University of Victoria

UVic’s Research and Learning Repository
Not a member yet
    19614 research outputs found

    Enhancing fact-checking in large language models: Cost-effective claim verification through first-order logic reformulation

    No full text
    In the realm of Large Language Models (LLMs), the ability to accurately perform Fact Checking (FC) tasks, which involves verifying complex claims against challenging evidence from multiple sources, remains a crucial yet under-explored area. Our study presents a comprehensive benchmarking of various LLMs, including GPT-4, on this critical task. We utilize a modern, challenging dataset designed explicitly for fact-checking, HOVER, which comprises thousands of evidence-claim pairs covering diverse aspects of life, history, and entertainment. This dataset differs from common datasets that evaluate the reading comprehension capabilities of LLMs, which are primarily composed of sets of question-and-answer pairs. Our findings demonstrate that GPT-4 not only decisively surpasses the current state-of-the-art (SOTA) models in FC tasks but also shows that other, open-source, LLMs (e.g. Mixtral and Llama-3) exhibit close-to-SOTA performance out-of-the-box. This implies that simply presenting these models with the evidence text and claim allows them to infer the claim’s veracity effectively. We contrast this with the existing SOTA methods, which involve complex, multi-step solutions, including the use of multiple LLMs to verify claims – a process that necessitates continuous updates and local execution, making it less accessible for regular users. Furthermore, we explore the impact of claim formulation on the FC task’s effectiveness. By converting complex claims into first-order logic (FOL) and then back into natural language, we observe improved performance in some LLMs, particularly with more challenging dataset subsets. This method, although utilizing GPT-4 for the FOL breakdown, serves as a practical guideline for users: more formally structured claims yield more reliable responses.Graduat

    School-based educational programs, cultural identity, and mental health among urban Indigenous youth

    No full text
    This study aimed to examine the association between school-based cultural connectedness programs in the public school system and mental health outcomes in urban Indigenous youth. Past research shows that school-based programs increase cultural connectedness. Additionally, cultural connectedness predicts positive mental health in Indigenous populations. However, little research has examined the link between school-based programs and mental health, especially in urban Indigenous populations. The researchers volunteered and attended events at the Victoria Native Friendship Centre (VNFC) to foster a positive rapport with the urban Indigenous community. Participants aged 13-18 were recruited to participate in focus groups. Semi-structured interviews were conducted in the focus groups, and the data was analyzed using thematic analysis. Anticipated themes included community belonging, cultural identity and pride, overall mood, and life satisfaction. Preliminary results indicated two main themes 1) Lack of guidance (uncertainty about how to access programs and resources) and 2) Cultural pride (increased cultural pride and connectedness upon program participation). In summary, this study focuses on the mental health and well-being of urban Indigenous youth in relation to cultural connectedness programs in the public school system. Preliminary results show that programs increase cultural pride, and youth desire more guidance for accessing cultural connectedness programs.Jamie Cassels Undergraduate Research Awards (JCURA)UndergraduateReviewe

    An Exploratory Study of Data Physicalization Using Household Objects

    No full text
    We explore people’s perceptions and ideas regarding creating data physicalizations using household objects such as chairs, flower pots, and photo frames to enable data-driven self-reflection. By conducting a sketching-based qualitative study with 11 participants we identified styles of physical encoding participants used, strategies for creating physicalizations they employed, and techniques for constructing physicalizations they relied on. From the study results we contribute i) a bottom-up list of physical variables people might use for different data types, ii) a comparison between the theory about visual variables and the empirical use of physical variables, iii) an identification of the need for flexible taxonomies for physical representations, and iv) a discussion of the relationship between social pressure and the location of physical representations in the household.Graduat

    Seasonal variations in anthropogenic and natural particles induced by rising C02 levels

    Full text link
    Using an aerosol–climate coupled model, this paper has investigated the changes in distributions of anthropogenic and natural particles due to 4 × CO2-induced global warming, under the low emission scenario of Representative Concentration Pathway 4.5 (RCP4.5). Special attention is paid to the seasonal variations of aerosol size modes. With rising CO2 levels, surface warming, and changes in atmospheric circulations and hydrologic cycles are found during both summer (JJA) and winter (DJF). For anthropogenic particles, changes in fine anthropogenic particulate matter (PM2.5, particles with diameters smaller than 2.5 μm) decrease over high-latitude regions and increase over the tropics in both DJF and JJA. Global mean column concentrations of PM2.5 decrease by approximately 0.19 mg m−2, and concentrations of coarse anthropogenic particles (CPM, particles with diameters larger than 2.5 μm) increase by 0.005 mg m−2 in JJA. Changes in anthropogenic particles in DJF are similar to those in JJA, but the magnitudes of maximum regional changes are much smaller than those in JJA. The coarse anthropogenic particles (CPM, particles with diameters larger than 2.5 μm) increase over northern Africa and the Arabian Peninsula during JJA, whereas changes in anthropogenic CPM during DJF are minimal. During both JJA and DJF, changes in anthropogenic CPM are about two orders of magnitude smaller than those of anthropogenic PM2.5. Enhanced wet deposition by large-scale precipitation under rising CO2-induced surface warming is the critical factor affecting changes in anthropogenic particles. For natural particles, the distribution of change in the natural PM2.5 burden is similar to that of natural CPM, but much larger than natural CPM during each season. Both natural PM2.5 and CPM burdens increase over northern Africa and the Arabian Peninsula during JJA, but decrease over most of the continental regions during DJF. Changes in surface wind speed, divergence/convergence of surface wind, and precipitation are primary reasons for the variation of natural particles.This research was funded by the National Key R&D Program of China (2022YFC3701202) and the National Key Natural Science Foundation of China (42275039).FacultyReviewe

