Our Work

This article presents a tailored research tool developed to explore and test the recent theory of deeply embedded core normative values, which asserts that beliefs about the legitimacy of law-enforcem…

This article presents a tailored research tool developed to explore and test the recent theory of deeply embedded core normative values, which asserts that beliefs about the legitimacy of law-enforcement authorities derive in part from profound cultural orientations through which individuals act in and interpret the world. The model is based on a typology of four core normative value systems, namely religious–traditional, liberal, republican–communitarian, and ethno-national, which are associated in different ways with legitimacy. We first conducted in-depth interviews with 45 participants to develop the research tool (questionnaire), then performed a large-scale representative survey among 1,617 Israeli respondents. The findings support the main premises of the model. They show that liberal, republican–communitarian, and religious–traditional values have significant associations with two measures of legitimacy (trust and obligation to obey the police), and that these associations generally hold beyond perceptions of police conduct and performance.Published version: Policing: A Journal of Policy and Practice, 17, paad027. https://doi.org/10.1093/police/paad027

Batch upload of files from directory 'nicotiana_nudicaulis_decontaminated_FSCR'.

The aim of this resource is to provide researchers a set of genomes for sex-aware study. Current genomes may be suboptimal in read detection, due to the fact that some sequences are shared between X a…

The aim of this resource is to provide researchers a set of genomes for sex-aware study. Current genomes may be suboptimal in read detection, due to the fact that some sequences are shared between X and Y chromosomes (PAR regions, XTR and other fine-grained regions). Therefore, to improve read alignment accuracy, the best strategy set is to map reads to XX or XY-adapted genomes depending on the sample karyotype. This resource includes genome .fa files of several assemblies (human GRCh38 and GRCh37 from ensembl, and T2T from UCSC; mouse GRCm39 and GRCm38 from ensembl) ready for use to create aligner indexes, as well as the most relevant genome annotations for each release, adapted to XX/XY masking. T2T release does not include annotations, although they can be generated from the source code in Github. 

Dataset Title Supporting Data for “Securing IoT–Cloud Sensing Systems for Viable Renewable Energy Supply Chains” General Description This repository contains expert comparison data a…

Dataset Title Supporting Data for “Securing IoT–Cloud Sensing Systems for Viable Renewable Energy Supply Chains” General Description This repository contains expert comparison data and associated consistency results supporting the study “Securing IoT–Cloud Sensing Systems for Viable Renewable Energy Supply Chains,” accepted for publication in Energy for Sustainable Development. The study examines implementation constraints affecting secure IoT–cloud sensing systems in renewable energy supply chains in Mexico. The repository documents the fuzzy comparisons used for constraint weighting and the corresponding expert-level consistency assessments under the fuzzy linear best–worst method (FLBWM). Repository Contents 1. Pairwise comparison data.xlsx This workbook contains fuzzy comparison evaluations provided by 15 domain experts. It includes two sheets: Best-to-Others Comparisons: preferences of the selected best (most important) criterion over the other criteria in each comparison set. Others-to-Worst Comparisons: preferences of the other criteria over the selected worst (least important) criterion in each comparison set. Each expert’s evaluations are presented in a separate column. 2. FLBWM_consistency_results.xlsx This supplementary workbook contains 90 records covering 15 anonymized experts, identified as S01–S15. Each expert has six records: one for the main category comparison and one for each of the five within-category comparison sets. The variables are defined below: Variable Description Expert Anonymized expert identifier, from S01 to S15. Set Comparison set: Main denotes the comparison of categories C1–C5; C1–C5 denote comparisons of the criteria within each respective category. Best Code of the criterion selected as most important within the comparison set. Worst Code of the criterion selected as least important within the comparison set. Best–Worst TFN Triangular fuzzy number expressing the preference of the selected best criterion over the selected worst criterion. ξ* Optimal deviation value obtained from the FLBWM optimization model for the corresponding expert and comparison set. CI Consistency index corresponding to the best–worst fuzzy preference. CR Consistency ratio, calculated as ξ*/CI. Time (s) Recorded computational solution time, in seconds, for the corresponding optimization run. Criterion codes follow the notation used in the accompanying article. The within-category code prefixes are SV for C1, EV for C2, OV for C3, OP for C4, and IV for C5. Numerical results are reported at the precision displayed in the workbook. Consequently, recalculating CR from the displayed, rounded ξ* and CI values may produce small differences from the reported CR. Values displayed as 0.0000 should be interpreted at the reported numerical precision. Expert Panel The comparison data were provided by 15 experts with academic and professional backgrounds relevant to renewable energy systems, supply chain management, logistics, cloud computing, IoT systems, cybersecurity, and digital infrastructure. Expert identifiers are anonymized, and personally identifiable information is not included in the shared files. Linguistic Scale The following linguistic terms and triangular fuzzy numbers were used in the fuzzy comparison process: Linguistic term Abbreviation Triangular fuzzy number Equally Important EI (1, 1, 1) Weakly Important WI (2/3, 1, 3/2) Fairly Important FI (3/2, 2, 5/2) Very Important VI (5/2, 3, 7/2) Absolutely Important AI (7/2, 4, 9/2) Version Update This version replaces the previously deposited dataset with an updated Excel file containing expert-level FLBWM best/worst selections and consistency results. The record title and description have been revised to reflect the current contents. Data Usage Notes The files support examination of the expert judgments and consistency assessments used in the FLBWM weighting analysis. The consistency-results workbook should be interpreted alongside the original comparison workbook and the methodological details in the accompanying article. Computational solution times depend on the hardware, software, solver, and execution conditions and should not be treated as general performance benchmarks. Reuse is governed by the license selected for this Zenodo record. Please cite the dataset and the accompanying article when using these materials.

