324139 research outputs found
Sort by
Who uses general-purpose AI? A typology of ChatGPT early adopters
Get PDFThe world continues to debate the benefits, possibilities, failures, and risks of general-purpose artificial intelligence (AI) tools such as ChatGPT. With new tools and features being released at a high frequency, early adopters are eager to utilize them in various ways. Yet the priorities of these early adopters vary widely depending on their specific needs, capabilities, motivations, and usage patterns. In this study, we therefore explore how and why early adopters choose to use general-purpose AI tools. To do so, we draw on data from an online survey conducted among early ChatGPT users (n = 344) in April 2023, shortly after its public release. Based on this data, we identify six main dimensions determining the adoption of general-purpose AI tools: Utilitarian Value, Trust in AI, Convenience Value, Specific Job Utility, Perceived Social Presence, and Privacy Concerns. We then extend theories of innovation diffusion and technology adoption by empirically characterizing four early adopter archetypes: AI Enthusiasts, Naïve Pragmatists, Cautious Adopters, and Reserved Explorers. Distinguishing these archetypes helps devise interventions for effective AI adoption from a dual-use (i.e., functional-emotional vs. social-relational) and risk-reward trade-off (e.g., utility vs. privacy) perspective. In light of these insights, we offer practical implications for the market design and commercialization of general-purpose AI tools tailored to the priorities of each adopter archetype
High-dimensional adaptive optics techniques and applications
No full textAdaptive optics (AO) has traditionally been associated with feedback correction of phase aberrations. However, in practical imaging systems, performance is also significantly limited by coupled vectorial errors, including retardance and diattenuation. This thesis advances AO toward a high-dimensional framework that treats these error channels jointly. As a first step, this work develops and validates the vectorial extension of AO – pushing correction beyond scalar phase error – and establishes methods, metrics, and devices that pave the way to unified, highdimensional AO toolkit.First, a vectorial adaptive optics (V-AO) correction module was implemented in a wide-field Stokes polarimetric microscope. Using a pupil-intensity feedback strategy, the system successfully compensated for spatially varying polarisation aberrations, improving the vectorial uniformity across the field of view from 77.1% to 91.2% and achieving a vectorial precision of over 95% across various samples. Then the vectorial aberration correction concept is subsequently realised from the object’s perspective, hence termed as object-wise adaptive optics (O-AO). In this approach, a tunable arbitrary retarder array, realised by cascading polarisation modulators, is configured to form an inverse object for directly compensating retardance aberrations from the object’s perspective.Following the investigation of retardance correction, the capability of V-AO to correct diattenuation aberrations in non-depolarising media is investigated. The correction performance is quantitatively assessed using advanced optical skyrmionic beams as indicators. Finally, I study the correctability limit of V-AO in the presence of diattenuation in non-depolarising media. The correction performance is assessed using optical skyrmion beams as sensitive indicators, and the role of intensity loss and detector noise is made explicit. For example, when the diattenuation is strong (extinction ratio =3.36), some states fall below the detection threshold and become unrecoverable, whereas for a weaker diattenuation (=1.65) the Stokes field can be effectively restored.A summary of this thesis and an outlook on future research directions is provided, focusing on further advancing high-dimensional AO methodology and expanding its applications to a broader range of practical optical system
Arithmetic progressions at the Journal of the LMS
No full textWe discuss the papers P. Erdős and P. Turán, On some sequences of integers, J. London Math. Soc. (1) 11 (1936), 261–264 and K. F. Roth, On certain sets of integers, J. London Math. Soc. (1) 28 (1953), 104–109, both foundational papers in the study of arithmetic progressions in sets of integers, and their subsequent influence
RETRACTED ARTICLE: DynaGraph: interpretable dynamic graph learning for temporal electronic health records
No full textElectronic health records (EHRs) capture evolving physiological processes, yet most machine learning models impose static or sequential assumptions that flatten their temporal and relational complexity. We introduce DynaGraph, a dynamic and interpretable graph learning framework that constructs evolving spatio-temporal graphs from multivariate clinical time-series. Unlike previous methods, DynaGraph learns the structure of relationships between different clinical variables over time without predefined graphs, integrates sequential embeddings with contrastive graph augmentation, and incorporates a pseudo-attention mechanism to reveal temporally resolved risk factors. Trained end-to-end with a novel multi-loss objective that combines focal, structural, and contrastive components, DynaGraph addresses two pervasive challenges in real-world clinical modelling: class imbalance and temporal instability. We evaluated DynaGraph on four large-scale EHR datasets totalling 40,856 patients: MIMIC-III (17,279 ICU admissions), eICU (1433 cardiac ICU patients), HiRID-ICU (33,000 patients), and EHRSHOT (2378 primary care patients). DynaGraph consistently outperforms 14 state-of-the-art baselines, achieving 6-8% relative improvements in area under the precision-recall curve (AUPRC) and significant gains in sensitivity (12-22% over leading methods). Beyond predictive performance, DynaGraph offers time-specific interpretability aligned with clinical reasoning, providing gradient-based feature importance scores at 3-hour intervals that identify which physiological relationships drive predictions. This framework explicitly models temporal attribution of risk factors across patient trajectories in a millisecond inference time
What role does temporal synchrony play in mid-level audiovisual crossmodal correspondences?
