Archive ouverte HAL de CY Cergy Paris Université
Not a member yet
38365 research outputs found
Sort by
Physical interpretation of the oscillation spectrum on the RGB and AGB
No full textInternational audienceContext. The high frequency resolution of the four-year time series collected by the space-borne telescope Kepler gives us an opportunity to study the seismic mode structure of highly luminous giants in great detail. Seismic observables can be used as to infer the interior structure through comparisons with stellar models. However, we still need to extend the physical interpretation of previously observed seismic differences between hydrogen-shell burning (Red-Giant Branch; RGB) and helium-burning (red clump and Asymptotic-Giant Branch; AGB) stars towards high luminosity stages.Aims. Here we aim to investigate which physical conditions differ between H-shell and He-burning stars in the heliumsecond ionisation zone, through the signature this zone leaves in mode frequencies. In addition, we explore the sensitivity of seismic parameters to the physics implemented in models. Methods. We used a grid of stellar models with mass between 0.8 M⊙ and 2.5 M⊙ and metallicity between -1.0 dex and 0.25 dex. Transfer mechanisms are implemented such as mass loss, core and envelope overshooting, and thermohaline mixing. We infer the p-mode frequencies of the models by artificially suppressing the gravity modes in the core. Results. In accordance with observations, we find that the main stellar properties affecting the seismic observables in the models are the stellar mass and metallicity. Mass loss on the RGB and rotation-induced mixing from the main sequence to the early-AGB cause a phase difference of the helium ionisation zone glitch signature between H-shell and He-burning stars. The amplitude of the glitch signature in the local large separation, ∆ν, is correlated with the density in the helium ionisation zone, which explains the different glitch amplitudes observed between H-shell and He-burning stars. The amplitude exceeds 10% of the observed value of ∆ν in high-luminosity red giants, which makes the asymptotic expansion less accurate when ∆ν ≤ 0.5 µHz.Conclusions. An efficient mass loss on the RGB, typically encountered when M ≤ 1.5 M⊙, can explain the classification of H-shell and He-burning stars based on the p-mode pattern. When M ≥ 1.5 M⊙, efficient mixing mechanisms might leave an important detectable signature in the p-mode frequencies, permitting a potential classification of these stars.</p
Euclid preparation. Predicting star-forming galaxy scaling relations with the spectral stacking code SpectraPyle
No full textInternational audienceWe introduce SpectraPyle, a versatile spectral stacking pipeline developed for the Euclid mission's NISP spectroscopic surveys, aimed at extracting faint emission lines and spectral features from large galaxy samples in the Wide and Deep Surveys. Designed for computational efficiency and flexible configuration, SpectraPyle supports the processing of extensive datasets critical to Euclid's non-cosmological science goals. We validate the pipeline using simulated spectra processed to match Euclid's expected final data quality. Stacking enables robust recovery of key emission lines, including Halpha, Hbeta, [O III], and [N II], below individual detection limits. However, the measurement of galaxy properties such as star formation rate, dust attenuation, and gas-phase metallicity are biased at stellar mass below log10(M*/Msol) ~ 9 due to the flux-limited nature of Euclid spectroscopic samples, which cannot be overcome by stacking. The SFR-stellar mass relation of the parent sample is recovered reliably only in the Deep survey for log10(M*/Msol) > 10, whereas the metallicity-mass relation is recovered more accurately over a wider mass range. These limitations are caused by the increased fraction of redshift measurement errors at lower masses and fluxes. We examine the impact of residual redshift contaminants that arises from misidentified emission lines and noise spikes, on stacked spectra. Even after stringent quality selections, low-level contamination (< 6%) has minimal impact on line fluxes due to the systematically weaker emission of contaminants. Percentile-based analysis of stacked spectra provides a sensitive diagnostic for detecting contamination via coherent spurious features at characteristic wavelengths. While our simulations include most instrumental effects, real Euclid data will require further refinement of contamination mitigation strategies
HiPP-Prune: Hierarchical Preference-Conditioned Structured Pruning for Vision-Language Models
No full textPruning vision-language models (VLMs) for efficient deployment is challenging because compression can affect not only task utility but also visual grounding, often amplifying object hallucinations even at the same sparsity level. We present HiPP-Prune, a hierarchical preference-conditioned structured pruning framework that treats pruning as conditional resource allocation under multiple objectives. HiPP-Prune makes plan-level decisions: a single policy invocation outputs a global pruning blueprint by factorizing decisions into an overall sparsity budget and a layer-wise allocation, enabling queryable trade-offs via a user-specified preference vector. To account for VLM-specific failure modes, our policy state integrates a visual sensitivity signal derived from attention flow between vision tokens and language hidden states, discouraging over-pruning of vision-critical layers that facilitate cross-modal fusion. We optimize pruning plans with plan-level Group Relative Policy Optimization (GRPO) under a multi-objective return that combines task utility, hallucination robustness (POPE), compression, and a synaptic-flow-inspired stability proxy to reduce unproductive exploration in high-sparsity regimes. Experiments on LLaVA with POPE and ScienceQA demonstrate that HiPP-Prune discovers diverse non-dominated pruning plans and provides controllable robustness--utility trade-offs under matched sparsity budgets
Causality and communication in physics
No full textUnderstanding the concept of causality is fundamental in physics, as it conditions the very possibility of any exchange of information between physical systems. This article examines and compares three main interpretations of the correlative operational concept of "causal separation" that have been respectively proposed in special relativity, quantum theory, and according to the statistical "non-signalling" condition. These interpretations, which can be classified regarding the range of CHSH correlation degrees they allow, lead to radically different results regarding the exchange of information between physical systems. Namely, it can be shown that, in opposition with the non-communication theorem of quantum theory, non-local communication can be established between the two entangled parts of a physical system when the relativistic criterium of causal separation is satisfied but not the quantum one