GPTQ-style quantization damps the calibration Hessian by a fixed fraction of its mean diagonal (one percent in the GPTQ authors' code, five percent in a maintained implementation), a choice that ignor…

GPTQ-style quantization damps the calibration Hessian by a fixed fraction of its mean diagonal (one percent in the GPTQ authors' code, five percent in a maintained implementation), a choice that ignores how many calibration rows there are. Small tabular, time-series and text-embedding foundation models can be calibrated with about as many rows as their layers are wide or fewer, and there no fixed fraction fits: on the layers we measure, the best fixed multiplier moves from 0.0064 where rows and width are of the same order to 0.77 where calibration is scarcer. Building on the known damping–ridge identity, we show that the Cholesky factor GPTQ already builds holds the d nested ridge regressions that the damping regularizes. Under an idealized rounding model, this reading decomposes the layer's population error exactly and bounds, from the population activation spectrum alone, what GPTQ can gain over rounding to nearest; it also makes the penalty estimable, because the leave-one-out error summed over the regressions (LOOCV) comes out of the same factorization with no extra quantization pass. On 175 real-layer cells with at most 512 calibration rows from four foundation models, LOOCV, chosen among six cross-validation criteria, has a median excess cost (extra error over the best error among the selection sweep's arms, as a fraction of its gain over rounding to nearest) of 0.0048 against 0.0843 for the one-percent default and 0.0401 for the five-percent rule, lower than every fixed, searched or hindsight rule we compare; a confirmation with the rule fixed in advance, on re-split calibration pools and a model absent from the selection, repeats its advantage over the one-percent default where calibration rows and layer width are of the same order. In whole-model 4-bit quantization of the Qwen3.5-0.8B language model, LOOCV also gives lower WikiText-2 perplexity than the one-percent default in each of the three calibration-size and format settings we test, and lower than the five-percent rule with 512 calibration tokens. Where calibration rows and layer width are of the same order, damping can therefore be estimated from each layer's calibration set rather than fixed.

Release accompanying the revised SoftwareX submission. spherical_shell: exact solution of the sphere for temperature and thermo-elastic stresses, structured hexahedral octant mesh, second-order mesh …

Release accompanying the revised SoftwareX submission. spherical_shell: exact solution of the sphere for temperature and thermo-elastic stresses, structured hexahedral octant mesh, second-order mesh convergence, plots. creep_shrink_fit_2D: Lamé signs in the elastic factor, eq. (20) for the hub creep rate, independent radial J2 reference (reference_1d.py) for the relaxation. mesh_sensitivity_2D: exact mean temperature in the stress reference; the stress converges at second order. coaxial_gap_3D: gold re-blessed. Teaching cases: 02_tensile_bar_3D and 03_plate_with_hole_2D. Documentation rewritten against the code: input-key reference with defaults, verification page, models under development, parallel runs, material-data disclaimer; docs build dependencies fixed. docs-auditor agent; the version check now covers the citation block.

Una  glosa de un Gran Libro :"Escuela de Mandarines" (1974), la novela de Miguel Espinosa, partiendo de la "anécdota" —la Feliz Gobernación y sus seis castas— para eleva…

Una  glosa de un Gran Libro :"Escuela de Mandarines" (1974), la novela de Miguel Espinosa, partiendo de la "anécdota" —la Feliz Gobernación y sus seis castas— para elevarla a categoría: el eón del mandarinazgo como forma permanente del poder que coloniza el lenguaje, convierte el saber en dogma y organiza la sociedad en castas de aspiración y sumisión. El Eremita, homo viator, abandona la Naturaleza y emprende un viaje iniciático no para conquistar el poder, sino para interrogarlo con serenidad y escucha. Su victoria no es la derrota de los mandarines, sino la de no dejarse absorber por ellos. La prosa practica un extrañamiento deliberado (neologismos, mayúsculas institucionales) puesto al servicio de una voluntad de medida clásica: hipérbole calculada, ironía que revela, belleza que resiste. Se analiza el poder como discurso que fabrica la realidad, la lógica de clausura de las castas, el tiempo hiperbolizado que relativiza la pretensión de eternidad del sistema, y la resistencia que nace de la palabra viva y de la ética del encuentro. La conclusión es clara: mandan los mandarines, pero no mandan del todo mientras quede alguien dispuesto a caminar, escuchar y no dañar la vida. El estilo del comentario sigue la vía de la glosa: de lo concreto a lo categorial, con claridad, medida y lucidez, sin nombrar a comentaristas ni expertos, y sin renunciar a la emoción gobernada por la forma.

Batch upload of files from directory 'nicotiana_noctiflora_decontaminated_FSCR'.

Supporting materials for the MethodsX manuscript by Kazuki Abe,Youhi Morii, and Kaoru Maruta, “Quasi-DNS with chemical kinetics for asingle multiple-injection burner element under stable lean co…

Supporting materials for the MethodsX manuscript by Kazuki Abe,Youhi Morii, and Kaoru Maruta, “Quasi-DNS with chemical kinetics for asingle multiple-injection burner element under stable lean conditions forfuture gas turbine applications.” This archive contains the OpenFOAM case files for the RANS and Quasi-DNScalculations, including the final computational meshes, boundary and operatingconditions, chemical-mechanism files, and numerical settings. It also includesthe processed data used to generate selected principal and supplementaryfigures, together with selected post-processing scripts. The final solution fields are provided separately as `2.63.tar.gz` for theRANS calculation and `0.0261.tar.gz` for the Quasi-DNS calculation. Extracteach archive in the root directory of its corresponding OpenFOAM case torestore the final-time directories, `2.63/` and `0.0261/`, respectively. The complete time-resolved three-dimensional field histories from theproduction simulation are not included because of their storage requirements.