No full textTemporal synchrony is widely recognized as one of the key factors facilitating the emergence of crossmodal correspondences and affecting their crossmodal effects. However, several issues regarding the definition of temporal synchrony and the mechanisms underlying its crossmodal effects remain open, depending on the specific experimental/perceptual context/stimuli used, as well as the influence of crossmodal congruency and structural (including isomorphic) crossmodal correspondences. In this review, we take a closer look at the literature that has been published in this area over recent decades in order to critically evaluate what is currently known concerning the crossmodal effects that are mediated by temporal synchrony. We focus especially on mid-level audiovisual crossmodal correspondences, defined as those that involve multi-element, or dynamic, auditory and visual stimuli. We examine the different experimental methodologies used and their limitations as well as the theoretical frameworks that have been proposed to account for the viewer’s impression of (and the meaning/affect that is associated with) such experimental audiovisual displays, including those that are based on the ‘Congruency-Associationist Model’, Gestalt perceptual grouping, as well as the phenomenon of multisensory emergence. Finally, we outline several directions for future research on temporal synchrony in the context of audiovisual crossmodal correspondences
Constraints on the Correlation of IceCube Neutrinos with a Tracer of Nearby Large-scale Structure
No full textThe IceCube Neutrino Observatory has observed extragalactic astrophysical neutrinos with an apparently isotropic distribution. Only a small fraction of the observed astrophysical neutrinos can be explained by known sources. Neutrino production is thought to occur in energetic environments that are ultimately powered by the gravitational collapse of dense regions of the large-scale mass distribution in the universe. Whatever their identity, neutrino sources likely trace this large-scale mass distribution. The clustering of neutrinos with a tracer of the large-scale structure may provide insight into the distribution of neutrino sources with respect to redshift and the identity of neutrino sources. We implement a two-point angular cross correlation of the Northern sky track events with an infrared galaxy catalog derived from the Wide-field Infrared Survey Explorer (WISE) and Two Micron All Sky Survey (2MASS) source catalogs, which trace the nearby large-scale structure. No statistically significant correlation is found between the neutrinos and this infrared galaxy catalog. We find that ≤54% of the diffuse muon neutrino flux can be attributed to sources correlated with the galaxy catalog with 90% confidence. Additionally, when assuming that the neutrino source comoving density evolves following a power law in redshift, dNs/dV ∝ (1 + z)k, we find that sources with negative evolution, in particular k < −1.75, are disfavored at the 90% confidence level
Contextual analysis of antibiotic prescribing practices for upper respiratory infections and diarrhoea in rural Punjab, Pakistan
No full textIntroduction: Antimicrobial resistance (AMR) poses a major global public health threat, with deaths projected to reach 10 million annually by 2050, disproportionately affecting LMICs. Almost 80% of prescriptions at primary healthcare facilities contain at least one antibiotic. This study explores the existing care context at rural health centres (RHC) in Punjab to inform the development of an intervention package aimed at reducing inappropriate antibiotic prescriptions for two most common infections in the outpatient department of RHCs i.e., upper respiratory tract infections (URTI) and diarrhoea. Methods: This study employed a care-cascade framework, selected for its suitability in resource-constrained settings and its applicability in contexts with limited evidence on AMR and stewardship practices. The mixed-methods design included: literature review; facility review (n = 3); key informant interviews with healthcare providers (n = 6) and patients (n = 6); and focus group discussions with health managers (n = 1), healthcare providers (n = 2), patients (n = 2), patients with experience of using mobile healthcare applications (n = 2) and community significants (n = 1). The data analysed at each step informed the next step and was populated in the pre-defined care tasks considered as themes. Results: Using the care-cascade framework, four major care tasks were identified in URTI and diarrhoea management at RHCs: (1) patient consultation and laboratory testing, (2) drug prescription, (3) drug dispensing, and (4) patient education and counselling. At the RHCs there were no standard guidelines or diagnostic protocols being followed, inadequate lab facilities, and very brief patient consultations. Antibiotic prescription was largely empirical, driven by doctor’s clinical acumen and patient demands. Drug dispensing practices were inconsistent, involving dispensing partial courses without counselling. Patients were insufficiently counselled on disease management and antibiotic use. Stakeholders made context specific recommendations for developing clinical protocols, patient counselling, and training both doctors and dispensers. They also informed the design and content of the digital tools for patient engagement. Conclusion: This study underscores the complexity of antibiotic stewardship in rural health systems. A care-cascade framework allowed us to unpack these multi-layered challenges and identify leverage points for improvement. Future implementation strategies must embed these insights into program design, ensuring that stewardship is not merely a clinical responsibility but a health system imperative
