties HealthDCAT-AP Release 7 makes mandatory on a dataset. Three of the eight are present on exactly zero records across five countries. Set beside the portal's own quality assessment, which validates DCAT-AP and never mentions the health profile, this is not a health extension skipped on top of sound generic practice: no Nordic catalogue reaches the assessment's top rating band, 5 of the 11 are reported at zero per cent DCAT-AP compliant, and the properties surviving in both layers are the ones a human types into a form, not the ones needing a value bound to a controlled vocabulary. Two further results follow. Finland contributes 2,259 descriptions to the European portal of which 1,146 carry a theme, and not one uses the EU theme authority vocabulary, so a European health filter returns no Finnish dataset at all. Separately, the authority namespace answers HTTP 200 with a well formed empty document for terms it never defined, letting 1,238 datasets across the wider portal carry theme IRIs that resolve to nothing while passing any status code check. We publish the vocabulary, the shapes and the harvesters, record every verdict as a dated observation rather than a property of the dataset, and report the five errors this discipline caught before publication.

cs.DL↗

arXiv2026

Comparing Apples to Oranges: A Taxonomy for Navigating the Global Landscape of AI Regulation

AI governance has transitioned from soft law, such as national AI strategies and voluntary guidelines, to binding regulation at an unprecedented pace. This evolution has produced a complex legislative landscape: blurred definitions of "AI regulation" mislead the public and create a false sense of safety; divergent regulatory frameworks risk fragmenting international cooperation; and uneven access to key information heightens the danger of regulatory capture. Clarifying the scope and substance of AI regulation is vital to uphold democratic rights and align international AI efforts. We present a taxonomy to map the global landscape of AI regulation. Our framework targets essential metrics-technology or application-focused rules, horizontal or sectoral regulatory coverage, ex ante or ex post interventions, maturity of the digital legal landscape, enforcement mechanisms, and level of stakeholder participation-to classify the breadth and depth of AI regulation. We apply this framework to five early movers: the European Union's AI Act, the United States' Executive Order 14110, Canada's AI and Data Act, China's Interim Measures for Generative AI Services, and Brazil's AI Bill 2338/2023. We further offer an interactive visualization that distills these dense legal texts into accessible insights, highlighting both commonalities and differences. By delineating what qualifies as AI regulation and clarifying each jurisdiction's approach, our taxonomy reduces legal uncertainty, supports evidence-based policymaking, and lays the groundwork for more inclusive, globally coordinated AI governance.

cs.CY↗

arXiv2026

Digital Engagement, Income Disparities, and Job Seeking in the United States since 2010

Surveys often record how frequently people use the internet without measuring the infrastructures, skills, and support systems that make digital participation possible. Using the U.S. National Longitudinal Survey of Youth 1997 cohort, we study how internet-use frequency relates to labor income, employment attachment, and job seeking after 2010. The main digital-engagement analysis uses the comparable 2011, 2013, and 2015 waves, with 2017 retained as later labor-market context. Across repeated cross sections, daily internet use consistently marks higher income and stronger employment attachment. Relative to daily use, less-than-daily use is associated with roughly 11 to 20 percent lower income, while nonuse is associated with about 18 to 21 percent lower income in 2011 and 2013. Respondents reporting no internet use are also 13 to 23 percentage points less likely to report full-year work. Job-search estimates reveal a distinct mechanism: active search is governed by employment status, search intensity, and application support, so a frequency item sorts respondents more sharply on durable labor-market attachment than on short-window search. Education accounts for a substantial share of the raw digital gradient, and pooled lagged-outcome and doubly robust transition estimates separate durable stratification from positive adoption margins. The results establish internet-use frequency as an informative behavioral marker of digitally mediated labor-market stratification and clarify why routine use should not be treated as a simple measure of digital access.

cs.CY↗

arXiv2026

Evaluating GNNs for Success Prediction in Artist Collaboration Networks

As the music industry becomes an increasingly collaborative effort, understanding the underlying structures of the artist network has become a focal point in cultural data analytics. This study expands on the previous analyses of the Italian and Danish networks by introducing a novel dataset of the Polish music scene. By utilizing methodologies used in the prior studies, this work enables a direct comparison between three distinct European music landscapes and allows to merged the created networks into one. Furthermore, this research introduces a framework to test the efficacy of Graph Neural Networks (GNNs) for artist popularity predictions based on the metadata and the position in the network. The statistical analysis revealed that the Polish and tri-national network exhibit similar properties and clustering behaviours, consistent with prior models. An evaluation of the predictive architectures reveals that while GNN models achieve a comparable F1-macro scores to the Multilayer Perceptron (MLP) in specific cases however the MLP remains a superior model regarding the success metric. The results suggest that internal node features - such as genre and label affiliation might carry more predictive capabilities than the topology of the network. The higher performance of the GNN models in the tri-national network might also suggests that the relational features become more informative when the network spans multiple linguistic and geographic boundaries, with the GNNs potentially capturing complex 'bridge' structures between the merged networks.

cs.SI↗