Buckley, Andy · Bierlich, Christian · Butterworth, Jonathan · et al.

1 files · 47 MB · gzipdeclared

The Rivet toolkit (Robust Independent Validation of Experiment and Theory) is a system for validation of Monte Carlo event generators. It provides a large (and ever growing) set of experimental analyses useful for MC generator development, validation, and tuning, as well as a convenient infrastructure for adding your own analyses. Rivet is the most widespread way by which analysis code from the LHC and other high-energy collider experiments is preserved for comparison to and development of future theory models. It is used by phenomenologists, MC generator developers, and experimentalists on the LHC and other facilities.

Buckley, Andy · Corpe, Louie · Gutschow, Christian · et al.

1 files · 1.2 MB · gzipdeclared

YODA -- Yet more Objects for Data Analysis -- is a lightweight C++ and Python package for weighted histogramming and reference data manipulation. It is particularly associated with the Rivet MC event analysis framework for High-Energy Physics, but has been written with general-purpose applications in mind. Version 2.0.0 and later are written for histogramming of arbitrary object types, in arbitrary dimensionalities, with statistically consistent fill volumes in mixes of continuous and discrete spaces.

Kvashin, Nikolai · Wolfram, Uwe

1 files · 8.2 MB · gzipdeclared

An open-source MOOSE application implementing a molecular-dynamics-to-finite-element pipeline for hierarchical biominerals: orthotropic crystal plasticity with tension-compression asymmetry, fabric-based elasticity, and a homogenized exponential cohesive-zone interface law. Primary case study: cold-water coral aragonite; validation target: trabecular bone.

Lesbre, Dorian · Lemerre, Matthieu

3 files · 1.3 GB · gzip, zipdeclared

This is the artifact for the paper A Lattice of Union-Finds by D. Lesbre and M. Lemerre , to be published as SAS 2026 . It contains our union-find lattice library, benchmarks, and the frama-c/codex abstract interpreter . We provide a pre-built docker image for x86 processors to easily run the benchmarks and interpreter. We also provide the source code for both the library and interpreter, as well as instruction on how to compile locally or how to rebuild the docker container. See the included README for instructions on how to run our artifact.

Street, Rachel · Bachelet, Etienne · Bozza, Valerio · et al.

4 files · 1.3 GB · gzipdeclared

Entries to the 2018 Microlensing Data Challenge This challenge was organized to stimulate research and development of software for the analysis of microlensing events at scale. The challenge itself is described here and a detailed description of the required contents for each entry can be found here . The original dataset for the challenge is available from this link . This dataset consists of the submitted entries from the four teams that participated, which were anonymized prior to evaluation.