    Examining psychosocial predictors of cognition in parkinsonism across time

    No full text
    Background: Parkinsonism is a clinical syndrome that encompasses a spectrum of pathological difficulties including tremor, rigidity, abnormal slowness of movement, and postural instability, and affects between 108 to 212 per 100,000 older adults in North America per year. Rates of Parkinson’s disease (PD) related disability and death are also increasing more rapidly than any other neurological disorder. In addition to mobility disability, non-motor symptoms of parkinsonism such as changes in cognition occur as motor symptoms progress, as PD is associated with an increased risk of dementia. The cognitive reserve theory suggests that psychological and social factors may mitigate declines in cognition. Objectives: There is currently a paucity of research examining relationships between psychosocial factors and cognition in the context of parkinsonism synergistically and across time. To address these gaps, this study aimed to examine relationships between mobility disability, parkinsonism, psychological well-being, social participation, and cognition both cross-sectionally and longitudinally. Methodology: Cross-sectional path analyses included 1634 older adults from the Rush Memory and Aging Project (MAP); 1288 of which had sufficient follow-up assessments to conduct longitudinal growth curve analyses. Cognition was assessed across five domains: episodic memory, semantic memory, working memory, perceptual speed, and visuospatial ability. Results: Cross-sectional analyses revealed that psychological well-being partially mediated increases in mobility disability associated with decreases in performance in episodic memory, semantic memory, working memory, and perceptual speed performance. Relationships between mobility disability and episodic memory, semantic memory, and perceptual speed, respectively were also partially mediated by social participation. Psychological well-being and social participation did not mediate declines in cognition associated with increases in parkinsonism. Longitudinal growth curve analyses supported the notion that more severe mobility disability and parkinsonism is associated with worse psychological well-being and social participation across time. The indirect paths within the longitudinal models also suggest that psychological well-being and social participation may mediate relationships between parkinsonism and mobility disability, respectively, and cognitive performance across all domains. Conclusions: Results indicate that psychological well-being and social participation influence the effects of mobility disability and parkinsonism on cognition in differing manners. Findings indicate that psychological well-being interventions targeting reductions in anxiety, depression, and negative mood may be best suited for individuals with more severe mobility disability and may be most beneficial across the domains of episodic memory, semantic memory, working memory, and perceptual speed. Similarly, interventions targeting social participation may also be helpful to individuals with more severe mobility disability, and are most likely to impact episodic memory, semantic memory, and perceptual speed. For those with more severe mobility disability and parkinsonism, long-term psychological well-being and social participation-based interventions may be beneficial in targeting declines across all cognitive domains.Graduate2025-08-1

    Approaches to extracting patterns of service utilization for patients with complex conditions: graph community detection vs. natural language processing clustering

    Full text link
    Background: As patients interact with a healthcare service system, patterns of service utilization (PSUs) emerge. These PSUs are embedded in the sparse high-dimensional space of longitudinal cross-continuum health service encounter data. Once extracted, PSUs can provide quality assurance/quality improvement (QA/QI) efforts with the information required to optimize service system structures and functions. This may improve outcomes for complex patients with chronic diseases. Method: Working with longitudinal cross-continuum encounter data from a regional health service system, various pattern detection analyses were conducted, employing (1) graph community detection algorithms, (2) natural language processing (NLP) clustering, and (3) a hybrid NLP–graph method. Result: These approaches produced similar PSUs, as determined from a clinical perspective by clinical subject matter experts and service system operations experts. Conclusions: The similarity in the results provides validation for the methodologies. Moreover, the results stress the need to engage with clinical or service system operations experts, both in providing the taxonomies and ontologies of the service system, the cohort definitions, and determining the level of granularity that produces the most clinically meaningful results. Finally, the uniqueness of each approach provides an opportunity to take advantage of the various analytical capabilities that each approach brings, which will be further explored in our future research.FacultyReviewe

    Understanding the elemental variation of threespine stickleback along the Sooke River

