179566 research outputs found
Sort by
Abnormal predicates:Learning categorical defaults from probabilistic rules
Get PDFLearning defaults is a longstanding goal in the field of knowledge representation and reasoning. We provide a novel method for learning defaults by way of introducing a new predicate: the abnormal predicate, which explicitly covers all the exceptions to a rule, thus forming a default theory. Our proposed method for learning defaults is sound and complete for all rule-exceptions, and can be extended for use on other frameworks
History, truthfulness, politicization, value judgements
Get PDFThis article attempts to establish clarity about the relationships between politicization, value judgements, and truthfulness in historical scholarship. It establishes at least one criterion for distinguishing between legitimate and illegitimate politicization based on the imperative of truthfulness. Then it argues that some value judgements and authorial prompts-to-judgement are often unavoidable and are compatible with historians’ commitment to truthfulness. Such judgements must be taken more seriously than they generally are as a principle of producing and consuming histories and not dismissed as the abrogation of a disciplinary norm
Commentary:Time to abandon the 'clinical high risk state for psychosis' (CHR-P) concept in adolescence? Commentary on Frearson et al. 'Efficacy of preventative interventions for children and adolescents at clinical high risk of psychosis: A systematic review and meta-analysis of intervention studies'
Get PDFThere has been much academic interest in 'the clinical high risk for psychosis' (CHR-P) concept. Indeed, as two child and adolescent psychiatrists interested in psychosis prediction and prevention, we enthusiastically embraced the paradigm in our clinical and academic work. However, despite more than two decades of research, there is no definition of CHR-P in adolescence that has proven to be able to usefully predict transition to psychosis. Indeed, research suggests that much, if not all, of the risk associated with CHR diagnoses in adolescents is captured by being help-seeking for mental health problems, rather than being associated with a CHR diagnosis itself. In this commentary, we critique the systematic review by Frearson et al. (2025). In particular, we challenge the conceptualisation around the CHR-P concept, as applied to under 18 s, and the assumptions underpinning it. We also highlight issues with the terminology used when describing the experiences of young people categorised as being at CHR-P. Rather, we make the case for understanding and supporting help-seeking young people with distressing perceptual and ideational disturbance employing a needs-based, person-centred approach.</p
PML mutants from arsenic-resistant patients reveal SUMO1-TOPORS and SUMO2/3-RNF4 degradation pathways
Get PDFArsenic effectively treats acute promyelocytic leukemia by inducing SUMO and ubiquitin-dependent degradation of the promyelocytic leukemia (PML)-retinoic acid receptor alpha oncogenic fusion protein. However, some patients relapse with arsenic-resistant disease because of missense mutations in PML. To determine the mechanistic basis for arsenic resistance, PML-/- cells were reconstituted with YFP fusions of wild-type PML-V and two common patient mutants: A216T and L217F. Both mutants were resistant to degradation by arsenic but for different biochemical reasons. Arsenic did not trigger SUMOylation of A216T PML, which failed to recruit the SUMO-targeting ubiquitin ligases RNF4 and TOPORS. L217F PML did respond with increased SUMO2/3 conjugation that facilitated RNF4 engagement but failed to reach the threshold of SUMO1 conjugation required to recruit TOPORS. Thus, neither mutant accumulated the appropriate polyubiquitin signal required for p97 binding. These PML mutants have revealed a convergence of SUMO1, SUMO2/3, TOPORS, and RNF4 that facilitates the arsenic-induced degradation of PML.</p
Calving from horizontal forces in a revised crevasse-depth framework
Get PDFCalving is a key process for the future of our ice sheets and oceans, but representing it in models remains challenging. Among numerous possible calving parameterisations, the crevasse-depth law remains attractive for its clear physical interpretation and its performance in models. In its classic form, however, it requires ad-hoc and arguably unphysical modifications to produce crevasses that are deep enough to result in calving. Here, we adopt a recent analytical approach accounting for the feedback between crevassing and the stress field and varying the density of water in basal crevasses, and show that it removes the need for such ad-hoc modifications. After accounting for ice tensile strength and basal friction, we show that the revised formulation predicts that full-thickness calving should occur at flotation when the calving front ice thickness is greater than around 400 m. It also predicts no calving for ice thinner than around 400 m, suggesting that calving at such glacier fronts is not driven purely by horizontal forces. We find good observational support for this analysis. We advance the revised crevasse-depth formulation as a step towards understanding differing calving styles and a better representation of calving in numerical models
Crossmodal ASR error correction with discrete speech units
Get PDFASR remains unsatisfactory in scenarios where the speaking style diverges from that used to train ASR systems, resulting in erroneous transcripts. To address this, ASR Error Correction (AEC), a post-ASR processing approach, is required. In this work, we tackle an understudied issue: the Low-Resource Out-of-Domain (LROOD) problem, by investigating crossmodal AEC on very limited downstream data with 1-best hypothesis transcription. We explore pre-training and fine-tuning strategies and uncover an ASR domain discrepancy phenomenon, shedding light on appropriate training schemes for LROOD data. Moreover, we propose the incorporation of discrete speech units to align with and enhance the word embeddings for improving AEC quality. Results from multiple corpora and several evaluation metrics demonstrate the feasibility and efficacy of our proposed AEC approach on LROOD data as well as its generalizability and superiority on large-scale data. Finally, a study on speech emotion recognition confirms that our model produces ASR error-robust transcripts suitable for downstream applications