Adapting everyday activities to summer heatwaves: a multi-country analysis of mobile phone location data
No full textIn the 21st century, record-breaking summer heatwaves have had devastating impacts on people’s health, well-being, and livelihoods. In light of this urgent threat, government institutions across the globe are developing guidelines and planned interventions to increase resilience to heatwaves. These measures require an understanding of how people adapt to extreme heat within the constraints of daily life. Existing studies have used large-scale mobility data to characterize heatwave adaptation at population levels, but skew towards cities and regions in high-income countries, have diverse methodologies that limit generalizability to other contexts, and focus on ‘activity level’ changes without discerning which activities are being altered. Addressing these gaps, this study combines climate reanalysis, mobile phone location, socio-demographic, and physical-geographical data across Brazil, France, India, Nigeria, Turkey, the USA, and China during 2022/2023 summer heatwaves. Google Community Mobility Reports data is used in multivariate multi-level modelling for the first six countries to examine daily activity changes during heatwaves (home, work, transit, grocery/pharmacy, retail/recreation, parks). In China, Baidu data on activity levels is analysed in a complementary multi-level model. The results show a widespread tendency to withdraw into homes but also highlight unequal substitutions of activities, and—under some circumstances—visits to potentially cooler locations away from home. This study highlights the global nature of heatwave adaptation and the value of considering adaptation within the context of people’s everyday lives
Development and validation of a health education module for parents of schoolchildren with overweight and obesity in the UAE
No full textBackground: Childhood overweight and obesity continue to rise globally and in the United Arab Emirates (UAE), underscoring the need for accessible, evidence-based parental education tools. Objective: This study aims to develop and validate a health educational module (HEM) tailored for parents of schoolchildren with overweight and obesity in the UAE. Methods: The study was conducted in two phases: module development, and content and face validation. The module development involved an extensive review of national and international guidelines and previous research, followed by the design of infographic-based educational messages. Content and Face Validity were performed by six nutrition experts and 16 parents, respectively, using the Content Validity Index (CVI) and Face Validity Index (FVI). Results: The final HEM consisted of 25 infographic messages covering dietary intake, physical activity, behavior change, and family engagement. The module achieved excellent content validity (S-CVI/Ave = 0.97) and face validity (S-FVI/Ave = 0.99). Participant feedback resulted in language refinement, improved illustrations, and removal of outdated elements such as the Food Pyramid. Conclusion: The validated HEM is culturally relevant, parent-friendly, and scientifically grounded. It offers a structured, high-quality educational tool suitable for use in research, school-based initiatives, and public-health programs aimed at reducing childhood obesity in the UAE
Diels-Alder reaction affords circumpyrene tetracarboxydiimide with excited state intramolecular charge transfer character
No full textLarge polycyclic aromatic hydrocarbon (PAH) imides are promising candidates for optoelectronic applications in view of their narrow optical gaps and/or excited state charge transfer character. Diels–Alder (D–A) reactions of PAHs at bay regions can enable simultaneous extension of the aromatic structure and introduction of the imide moieties, but the actual use of this strategy has been limited. Herein, we demonstrate the D–A cycloaddition of dibenzo[hi, st]ovalene, one of the largest PAHs functioning as bisdiene, with maleimides to afford circumpyrene tetracarboxydiimides. Notably, efforts to optimize the yield of di-adduct revealed that the fully aromatized mono-adduct is inert toward further D–A reaction, and density functional theory (DFT) calculations instead indicated a partially dehydrogenated mono-adduct as the key intermediate enabling the second cycloaddition. The resulting product represents a rare example of PAH diimide featuring an acceptor-donor-acceptor type structure. Detailed spectroscopic and theoretical studies, including transient absorption and two-dimensional electronic spectroscopy, revealed the emergence of a bright intramolecular charge transfer state that could be directly excited to demonstrate distinct photophysical dynamics. These findings provide deep mechanistic insights into the D–A reactivity of large PAHs and underscores the potential of such PAH imides for advanced optoelectronic and photonic applications