Engineering Highly Conductive COF-5-Based Architectures: A Strategy to Capitalize on Pore Structure for High-Performance Ion Storage
No full textInternational audienc
Seeing in the dark: Towards a broad construction of the access to data provisions of the DSA
No full textInternational audienceThe Digital Services Act (DSA) is the most ambitious effort taken by liberal democratic nations to regulate social media platforms. One of the main ways it does this is by establishing various transparency obligations applicable to all platforms and search engines, as well as a specific transparency and data-sharing regime for the largest platforms and search engines, defined by the number of users. Specifically, Article 40 of the DSA grants vetted researchers access to data "for the sole purpose of conducting research that contributes to the detection, identification and understanding of systemic risks in the Union. " This article argues that Article 40's requirement for the requested data to be "necessary and proportionate" to conduct a specific type of research may hinder effective research. Indeed, researchers have been denied broad access -or any access to data -on privacy or confidentiality grounds before (Bontcheva, 2024). Consequently, researchers have limited prior knowledge of social media impacts and thus may have limited knowledge about what data, exactly, they need. Drawing on cutting-edge social media research, we explain why vetted researchers may thus need broad access to social media data to meet the objectives of the DSA. More specifically, we argue that researchers need access to system-level data, meaning data that captures how an entire digital system or platform operates, not just the activity of individual users. Consequently, we propose an interpretation of the DSA's data request requirements that would enable researchers to study the online political landscape of EU member states effectively
Unlocking Rare 3- O -Sulfation Patterns in Heparan Sulfate via ARSG-Directed Regioselective Desulfation and HILIC-MS
No full textInternational audienceDeciphering rare 3-O-sulfated motifs in heparan sulfate (HS) is the key to understanding critical biological processes such as viral infection and tumor progression. However, the structural analysis of these elusive structures remains highly challenging due to their low abundance and the limitations of conventional analytical approaches. In this study, we report the application of human lysosomal arylsulfatase G (ARSG), a sulfatase specifically targeting 3-O-sulfated glucosamine residues, as a novel enzymatic tool, which in combination with heparinase and heparanase ¬¬-digestion, facilitates the identification of 3-O-sulfated glucosamine moieties within HS. Coupling ARSG-mediated regi-oselective desulfation with hydrophilic interaction liquid chromatography-mass spectrometry (HILIC-MS) enables the efficient detection and quantification of 3-O-sulfated oligosaccharides in HS and heparin derivatives. This approach overcomes a major analytical barrier in HS characterization and also provides new insights into the distribution and functional roles of 3-O-sulfation. The integration of ARSG expands the enzymatic toolkit available for HS structural characterization and represents a significant advance toward elucidating the structure-function relationships of HS in health and disease
Les petites sources de l’intégrité scientifique
No full textInternational audienceThe field of research integrity is currently witnessing a proliferation of “small-scale normative texts,” such as charters, codes of conduct, and institutional guidelines. These texts are distinctive in that they seek to regulate science—a social institution historically characterized by its autonomy (R. K. Merton)—through professional standards developed “outside state law.” Scientists, as the primary architects of these “small sources,” perceive their involvement as both an expression of academic freedom and a mechanism for self-regulation. Yet, these standards also serve as tools for disciplining scientific behavior, particularly when breaches of integrity are subject to formal sanctions. Moreover, the scope of these normative texts varies widely: from global frameworks, such as the Declarations of the World Conferences on Research Integrity, to highly localized rules specific to individual journals or research institutions. This multiplicity of scales challenges the relevance of a purely national approach to scientific integrity. Against this backdrop, this article examines the role of the actors involved in crafting these small-scale normative texts, with the aim of shedding light on the complex dynamics of normative production in the regulation of scientific activity.Le domaine de l’intégrité scientifique connaît une prolifération de « petites sources » textuelles, telles que des chartes ou codes de bonne conduite. La particularité de ce domaine étant qu’il vise à réguler la science, cette « institution sociale autonome » (R. K. Merton), pour laquelle les scientifiques revendiquent la création de normes professionnelles « hors du droit de l’État ». Cette autonomie des activités scientifiques, et le rôle central qui revient aux scientifiques dans l'élaboration des « petites sources de l’intégrité scientifique », sont vus par ces derniers comme une composante de la liberté académique. Mais les normes de l'intégrité scientifique sont aussi un moyen de discipliner les comportements, notamment lorsque les manquements à l’intégrité scientifique font l'objet de sanctions disciplinaires. Par ailleurs, elles ne se recoupent pas avec le périmètre d’application du droit de l’État : à un extrême, elles sont mondiales (ex. les Déclarations de la World Conference on Research Integrity), et à l'autre, elles sont locales et propres à certaines revues ou établissements de recherche. Il apparaît donc difficile de ne retenir qu’une approche nationale. Le présent article vise ainsi à analyser le rôle des acteurs impliqués dans la mise en place de ces petites sources qui viennent encadrer le fonctionnement de l’activité scientifique, afin d’affiner l’étude des dynamiques de production normative