Archives ouvertes de Paris-Saclay
Not a member yet
272730 research outputs found
Sort by
Bootstrapping ABJM theory
No full textInternational audienceSupersymmetric localization reduces the computation of protected observables in ABJM theory to finite-dimensional matrix integrals. Building on the techniques introduced in arXiv:2512.02119, we develop a bootstrap framework for the systematic calculation of instanton corrections to the free energy and to supersymmetric Wilson loops. Exploiting exact functional relations and consistency conditions satisfied by grand-canonical observables, in the Fermi-gas formulation of the ABJM matrix model, we provide analytic derivations of several relations for the free energy that were previously known only conjecturally, either from refined topological string theory or from high-precision numerical studies. We apply the same framework to determine the nonperturbative corrections to and BPS Wilson loops, elucidating their qualitative differences and uncovering novel structural features of the instanton effects. These results further highlight the intricate nonperturbative structure and network of dualities underlying ABJM theory
NMR methods for characterizing molecular species within two immiscible solvents: application to SABRE-hyperpolarised species
No full textInternational audienceThe potential of 1 H NMR for studying the structure and dynamics of solutes on either side of an interface between two immiscible liquids is analysed. First, monitoring via localised spectroscopy of the migration of a solute from an aqueous phase to an organic phase, and the measurement of its self-diffusion coefficient slice by slice within the two phases, validated the NMR approach. Next, a fast version of localised spectroscopy (Radsl-CSI) was designed and applied to track the fate of species hyperpolarized by para-hydrogen from the organic phase where they are produced to an aqueous phase.This sequence evidenced the transfer of hyperpolarised pyridine from the organic phase to the aqueous phase, with a substantial signal gain in the latte
Bounds on Lorentz invariance violation from muon fluctuations at the Pierre Auger Observatory
No full textInternational audienceQuantum gravity theories often modify spacetime symmetries. In particular, Lorentz invariance may be violated when approaching the Planck scale. Although the scales at which interactions occur in extensive air showers induced by ultra-high-energy cosmic rays in the atmosphere are many orders of magnitude below the Planck scale, these violations might still be observable. In this work, the fluctuations in the number of muons in the extensive air showers measured at the Pierre Auger Observatory are exploited, for the first time, to constrain Lorentz invariance violations. The bounds derived in the hadronic sector are the strongest ever obtained, and do not rely on assumptions about the mass composition of ultra-high-energy cosmic rays. The fluctuations in the number of muons constitute a new and powerful observable to further explore Lorentz invariance in a region of the parameter space not accessible to other observables
High-level hadronic tau lepton triggers of the CMS experiment in proton-proton collisions at = 13.6 TeV
No full textInternational audienceThe trigger system of the CMS detector is pivotal in the acquisition of data for physics measurements and searches. Studies of final states characterized by hadronic decays of tau leptons require the reconstruction and the identification of genuine tau leptons against quark- and gluon-initiated jets at the trigger level. This is a difficult task, particularly as improvements to the LHC have resulted in an increased number of interactions per bunch crossing in recent years. To address this challenge, a series of machine-learning algorithms with high identification efficiency and low computational cost have been incorporated into the high-level trigger for hadronically decaying tau leptons. In this paper, these developments and the trigger performance are summarized using data collected by the CMS experiment in proton-proton collisions at = 13.6 TeV in 20222023, corresponding to an integrated luminosity of 62 fb
Bernoulli and Gauss Take a Look at the MAP
