Search for: All records

To recognize and mitigate the harms of generative AI systems, it is crucial to consider who is represented in the outputs of generative AI systems and how people are represented. A critical gap emerges when naively improving who is represented, as this does not imply bias mitigation efforts have been applied to address how people are represented. We critically examined this by investigating gender representation in occupation across state-of-the-art large language models. We first show evidence suggesting that over time there have been interventions to models altering the resulting gender distribution, and we find that women are more represented than men when models are prompted to generate biographies or personas. We then demonstrate that representational biases persist in how different genders are represented by examining statistically significant word differences across genders. This results in a proliferation of representational harms, stereotypes, and neoliberalism ideals that, despite existing interventions to increase female representation, reinforce existing systems of oppression. 

Introduction: Coronavirus disease 2019 (COVID-19) has had a profound impact globally, causing the death of millions of people and deeply affecting socio-psychological, human health, and economic systems, with some nations bearing a disproportionate burden. Despite obesity having been established as one of the major risk factors of COVID-19 severity and other degenerative diseases, the effects that dietary pattern plays in COVID-19 outcomes remain poorly understood. The goal of this study is to look into the connection between eating habits, the number of non-obese and obese people, and COVID-19 outcomes in countries with populations exhibiting normal Body Mass Index (BMI), which is an indicator of obesity. Methods: The analysis includes data from 170 countries. From the 170 countries, we focused on 53 nations where the average, BMI falls within the normal range (18.5 to 24.9). A subset of 20 nations was selected for a more detailed examination, comprising 10 nations with the lowest BMI values within the normal range (18.5-19.8) and 10 nations with the highest BMI values within the normal range (23.5-24.9). We used Artificial Intelligence (AI) and Machine Learning (ML) applications to evaluate key metrics, including dietary patterns (sugar and vegetable intake), obesity prevalence, incidence rate, mortality rate, and Case Fatality Rate (CFR). Results: The results demonstrate a significant correlation between higher obesity prevalence and increased COVID-19 severity, evidenced by elevated incidence, mortality, and CFRs in countries like North Macedonia and Italy. In contrast, nations such as Iceland and New Zealand with well-established healthcare systems revealed low mortality rates and case fatality rates despite variations in dietary habits. The study also revealed that vegetable consumption appears to provide a slight to significant protective effect, suggesting that dietary patterns alone do not consistently predict COVID-19 Outcomes. Conclusion: Data generated from this study showed the crucial role of healthcare infrastructure along with the testing capacity and data reporting in influencing the success of pandemic responses. It also highlights the need to integrate public health strategies, which focus on obesity management and improvement of healthcare preparedness. In addition, AI-driven predictive modeling offers valuable insights that may guide pandemic response efforts in the future, thereby enhancing global health crisis management and mitigating the impact of future health emergencies. 

Acer(Sapindaceae) is a major genus of broadleaf trees dominating deciduous forests in the Northern Hemisphere, with Asia exhibiting the highest species diversity. Many economically importantAcerspecies are cultivated for ornamental or timber purposes.Acerpowdery mildew, caused by fungi in the tribeCystotheceae, poses significant global economic and ecological threats. The pathogenicity spectrum remains unclear due to taxonomic uncertainties in its primary causal genera,SawadaeaandTakamatsuella. This study presents a comprehensive phylogenetic-taxonomic analysis of the two genera across East Asia, Europe, and North America. Using 75 ITS and 58 28S rDNA newly obtained sequences, we resolved 12Sawadaeaspecies and oneTakamatsuellaspecies into nine monophyletic clades, revealing marked cryptic diversity (three new species:S. acerina,S. aceris-arguti,S. taii) and two paraphyletic groups (S. bifida/S. negundinis). Taxonomic revisions include:S. bicornissplit into twoformae(f. bicornisandf. polyphaga f. nov.) with distinct host preferences;S. tulasnei(sensu stricto) restricted to Europe/North America, invalidating previous Asian records;S. nankinensisandS. koelreuteriaeform two basal lineages. Phylogenetic positioning confirmedTakamatsuellaas a distinct genus sister toSawadaea, supported by an ITS1 26 bp deletion. Host specificity analysis revealed narrow host ranges (primarilyAcer) with two evolutionary host expansions toKoelreuteria,Aesculus, andLiquidambar. This study also newly describes the asexual morphs of four species (S. aesculi,S. bifida,S. bomiensisandS. kovaliana) and establishes a molecular framework for disease management through clarified phylogeny and taxonomy. Our findings provide critical insights into fungal evolution, host-pathogen interactions, and strategies for mitigating powdery mildew impacts in forest ecosystems. 

