Search Results (502)

Large language models (LLMs) have advanced geospatial artificial intelligence; however, geospatial knowledge-base question answering (GeoKBQA) remains underdeveloped. Prior systems have relied on handcrafted rules and have omitted the splitting of datasets into training, validation, and test sets, thereby hindering fair evaluation. To address these gaps, we propose a prompt-based multi-agent LLM framework (based on GPT-4o) that translates natural-language questions into executable GeoSPARQL. The architecture comprises an intent analyzer, multi-grained retrievers that ground concepts and properties in the OSM tagging schema and map geospatial relations to the GeoSPARQL/OGC operator inventory, an operator-aware intermediate representation aligned with SPARQL/GeoSPARQL 1.1, and a query generator. Our approach was evaluated on the GeoKBQA test set using 20 few-shot exemplars per agent. It achieved 85.49 EM (GPT-4o) with less supervision than fine-tuned baselines trained on 3574 instances and substantially outperformed a single-agent GPT-4o prompt. Additionally, we evaluated GPT-4o-mini, which achieved 66.74 EM in a multi-agent configuration versus 47.10 EM with a single agent. The observations showed that the multi-agent gain was higher for the larger model. Our results indicate that, beyond scale, the framework’s structure is important; thus, principled agentic decomposition yields a sample-efficient, execution-faithful path beyond template-centric GeoKBQA under a fair, hold-out evaluation protocol. Full article

Whilst there has been substantial attention to care-experienced parents’ needs and experiences in the academic literature internationally, understandings of nascent services, their characteristics and implementation processes are more limited. With an overarching socioecological resilience systems framing and drawing on an innovation perspective, we aim to develop understanding of how to design and develop innovations to enable care-experienced parents’ successing. We conducted a narrative literature review that included 33 sources published internationally between 2017 and 2025. We conducted thematic analysis to identify adversities experienced by and innovations developed for care-experienced parents. We authenticated the themes in a workshop with members of the practice community and developed frameworks to represent the themes. Findings are represented in three sections. First, we consider parental needs, with an overview of adversities experienced by care-experienced parents together with individual protective factors and required service responses, framed by psychological, social and structural domains. Second, drawing upon such understanding, we consider intervention design, with a focus on exemplar innovations and the characteristics that are represented by five service delivery models: therapeutic; social; partnership; advocacy; and co-production. Third, with a need to ensure that service intervention is effective, we examine the process of developing service innovations and consider five dynamic ingredients that enable implementation success: shared leadership; receptivity of context; co-production; learning and adaption; and outcome measurement. Our review contributes new understanding to inform processes of designing and implementing innovations to enable care-experienced parents’ successing. We offer a framework that represents a starting point towards enabling care-experienced parents’ successing that can be applied in policy and practice, although more research is needed. Full article

Recent advances in large language models (LLMs) have driven substantial progress in knowledge base question answering (KBQA), particularly under few-shot settings. However, symbolic program generation remains challenging due to its strict structural constraints and high sensitivity to generation errors. Existing few-shot methods often rely on multi-turn strategies, such as rule-based step-by-step reasoning or iterative self-correction, which introduce additional latency and exacerbate error propagation. We present CBR², a case-based reasoning framework with dual retrieval guidance for single-pass symbolic program generation. Instead of generating programs interactively, CBR² constructs a unified structure-aware prompt that integrates two complementary types of retrieval: (1) structured knowledge from ontologies and factual triples, and (2) reasoning exemplars retrieved via semantic and function-level similarity. A lightweight similarity model is trained to retrieve structurally aligned programs, enabling effective transfer of abstract reasoning patterns. Experiments on KQA Pro and MetaQA demonstrate that CBR² achieves significant improvements in both accuracy and syntactic robustness. Specifically on KQA Pro, it boosts Hits@1 from 72.70% to 82.13% and reduces syntax errors by 25%, surpassing the previous few-shot state-of-the-art. Full article