No full textThis article discusses two Maximum a Posteriori (MAP) interpretations for state-of-the-art methods used in sparse inverse problems: the joint-MAP and the Marginal-MAP. Canonically rooted in a Bayesian framework, sparsity is modeled by a general spike and slab distribution. The focus is on the recovery of the solution support rather than on signal amplitudes. We study the prominent Bernoulli-Gaussian model leading to NP-hard optimization problems. We show that a judicious re-parametrization of the joint-MAP may indeed be a nice surrogate of the marginal-MAP. Additionally, we explore common continuous relaxations of the support and encompass them under the scope of a parametrized distribution. Upon describing the behavior of a few relaxations, strong links are established between the Bernoulli-Gaussian joint-MAP, marginal-MAP, and well-studied methods such as the Lasso and Sparse Bayesian Learning. Finally, the utilization of randomized rounding for both joint-MAP and marginal-MAP problems yields valuable insights into obtaining sparse solutions with an emphasis on support recovery
从低空经济看战略性新兴产业换道超车
No full text近年来，低空经济上升为我国战略性新兴产业，呈高速发展态势。国内学界与业界围绕这一领域已形成丰富研究与讨论，然而对其演进方式仍存在捷足先登抑或后发制人的认知分歧。本文基于战略管理范畴的先动优势与后动优势理论，对低空经济跃迁机制展开系统分析，进而提炼出双优势协同换道超车路径。理论层面，本文构建了融通微观与宏观视域的中观产业分析框架，并揭示了先动优势与后动优势的辩证统一关系，为相关研究提供新思路。实践层面，本文为战略性新兴产业实现换道超车贡献可借鉴进路，亦反驳海外舆论对我国产业发展模式的片面质疑。结语对低空经济未来市场空间与演化趋势进行展望
Triaxiality of neutron-rich ruthenium nuclei studied by lifetime measurements
No full textInternational audienceThe breaking of axial symmetry in nuclei enables otherwise precluded behaviours, making it an interesting phenomenon to study. Experimental fingerprints such as very low-lying states suggest pronounced triaxial deformation for the neutron-rich ruthenium isotopes. Nevertheless, theoretical calculations differ in the description of the triaxial deformation and its evolution with neutron number, making experimental data crucial to understanding it. We investigated the evolution of the degree of triaxiality and rigidity in neutron-rich ruthenium isotopes by measuring lifetimes of excited states in Ru with the recoil distance Doppler-shift method. The experiment was carried out at the Grand Accélérateur National d’Ions Lourds using the Advanced Gamma Tracking Array coupled to the Variable Mode Spectrometer. We obtained B(E2) values for 29 transitions in the studied nuclei and compared them with fully microscopic symmetry conserving configuration mixing calculations, and phenomenological generalized triaxial rotor and triaxial particle-rotor models. The models generally reproduce the measured transition strengths, and show an increase in triaxiality with neutron number, reaching near maximum triaxiality in Ru. The results are consistent with a transition from soft to rigid motion as the neutron number increases
Regional source attribution of tropospheric ozone to NOx and volatile organic compounds in the Beijing-Tianjin-Hebei region using the WRF-Chem model
No full textInternational audienceAs one of China’s densely populated and economically developed regions, the Beijing-Tianjin-Hebei (BTH) area faces severe ozone (O3) pollution, largely driven by nitrogen oxides (NOx) and volatile organic compounds (VOCs) through complex chemical reactions. A detailed quantification of the specific contributions of various NOx and VOCs sources to O3 levels under meteorological and boundary-layer dynamics is crucial for effective pollution control in the BTH region. In this study, we developed and implemented an explicitly tagging approach and process analysis method in the WRF-Chem model to attribute O3 formation separately to NOx and VOCs from various sources, thereby quantifying the contributions of anthropogenic and natural emissions in O3 budget in the BTH region. The results reveal that NOx emissions primarily contribute to O3 through anthropogenic sources (63.5%), while for VOCs, the contributions from anthropogenic emissions (29.3%), background methane (15.8%), and biogenic sources (11.8%) are comparable. Notably, regional transport significantly contributes to O3 levels in the BTH region through both VOCs (80.4%) and NOx (71.2%), with emissions from Shandong, Henan, northeast China, and the Yangtze River Delta, identified as key anthropogenic sources regions of the interprovincial transport. The northeastern low-pressure system and the western Pacific subtropical high drive interprovincial pollutant transport through the free troposphere, with subsequent downward mixing of O3 formed in the free troposphere, contributing significantly to regional transport’s impact on O3 pollution in the BTH. This study provides a comprehensive assessment of ozone source attribution in the BTH area, emphasizing the importance of coordinated regional strategies for O3 mitigation, with particular attention to synoptic influences and planetary boundary layer processes.Regional pollutant transport strongly affects BTH ozone levels, particularly from Shandong, Henan, and the Yangtze River Delta. Large-scale weather systems further enhance long-range ozone transport through vertical mixing. These findings highlight the need for coordinated regional and cross-sector emission control strategies to effectively reduce ozone pollution in northern China