Usai, Luigi

22 files · 467 KB · gzip, pdfdeclared

DOCUMENT METADATA & SUBMISSION SHEET Target Journal: IEEE/ACM Transactions on Computational Biology and Bioinformatics / Journal of Chemical Information and Modeling Classification: Biophysical Software and Engineering / Riemannian Geometry / Causal Inference Framework Status: Phase 5+ Production Engine (Universal Deployment) SECTION 1: ENGLISH VERSION Title Accelerated Transition Path Sampling via Neuro-Symbolic Molecular Dynamics (NSMD) Phase 5+: A Functorial Sheaf and $SE(3)$ Principal Gauge Bundle Engine with Quantitative Liang-Kleeman Causal Inference Abstract This work presents the Phase 5+ production-grade architecture of the Neuro-Symbolic Molecular Dynamics (NSMD) framework, engineered for the deterministic optimization of the Onsager-Machlup action functional on non-Euclidean manifolds. By quotiencing the Cartesian configuration space $\mathbb{R}^{3N}$ by the action of the non-Abelian Lie group $SE(3)$ , we formalize macromolecular dynamics within an $SE(3)$ -principal gauge bundle. Horizontality along the manifold is rigorously preserved via an analytical Ehresmann connection form and equivariant parallel transport of the L-BFGS Riemannian history, completely eliminating gauge drift. To bypass the severe latency of textual hypergraph parsing (e.g., NDJSON-LD), we introduce a Category-Theoretic Functorial Presheaf Tensor (FPT). This operator maps abstract symbolic and secondary structure constraints into differentiable geometric potentials evaluated natively on GPU architectures in $\mathcal{O}(1)$ time complexity. This continuous landscape is coupled with Usai's Semantic Pixel Mapping (USPM) grid to modulate electro-topological anchors natively on CUDA registers. Furthermore, directional allosteric causality is computed on the tangent bundle via the exact mathematical formulation of the Liang-Kleeman Information Flow Rate, dynamically scaling constraint weights based on causal centralities. Benchmarked on Polyubiquitin-B (2,055 backbone atoms), the fully vectorized engine yields full path convergence in 168.0060 seconds with sub-0.03 Å ( $0.02706 \text{ Å}$ ) holonomic bond deviations, validating the numerical accuracy of the odd-even checkerboard retraction solver and establishing a new paradigm for high-throughput macromolecular transition sampling. Keywords Neuro-Symbolic Molecular Dynamics; $SE(3)$ Principal Bundle; Functorial Presheaf Tensor; Usai's Semantic Pixel Mapping; Liang-Kleeman Information Flow; Onsager-Machlup Functional. SECTION 2: ITALIAN VERSION Titolo Campionamento Accelerato dei Cammini di Transizione via Dinamica Molecolare Neuro-Simbolica (NSMD) Fase 5+: Un Motore basato su Fasci Funttoriali e Fibrati Principali di Gauge $SE(3)$ con Inferenza Causale Quantitativa di Liang-Kleeman Descrizione (Abstract Esteso) Il presente lavoro definisce l'architettura strutturale di livello produttivo Fase 5+ del framework di Dinamica Molecolare Neuro-Simbolica (NSMD), esplicitamente sviluppata per l'ottimizzazione deterministica del funzionale d'azione di Onsager-Machlup su varietà non euclidee. Quozientando lo spazio delle configurazioni cartesiane $\mathbb{R}^{3N}$ rispetto all'azione del gruppo di Lie non Abeliano $SE(3)$ , la dinamica macromolecolare viene formalizzata all'interno di un fibrato principale di gauge $SE(3)$ . L'orizzontalità lungo la varietà viene preservata mediante una forma di connessione analitica di Ehresmann e il trasporto parallelo equivariante della storia Riemanniana L-BFGS, annullando sistematicamente la deriva geometrica di gauge. Al fine di eradicare i colli di bottiglia computazionali indotti dal parsing ipergrafico testuale (es. NDJSON-LD), l'infrastruttura introduce un Tensore dei Fasci Funttoriali (FPT) basato sulla teoria delle categorie. Tale operatore mappa vincoli simbolici astratti e asserzioni strutturali in potenziali geometrici differenziabili continui, valutati direttamente su GPU con complessità temporale $\mathcal{O}(1)$ . Questo spazio continuo è accoppiato alla griglia tensoriale di Usai's Semantic Pixel Mapping (USPM), la quale alloca le feature elettro-topologiche ed energetiche della catena direttamente sui registri di esecuzione CUDA. La componente causale allosterica direzionale viene risolta sul fibrato tangente implementando la formulazione esatta del Tasso di Flusso Informativo di Liang-Kleeman, che rimodula dinamicamente i pesi dei vincoli in base all'asimmetria termodinamica del gradiente. Sottoposto a benchmark prestazionale sulla Poliubiquitina-B ( 2055 atomi di backbone ), il motore interamente vettorializzato ha completato la convergenza del cammino d'azione in 168,0060 secondi , registrando una massima deviazione sugli invarianti olonomi pari a soli $0,02706 \text{ Å}$ . L'esito convalida il rigore numerico del solutore di retrazione alternata a scacchiera (odd-even checkerboard relaxation), ponendo il framework come nuovo standard per la simulazione ad alto rendimento delle transizioni conformazionali. Parole Chiave Dinamica Molecolare Neuro-Simbolica; Fibrato Principale $SE(3)$ ; Tensore dei Fasci Funttoriali; Usai's Semantic Pixel Mapping; Flusso Informativo di Liang-Kleeman; Funzionale di Onsager-Machlup.

EASTERN PUBLICATIONS

8 files · 3.1 GB · gzipdeclared

This dataset (from 2017 to 2025) of Administrative Reference Framework is aiming to help bring transparency and significative impartiality to the Opinion emitted on 9.VI.2026 by the French College of Magistrate Deontology stating in its summary, that membership of Freemasonry is incompatible with the ethical obligations incumbent upon any judge where the oath taken entails allegiance or a primary duty of solidarity. Failing that, it gives rise to serious reservations and requires the judge to exercise the utmost vigilance as regards compliance, both in appearance and in fact, with the principles of independence, impartiality and neutrality.