    Full text link
    Fish play an important role in nutrient cycling in freshwater ecosystems. They convert a subset of dietary nutrients like phosphorus into waste products, which serve as fuel for primary producers like algae. The composition of waste products depends on what that organism requires for maintenance, growth, and reproduction. Different fish have different nutrient requirements. For example, if a fish has a relatively high proportion of bone, which is very phosphorus rich, you would expect that fish to use as much of the dietary phosphorus as possible and waste very little. Current studies compare the nutrients required and wasted by different species, but my research focuses on a single species: the threespine stickleback. Threespine stickleback display a wide range of variation in their phosphorus-rich bony armour. I hope to predict the proportions of nutrients in stickleback waste products based on the proportions in their bodies and diets. I collected stickleback from five sites along the Sooke River, measured their excretion rates in the field, and quantified the nutrient contents of their bodies and waste products. By understanding how the variation in stickleback bodies and diets relate to the contents of their wastes, we can better understand how animals affect their environments.Jamie Cassels Undergraduate Research Awards (JCURA)UndergraduateReviewe

    Telling My Auto EthnoGRAPHIC Story through My Drawings of Stó:lō and Sq'éwqel Archival History

    Full text link
    I used visual research methods such as drawing for inquiry and creating auto-ethnographic multimedia graphic stories about my experiences with Sq'éwqel Seabird Island First Nation and Stó:lō First Nation archival history (including archival audio recordings and photos), my reflections, and memories as a Sq'éwqel Seabird Island community member. Archie Charles’ oral stories from the Seabird Island Strength of Claim Project Database are central to this project. I explored drawing as a research method to establish a personal connection with archival history and community and create community-accessible resources for future education initiatives. Through this research, I found that I, the participant/researcher, became very curious and inspired by the inquiry, which forged a stronger connection to my community. In sharing my Indigenous graphic stories, I hope to find a way to educate on the diversity of Indigenous perspectives and engage readers in a way that connects them to Indigenous archival history and culture. These findings will be put forward to create unique Sq'éwqel Seabird Island First Nation educational programming.Graduat

    Education as strategy: Vocational reform and social mobility in neoliberal China

    No full text
    This thesis explores the evolution of vocational education policies in China, analyzing how market forces, neoliberal ideology, and the centralized control of the Chinese Communist Party (CCP) interact. I argue that while China aims to modernize and enhance its vocational education system to meet changing economic demands, these reforms serve dual purposes. On one hand, they aim to create a market-oriented education system that supports China’s broader economic objectives; on the other, they direct migrant and rural populations into vocational tracks, masking deeper socio-economic divides and using education reform as a tool for political stability rather than social equity. I situate China within the global trends of ‘vocationalism’ which promotes vocational education as a solution to economic and employment challenges. I then explore how vocational education reforms in China, articulated through the 1996 Vocational Education Law and its amendments, align with neoliberal trends that promote “suzhi” (quality) and “talent” to meet industrial demands. I also probe into China’s unique governance model, which combines market- driven reforms with authoritarian controls to shape its education reform. This governance strategy allows for a prioritization of national economic objectives over educational equality and perpetuates class distinctions by directing disadvantaged groups into vocational paths. I conclude that these reforms fail to uplift disadvantaged groups as claimed by state propaganda but only reinforce existing social stratifications. Empirical data for the thesis come from government reports, public media, secondary ethnographic literature and legal research.Graduat

    Mass spectrometry imaging methods for visualizing tumor heterogeneity

    No full text
    The authors would also like to thank Angela Jackson of the UVic Genome BC Proteomics Centre and Nathalie Costie, formally of Trev and Joyce Deeley Research Centre, BC Cancer, for preparing samples for mass spectrometry images used in Figure 2.Profiling spatial distributions of lipids, metabolites, and proteins in tumors can reveal unique cellular microenvironments and provide molecular evidence for cancer cell dysfunction and proliferation. Mass spectrometry imaging (MSI) is a label-free technique that can be used to map biomolecules in tumors in situ. Here, we discuss current progress in applying MSI to uncover molecular heterogeneity in tumors. First, the analytical strategies to profile small molecules and proteins are outlined, and current methods for multimodal imaging to maximize biological information are highlighted. Second, we present and summarize biological insights obtained by MSI of tumor tissue. Finally, we discuss important considerations for designing MSI experiments and several current analytical challenges.K.D.D acknowledges support from Vancouver Island University and Natural Sciences and Engineering Research Council of Canada (NSERC) (RGPIN-2022-03696). D.R.G was supported by the Metabolomics Innovation Centre (TMIC) from Genome Canada and Genome British Columbia, through the Genomics Technology Platform (GTP) (265MET), the Canadian Foundation for Innovation Major Sciences Initiative (CFI- MSI) program (35456), PacifiCan (Project No: 22591), Candian Foundation for Innovation John R. Evens Leaders Fund (Project No: 41124), the US Natioanl Institute of Health (NIH) and National Institute of Allergy and Infections Diseases (NIAID) award (1R01AI147314), and NSERC (RGPIN-2022-04433). J.J.L. thanks Canadian Institutes of Health Research (PJT-162279) for support. This work was also supported by the Terry Fox Research Institute New Frontiers Program Project and The Lotte and John Hecht Memorial Foundation (J.J.L, D.R.G, and K.D.D).FacultyReviewe

    11,522

    full texts

    19,614

    metadata records
    Updated in last 30 days.
    UVic’s Research and Learning Repository
    Access Repository Dashboard
    Do you manage Open Research Online? Become a CORE Member to access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard! 👇