Self-supervised training methods for transformers have demonstrated remarkable performance across various domains. Previous transformer-based models, such as masked autoencoders (MAE), typically utilize a single normalization layer for both the class token [CLS] and the tokens. We propose in this paper a new yet simple normalization method that separately normalizes embedding vectors respectively corresponding to normal tokens and the [CLS] token, in order to better capture their distinct characteristics and enhance downstream task performance. Our empirical study shows that the [CLS] embeddings learned with our separate normalization layer better encode the global contextual information and are distributed more uniformly in its anisotropic space. When the conventional normalization layer is replaced with a separate normalization layer, we observe an average 2.7% performance improvement in learning tasks from the image, natural language, and graph domains. 

The MicroBooNE experiment is an 85 tonne active mass liquid argon time projection chamber neutrino detector exposed to the on-axis Booster Neutrino Beam at Fermilab. One of MicroBooNE’s physics goals is the precise measurement of neutrino interactions on argon in the 1 GeV energy regime. Building on the capabilities of the MicroBooNE detector, this analysis identifies $K^{+}$mesons, a key signature for the study of strange particle production in neutrino interactions. This measurement is furthermore valuable for background estimation for future nucleon decay searches and for improved reconstruction and particle identification capabilities in experiments such as the Deep Underground Neutrino Experiment. In this Letter, we present the first-ever measurement of a flux-integrated cross section for charged-current muon neutrino induced $K^{+}$production on argon nuclei, determined to be $7.93\pm 3.22\left(\mathrm{stat}\right)\pm 2.83\left(\mathrm{syst}\right)\times {10}^{-42}{\mathrm{cm}}^2/\mathrm{nucleon}$based on an analysis of $6.88\times {10}^{20}$protons on target. This result was found to be consistent with model predictions from different neutrino event generators within the reported uncertainties. 

Abstract The existence of three distinct neutrino flavours,ν_e,ν_μandν_τ, is a central tenet of the Standard Model of particle physics^1,2. Quantum-mechanical interference can allow a neutrino of one initial flavour to be detected sometime later as a different flavour, a process called neutrino oscillation. Several anomalous observations inconsistent with this three-flavour picture have motivated the hypothesis that an additional neutrino state exists, which does not interact directly with matter, termed as ‘sterile’ neutrino,ν_s(refs. ^3–9). This includes anomalous observations from the Liquid Scintillator Neutrino Detector (LSND)³experiment and Mini-Booster Neutrino Experiment (MiniBooNE)^4,5, consistent withν_μ → ν_etransitions at a distance inconsistent with the three-neutrino picture. Here we use data obtained from the MicroBooNE liquid-argon time projection chamber¹⁰in two accelerator neutrino beams to exclude the single light sterile neutrino interpretation of the LSND and MiniBooNE anomalies at the 95% confidence level (CL). Moreover, we rule out a notable portion of the parameter space that could explain the gallium anomaly^6–8. This is one of the first measurements to use two accelerator neutrino beams to break a degeneracy betweenν_eappearance and disappearance, which would otherwise weaken the sensitivity to the sterile neutrino hypothesis. We find no evidence for eitherν_μ → ν_eflavour transitions orν_edisappearance that would indicate non-standard flavour oscillations. Our results indicate that previous anomalous observations consistent withν_μ → ν_etransitions cannot be explained by introducing a single sterile neutrino state.