Effective risk communication designed for risk management increases concern and motivates action by providing guidance and specific actions that can be taken. When exposures to environmental contaminants or stressors are ubiquitous or pollutant sources are not easily controlled, also decreasing sustainability, risk communication is focused on actions for risk reduction and avoidance. Three recommended practices (use of virtual exemplars, narrative, and social media) are discussed as tactics and platforms to inform public beliefs and behaviors and to encourage adoption of long-term planning goals that avoid the consequences of future risks. These risk communication strategies appeal broadly to lay audiences, are not limited to scientists and science-trained risk communicators, and are consistent with the US EPA’s SALT Framework, a research-based approach with recommended practices to guide risk communication. The overall strategy is to make risk communication more effective by using approaches that are dynamic, interactive, engaging, and relatable. Full article

The purpose of this study was to engage high school science teachers as co-design partners in refining and extending instructional frameworks to support multiple-document reading and writing in science classrooms. Using a participatory mixed-methods design, the project adapted the InSPECT framework for secondary science, developed professional development (PD) materials to introduce the framework, and explored the role of generative artificial intelligence (AI) in lesson planning. Five virtual focus group sessions guided the co-design of PD activities, followed by a pilot implementation in one biology classroom. Data included focus group and interview transcripts, surveys, and student work artifacts. Analyses examined teachers’ perceptions of PD features, framework usability, and student engagement. Teachers valued PD that was practical, relevant, and feasible within classroom constraints and described the frameworks as clear, stepwise structures that supported lesson design and literacy integration. Student work showed that paraphrasing was an accessible entry point, while bridging, elaboration, and source evaluation required additional modeling. Teachers viewed generative AI as a promising planning aid but expressed concerns about accuracy and ethics. Findings informed revisions emphasizing discipline-specific exemplars, scaffolds for higher-order strategies, and AI-literacy modules, illustrating how participatory design can yield feasible, teacher-centered PD. Full article

The environmental crisis precipitated by climate change has accelerated the urgency of urban green and low-carbon transformation. In 2024, China’s Action Plan for the National Standardization Development Outline (2024–2025) stipulated requirements for continuously improving the standard system for carbon peaking and carbon neutrality in public institutions. As key venues for knowledge innovation and energy consumption, the low-carbon transformation of higher education institutions holds significant importance for China’s achievement of its dual carbon goals. However, China lacks a systematic evaluation framework specifically designed for university campus carbon emissions. Existing green campus assessment standards often suffer from inadequate indicator adaptability, a lack of update mechanisms, and limited coverage. The STARS sustainability assessment system, widely adopted in North America, offers valuable reference points for establishing campus carbon emissions evaluation frameworks due to its features of indicator adaptability, dynamic update mechanisms, and comprehensive evaluation dimensions. This paper conducts an exploratory comparative case study of Princeton University (USA) and Tianjin University (China)—two leading research-intensive institutions—within the STARS 2.2 framework. It systematically analyses their divergent approaches to carbon management and evaluation, not as representatives of their respective continents, but as exemplars of how advanced universities operationalize low-carbon transitions. Based on this analysis and a review of domestic Chinese standards, it proposes a development pathway for China’s university campus carbon emissions evaluation system: (1) Establish a differentiated indicator system combining ‘universal fundamentals with discipline-specific types’ to enhance adaptability to campus characteristics; (2) Establish a mechanism for periodic version updates to the evaluation standard itself, ensuring alignment with evolving national carbon goals and technological advancements; (3) Develop a comprehensive and transparent carbon accounting framework that integrates direct emissions, purchased energy, and indirect sources. This research provides theoretical foundations and methodological support for institutional development and practical optimization in carbon emissions evaluation within Chinese higher education institutions. Full article