Pronostic hybride à l'aide de codes de simulation et de modèles statistiques : Application à l'étude du colmatage des générateurs de vapeur
No full textThis PhD thesis focuses on the development of hybrid methods for degradation prognostics in industrial systems. The main application concerns the clogging of steam generators (SGs) in pressurized water reactors operated by Électricité de France (EDF). Two main families of models are used in prognostics: physics-based models and statistical, data driven models. Hybrid approaches aim to combine both to leverage their respective strengths and improve robustness, particularly regarding predictive uncertainty. These tools are essential for planning maintenance or deciding on component replacement in long-lifespan critical infrastructure. In scenarios where degradation data is regularly available over time, filtering techniques (e.g., Kalman, particle filters) are effective at correcting simulation-based predictions using new observations. However, for complex systems like SGs, sparse data and complex physics make the prediction task strongly context-dependent. Still, general methodological principles can be established.Physics-based models come with structural biases and parametric uncertainty due to incomplete knowledge of input variables. Their use requires sensitivity analysis and rigorous uncertainty quantification, assuming the physical process is well modeled. When simulations are computationally expensive, surrogate models (or emulators) become necessary. The first part of this thesis develops a non intrusive UQ method applied to an industrial clogging prediction code developed by EDF. Results align with expert knowledge and reveal significant prognostic uncertainty. It is then crucial to evaluate the predictive quality of the emulators. Conformal prediction offers a robust distribution-free framework to construct prediction intervals with guaranteed coverage. We develop estimators suited for limited-data settings, producing intervals for scalar Gaussian processes. Unlike Bayesian credible intervals, our bounds are less sensitive to prior misspecification. For deterministic codes, interval width reflects surrogate approximation error, making them useful diagnostic tools. The next phase involves conditioning the prior distributions on available heterogeneous data to improve predictive robustness. Unlike standard Bayesian calibration, typically applied using lab data, the goal here is to adapt probabilistic predictions to operational field contexts. We propose a data fusion approach inspired by data assimilation, tailored to sparse and heterogeneous sources (e.g., operator measurements, statistical models). Applied to a synthetic crack propagation case and SG clogging, the method significantly improves predictive performance. Open questions remain regarding latent variable uncertainty and discrepancy modeling. Finally, in an exploratory direction, recalibrated simulations can generate degradation trajectories suitable for time series learning. Real-world systems often rely on sensor data that do not directly measure degradation. A key research question is whether such exogenous signals can predict future degradation states. If correlation exists, unobserved degradation may be inferred through features extracted from sensor signals. This work contributes to the development of digital twins, where hybrid modeling, uncertainty quantification, and data integration enable the construction of robust and certifiable predictive frameworks for industrial components such as those in nuclear power plants.Ces travaux de thèse portent sur le développement de méthodes hybrides pour le pronostic de dégradation dans les systèmes industriels. L'application principale concerne le colmatage des générateurs de vapeur (GV) des réacteurs à eau pressurisée opérés par Électricité de France (EDF). Deux grandes familles de modèles sont mobilisées : les modèles physiques et les modèles statistiques. Les approches