An educator framework for organizing Wikipedia editathons for computational biology
Get PDFMotivation: Wikipedia is a vital open educational resource in computational biology; however, a significant knowledgegap exists between English and non-English Wikipedias. Reducing this knowledge gap via intensive editing events, or‘editathons’, would be beneficial in reducing language barriers that disadvantage learners whose native language is notEnglish. Results: We present a framework to guide educators in organizing editathons for learners to improve andcreate relevant Wikipedia articles. As a case study, we present the results of an editathon held at the 2024 ISCB LatinAmerica conference, in which ten new articles were created for the Spanish-language edition of Wikipedia. We alsopresent a web tool, ‘compbio-on-wiki’, which identifies relevant English Wikipedia articles missing in other languages.We demonstrate the value of editathons to expand the accessibility and visibility of computational biology content inmultiple languages. Availability and Implementation: Source code for the compbio-on-wiki Toolforge site is available at:https://github.com/lubianat/compbio-on-wik
Emotional contagion, empathy, and sympathy as responses to verbal and visual narratives:Some conceptual and methodological issues
Get PDFThis chapter seeks to air the following basic questions: what is the relationship between the emotions of characters and audiences in genres such as epic and tragedy; and what role do vision and visualization play in the mediation of this relationship? But I want to approach these questions by means of a more limited one: do the emotional responses of audiences depend to any significant extent on forms of mimicry or mirroring of the emotions of characters, and if they do, must these emotions be elicited by direct visual stimuli or can they be set in train by the imagination? This will then lead to a further question: how (if at all) does this issue bear on the theory and practice of ancient literature, and can ancient theory and practice help us frame the questions that we should like to see answered with the help of modern research in psychology and cognitive science
Original vegetation condition and precipitation growth rate bifurcate sediment flux trend on the Qinghai-Tibet Plateau
Get PDFRiverine sediment flux on the Qinghai-Tibet Plateau follows separate trends in different headwater basins because of complicated pathways under global warming. Here we successfully reconstructed historical sediment fluxes from seven pathways at 25 hydrological stations during 1982 ~ 2022 using a conceptual Multivariate Climate Elasticity Model based on Taylor expansion and revealed the bifurcation in sediment flux trend. Significantly increasing trends occurred at five stations, caused either by elevated runoff from increased precipitation (contributing -10 ~ 50%) or by temperature modifying the underlying surface (-12 ~ 100%). Significantly decreasing trends were found at six stations, primarily owing to vegetation expansion (11 ~ 100%). Combinations of original vegetation condition and increasing rate of precipitation are pivotal for bifurcating the sediment flux trend. In the future, slowly increasing precipitation with greatly increasing temperature or poor original vegetation condition followed by rapid vegetation expansion may decelerate sediment yield, whereas rapidly increasing precipitation may overwhelm the marginal conservation effect of plentiful vegetation
Long-term benefits of atorvastatin on the incidence of cardiovascular events:the ASCOT-Legacy 20-year follow-up
Get PDFAIMS: Cardiovascular (CV) deaths were reduced by atorvastatin during a 16-year follow-up of participants in the Anglo-Scandinavian Cardiac Outcomes Trial-lipid-lowering arm. We now extend these observations over 20 years and report both non-fatal and fatal CV outcomes.METHODS: A cohort of 4605 UK hypertensive participants with total cholesterol <6.5 mmol/L (2317 atorvastatin vs 2288 placebo) was followed for up to 21 years (IQR 9.1-19.3). Cox proportional hazard models assessed HRs for non-fatal and fatal CV events. At the end of the original trial (3.3 years), all participants were offered atorvastatin. Lipid profiles were obtained from all subjects 2 years later and from subgroups approximately 9 years post-trial.RESULTS: Patients allocated to atorvastatin had a significant reduction in non-fatal myocardial infarction (MI) and fatal coronary heart disease (CHD) events (HR (95% CI) 0.81 (0.69 to 0.94, p=0.006)), total coronary events (0.88 (0.80 to 0.98, p=0.017)) and CV deaths (0.86 (0.74 to 0.99, p=0.048)). No significant reduction in heart failure (HF), strokes, total CV events and all-cause mortality was observed.In participants assigned atorvastatin in the trial, 3-year mean low-density lipoprotein-cholesterol was strongly associated with long-term CV outcomes. The HRs per 1 mmol/L decrease were for non-fatal MI and fatal CHD (0.69 (0.57 to 0.85, p<0.001)), total coronary events (0.70 (0.61 to 0.79, p<0.001)), non-fatal and fatal HF (0.68 (0.57 to 0.81, p<0.001)), non-fatal and fatal stroke (0.74 (0.59 to 0.92, p=0.006)), total CV events and procedures (0.74 (0.66 to 0.81, p<0.001)), CV mortality (0.66 (0.55 to 0.81, p<0.001)) and all-cause mortality (0.81 (0.71 to 0.90, p<0.001)).Two years after the trial, approximately two-thirds of subjects in each arm were taking atorvastatin. At this time point and approximately 9 years post-trial, lipid profiles were similar between those formerly assigned atorvastatin or placebo.CONCLUSIONS: These observations provide further evidence for the long-term legacy effects of statins and have implications for the early introduction of statins to prevent CV events and mortality.</p
Access Repository Dashboard
Do you manage Edinburgh Research Explorer? Access insider analytics, issue reports and manage access to outputs from your repository in the CORE Repository Dashboard!