High-grade gliomas—particularly glioblastoma (GBM)—remain refractory to standard-of-care surgery followed by chemoradiation, with a median overall survival of ~15 months. Oncolytic viruses (OVs), which selectively infect and lyse tumor cells while engaging antitumor immunity, offer a mechanistically distinct therapeutic modality. This review synthesizes clinical progress of OVs in GBM, with emphasis on oncolytic herpes simplex virus (oHSV) and coverage of other vectors (adenovirus, reovirus, Newcastle disease virus, vaccinia virus) across phase I–III trials, focusing on efficacy and safety. Key observations include the encouraging clinical trajectory of oHSV exemplars—T-VEC (approved for melanoma) and G47Δ (approved in Japan for recurrent GBM)—the multi-center exploration of the adenovirus DNX-2401 combined with programmed death-1 (PD-1) blockade, and the early-stage status of reovirus (pelareorep) and Newcastle disease virus programs. Emerging evidence indicates that oHSV therapy augments immune infiltration within the tumor microenvironment and alleviates immunosuppression, with synergy when combined with chemotherapy or immune checkpoint inhibitors. Persistent challenges include GBM’s inherently immunosuppressive milieu, limitations imposed by the blood–brain barrier, intrapatient viral delivery and biodistribution, and concerns about viral shedding. Future directions encompass programmable vector design, optimization of systemic delivery, biomarker-guided patient selection, and rational combination immunotherapy. Collectively, OVs represent a promising immunotherapeutic strategy in GBM; further gains will hinge on vector engineering and precision combinations to translate mechanistic promise into durable clinical benefit. Full article

Background: Women’s health has historically served as an incubator for major medical innovations yet often faces relative neglect in sustained funding and implementation. The rise of artificial intelligence (AI) and machine learning (ML) presents both opportunities and risks for diagnostics in obstetrics and gynecology (OB/GYN). Methods: A narrative review (January 2018–August 2025) integrating peer-reviewed literature and clinical exemplars was conducted. OB/GYN relevance, clinical validation/scale, near-term outcome impact, and domain diversity were prioritized in selection. Results: We highlight ten promising AI applications across imaging, laboratory diagnostics, patient monitoring/digital biomarkers, and decision support, including AI-enhanced fetal ultrasound, cervical screening, preeclampsia prediction with cell-free RNA, noninvasive endometriosis testing, remote maternal–fetal monitoring, and reinforcement-learning decision support in gynecologic oncology. Conclusions: AI shows transformative potential for women’s health diagnostics but requires attention to bias, privacy, regulatory evolution, reimbursement, and workflow integration. Equity-focused development and diverse datasets are essential to ensure benefits accrue broadly. Full article

To better address the challenge of corneal ulcer healing, with already available standard agents, and those recently introduced, such as stable gastric pentadecapeptide BPC 157, we introduced a novel conceptual framework—the “triad” of corneal ulcer healing↔corneal neovascularization↔intraocular pressure—and extended it to avascular tissues such as tendon. Within this framework, cytoprotection serves as the unifying principle, underscoring that therapeutic effects are not isolated but interconnected. Preclinical studies with BPC 157 therapy, as a cytoprotection agent, illustrate this integration. BPC 157 rapidly normalizes elevated intraocular pressure in glaucomatous rats, preserves retinal integrity, restores pupil function, maintains corneal transparency during ulcer or abrasion healing, and counteracts both corneal neovascularization and dry eye. In parallel, its consistent efficacy in tendon injury models highlights a cytoprotective specificity across avascular tissues. The cornea’s “angiogenic privilege,” preserved during healing and tendon recovery together, provides strong proof of concept. Furthermore, mapping standard therapeutic agents used for corneal ulcers, neovascularization, or glaucoma onto this triad, and linking them with tendon healing, reveals both shared pathways and inconsistencies across existing drug classes. Analyzed were the ascorbate, fibronectin, hyaluronic acid, metalloproteinase inhibitors, EGF, FGF, NGF, insulin, and IGF-1 (corneal ulcer healing), the antiangiogenic agents (endostatin, PAI-1, PEDF, angiostatin, TSP-1, TSP-2, IFN-α), corticosteroids, NSAIDs, cyclosporine A, anti-VEGF drops (treatment of corneal neovascularization), and alpha 2-agonists, beta-blockers, carboanhydrase inhibitors, muscarinic agonists, Rho-kinase inhibitors, and prostaglandin analogs (glaucoma). Taken together, these findings advance cytoprotection as a unifying therapeutic paradigm, with BPC 157 emerging as its first exemplar, and encourage further translational research toward clinical application. Full article