hybrides cherchent à combiner ces paradigmes pour tirer parti de leurs complémentarités, en particulier pour mieux maîtriser l'incertitude prédictive. Ces outils sont cruciaux pour la maintenance préventive ou le remplacement d'équipements dans des infrastructures critiques. Lorsque des données de dégradation sont disponibles régulièrement, les techniques de filtrage (Kalman, particulaire) sont efficaces pour corriger les prédictions issues des simulations. Toutefois, dans le cas des GV, les données sont rares et les phénomènes physiques complexes, rendant le pronostic fortement dépendant du contexte. Des lignes directrices générales peuvent cependant être définies. Les modèles physiques présentent un biais structurel et une incertitude paramétrique liée à la méconnaissance des entrées. Leur bon usage nécessite une analyse de sensibilité et une quantification d'incertitude rigoureuse. Lorsque les simulations sont coûteuses, des métamodèles deviennent nécessaires. La première partie de la thèse propose une méthode non intrusive de quantification d'incertitude appliquée à un code industriel EDF. Les résultats mettent en évidence une incertitude de pronostic significative, en cohérence avec l'expertise métier. Il est ensuite crucial d'évaluer la qualité prédictive des émulateurs. Les méthodes de prédiction conforme offrent une approche robuste, sans hypothèses fortes, pour produire des intervalles de prédiction. Nous développons des estimateurs adaptés aux faibles régimes de données, appliqués à des processus Gaussiens scalaires. Contrairement aux intervalles bayésiens, ceux-ci sont peu sensibles aux priors mal spécifiés. Dans le cas déterministe, leur taille reflète l'erreur d'approximation, ce qui en fait un outil diagnostic pertinent. La suite du travail consiste à ajuster les lois a priori à l'aide des données disponibles, afin d'adapter la prédiction probabiliste au contexte opérationnel réel. Nous proposons une méthode de fusion de données inspirée de l'assimilation, adaptée à des sources d'information hétérogènes et peu nombreuses (modèles statistiques, observations terrains). Appliquée à un cas jouet de fissuration et au colmatage des GV, cette approche améliore les performances prédictives. Des questions subsistent sur la prise en compte des variables latentes et la modélisation de la discrépance. Enfin, dans une perspective exploratoire, les simulations recalibrées peuvent générer des trajectoires utiles à l'apprentissage temporel. Les données capteurs disponibles sur les systèmes réels ne mesurent pas directement la dégradation. Une question centrale est de savoir si ces signaux permettent d'inférer les états futurs de dégradation. En cas de corrélation, l'information exogène peut être exploitée via des features extraits des signaux. Ces travaux s'inscrivent dans le développement des jumeaux numériques : l'intégration de modèles hybrides, la quantification d'incertitude et la fusion de données permettent de construire des cadres prédictifs robustes et certifiables pour les composants industriels des centrales nucléaires
[French recommendations for clinical practice, Nice/Saint-Paul-de-Vence 2024-2025: Management of localized cervical cancer].
No full textInternational audienceLocalized or locally advanced cervical cancer is treated with a curative intent. Its management requires multidisciplinary expertise and a rigorously structured approach to optimize the probability of success. Initial workup (clinical examination, imaging, pathology) allows precise characterization of the tumour and staging according to TNM and FIGO classifications. Surgical management of early stage cancers, ranging from conization for small tumour to hysterectomy, sometimes including sentinel lymph node biopsy, is based on therapeutic algorithms that take into account stage, pathological criteria (invasion, margins, node involvement) and risk category. Postoperative treatment, when required, includes radiochemotherapy, that can be followed by brachytherapy. In locally advanced cancers, treatment consists of radiochemotherapy followed by uterovaginal brachytherapy and immunotherapy that has recently demonstrated its benefits. Since cervical cancer often develops in young women, its management raises important questions related to fertility and sometimes, to the management of cancer during pregnancy. Finally, although it is not the topic of these recommendations, it is important to highlight the major role of vaccination to avoid the vast majority of these cancers