The importation of prefabricated buildings into colonies was a prevalent practice during the British colonial expansionist venture. However, in post-colonial Nigeria today, many of these prefabricated houses have either been largely modified or have vanished without architectural or written records. This undocumented disappearance poses a challenge to the development of architectural restoration proposals for the remaining few, especially with the authenticity of materials, as well as their morphology, configuration, use, and function being heavily contested. Among the remaining few that have undergone layers of modifications and are on the verge of total collapse is the Egbo Egbo Bassey House, imported and built in Old Calabar between 1883 and 1886 and declared a National Monument of Nigeria in 1959. Given the dearth of architectural and historical data, this paper aims to reconstruct its architectural morphology, chronological modification, and historical uses and functions, with the view of developing an evidence-based architectural restoration proposal for its adaptive reuse. The data was collected through semi-structured interviews (n = 16), archival research at the National Museum (archival file ID: TF128/C.25/A and TF120/C.20/A), and a measured architectural survey, which was performed using laser tapes and laser rangefinders. Annotated building images were captured using a Canon 5D Mark III and a DJI Marvic 3 drone. Comparative analysis with two other exemplars of prefabricated houses in the region was also conducted to consolidate oral, archival, and field data. Three architectural modification stages, namely 1886, 1959, and 2012, were determined for the analytical framework. Architectural outputs include measured 2-dimensional drawings (scale 1:50) and 3-dimensional models for the three historical stages. The accuracy of each model was ensured through methodical triangulation and confidence rubric ratings. The result of this paper provides a replicable inquiry methodology, which can be used to develop an evidence-based workflow for developing a restoration proposal for architectural heritage in contexts where architectural and historical data are not available or contested. As a limitation, this research does not include an analysis of wood typology, structural testing, and statistical analysis of material. Full article

A number of collector’s cabinets known as pronk or luxury dollhouses were formed in the seventeenth and eighteenth centuries by women in the Netherlands. The present study examines the dollhouse cabinets as exemplars of material culture collections assembled by female collectors. Primary sources give outsized attention to the materiality of these structures, often noting types of substance, quality, and craft. Despite what appears to be a straightforward transcription of the domestic world in miniature, the dollhouses are a multifaceted intersection of authentic materials as well as clever imitations or substitutions. The dollhouse collections are themselves predicated on the notion of reproduction as they replicate the home in small scale. Documents from the period provide a rich source from which to probe the meanings invested in the materiality of these dollhouses as sources of wonder. Economic theory from the period sheds new light on the dollhouses as forums for imitation and novelty, concepts that inform the innovative nature of these collections as it intertwined with issues of multiples and miniaturization. Full article

The proposed Adaptive RTR with Quarantine (ARQ) method integrates, within a single evolutionary scheme for continuous optimization, three mature ideas of pbest differential evolution with an archive, success-history parameter adaptation, and restricted tournament replacement (RTR) and extends them with a novel outlier quarantine mechanism. At the heart of ARQ is a combination of the following complementary mechanisms: (1) an event-driven outlier-quarantine loop that triggers on robustly detected tail behavior, (2) a robust center from the best half of the population to which quarantined candidates are gently repaired under feasibility projections, (3) local RTR-based replacement that preserves spatial diversity and avoids premature collapse, (4) archive-guided trial generation that blends current and archived differences while steering toward strong exemplars, and (5) success-history adaptation that self-regulates search from recent successes and reduces manual fine-tuning. Together, these parts sustain focused progress while periodically renewing diversity. Search pressure remains focused yet diversity is steadily replenished through micro-restarts when progress stalls, producing smooth and reliable improvement on noisy or rugged landscapes. In a comprehensive benchmark campaign spanning separable, ill-conditioned, multimodal, hybrid, and composition problems, ARQ was compared against leading state-of-the-art baselines, including top entrants and winners from CEC competitions under identical evaluation budgets and rigorous protocols. Across these settings, ARQ delivered competitive peak results while maintaining favorable average behaviour, thereby narrowing the gap between best and typical outcomes. Overall, this design positions ARQ as a robust choice for practical performance and consistency, providing a dependable tool that can meaningfully strengthen the methodological repertoire of the research community. Full article

An experiment was conducted to examine whether knowledge of word meanings enables learners to infer the meanings of related words, and whether such transfer is based on memory for related exemplars or for abstract knowledge. Participants completed a word root learning task in which they learned definitions of several English words derived from a shared root (e.g., ambler, noctambulant). At an immediate test, they were assessed on definitions of studied words, new unstudied derivatives (e.g., ambulate), and word roots (e.g., ambul). A multiple regression analysis showed that accuracy on word roots, but not on studied words, predicted performance on new derivatives. These results suggest that transfer of learning was based primarily on more abstract knowledge of word root meanings rather than on memory for specific words. These findings provide novel evidence that learners can apply root-based knowledge to new word forms, and are consistent with theories proposing that transfer is supported by abstract representations. Full article

Figurative language (metaphor, idiom, irony/sarcasm) is central to pragmatic communication but is frequently challenging for children and adolescents with autism spectrum disorder (ASD). A scoping review was conducted to map pedagogical and clinical interventions that target figurative-language skills in school-age learners with ASD and to summarize reported outcomes. Following a PCC (Population–Concept–Context) framework and PRISMA-ScR reporting, systematic searches were performed in ERIC and Google Scholar (2010–2025). Eligibility required an ASD sample (ages 5–18), an intervention explicitly addressing figurative-language comprehension, and empirical outcome data from educational or related practice settings. Seven studies met inclusion criteria: five targeting metaphors, one targeting idioms, and one targeting sarcasm/irony. Interventions were predominantly delivered one-to-one or in small groups and emphasized structured, explicit instruction with visual scaffolds and stepwise prompting. Across studies, participants demonstrated clear gains on trained items. Generalization beyond trained material was most often observed for metaphor and sarcasm interventions, particularly when instruction highlighted underlying semantic relations or cue-based pragmatic signals; by contrast, the idiom program yielded item-specific learning with minimal near-term transfer. Limited follow-up data suggested short-term maintenance where assessed. Reported variability across individuals was substantial, underscoring the influence of underlying structural-language skills and social-pragmatic demands. Overall, the evidence indicates that figurative-language skills in ASD are amenable to targeted intervention; effective programs tend to combine explicit teaching, visual supports, multiple exemplars, and planned generalization opportunities. Given small samples and methodological heterogeneity, further classroom-based trials with longer follow-up and detailed learner profiles are needed. The findings support integrating figurative-language goals within individualized education and speech-language therapy plans, while aligning instructional complexity with each learner’s linguistic and pragmatic profile. Full article

Background: Germline genetic variants impacting splicing are a frequent cause of disease. The clinical interpretation of such variants is challenging for many reasons including the immense complexity of splicing mechanisms. While recent advances in splicing algorithms have improved the accuracy of splice prediction, predicting the nature and abundance of aberrant splicing remains challenging. As RNA testing becomes more mainstream in the clinical diagnostic setting, the complexities of interpretation are coming to light. Methods: Data from patients undergoing concurrent DNA and RNA testing were retrospectively reviewed for unusual splicing impacts to underscore some of these complexities and serve as exemplars in how to avoid pitfalls in the interpretation of sequence variants. Results: Seven rare variants with unusual splicing impacts are presented: a variant at a consensus donor nucleotide position lacking a splice impact (NF1 c.888+2T>C); a mid-exonic missense variant creating a novel donor site and a cryptic acceptor site resulting in pseudo-intronization (BRIP1 c.727A<G p.Ile243Val); one variant creating a spliceosome switch from U12 to U2 (LZTR1 c.2232G>A p.Ala744Ala); two variants that would be expected to result in nonsense-mediated-mRNA-decay triggering splicing impacts that obviated nonsense-mediated-decay (APC c.1042C>T p.Arg348Ter and BRCA2 c.6762del; c.6816_6841+1534del); and two variants causing splicing impacts through pyrimidine tract optimization (NF1 c.5750-184_5750-178dup and ATM c.3480G>T p.Val1160Val). Conclusions: Paired DNA and RNA testing revealed unexpected splice events altering variant interpretation, expanding our knowledge of clinically important splicing mechanisms and highlighting the benefit of RNA testing. Full article