HAL Portal Univ-Côtedazur
Not a member yet
95391 research outputs found
Sort by
Göpel Varieties
No full text48 pages, comments welcome!International audienceWe show that the Coble hypersurfaces, uniquely characterized by the remarkable property that their singular loci are an abelian surface and a Kummer threefold, respectively, belong to a family of hypersurfaces exhibiting similar behavior, but defined in various types of homogeneous spaces. With the help of Jordan-Vinberg theory, we show how these hypersurfaces can be parametrized by Göpel type varieties inside projectivized representations of complex reflection groups
Seismotectonics of the Oriente Transform Fault revisited
No full textInternational audienceTransform faults are often considered to be geometrically simple, nearly linear, vertical structures that localize crustal deformation within a narrow zone surrounding the fault. The deformation kinematics are typically purely strike-slip, parallel to far-field plate motion, with seismic slip above the brittle-ductile transition, near the 600 °C isotherm, which is well predicted by thermal models. Although deviations from these simplified features have been described, much remains to be learned about the seismogenic behavior of transform faults, for example, why they release much less seismic moment than predicted by plate motion models, or why they so rarely produce earthquakes of magnitudes as large as would be expected given their geometric segmentation (>M7). The Oriente Transform Fault (OTF) along the southern margin of eastern Cuba, at the boundary between the Caribbean and North American plates, is a particularly relevant example to inform on the seismogenic behavior of transform faults for at least 5 reasons: (1) the OTF geometry changes from a nearly continuous trace along the Cayman Ridge to a highly segmented one westward along eastern Cuba, (2) the geometrically continuous segment was the location of a magnitude 7.8 supershear earthquake in January 2020, (3) GNSS-derived strain measurements indicate that this segmentation variation corresponds to a transition from very shallow
Digitally-enabled experiential learning: marine pollution through the lenses of augmented reality
No full textInternational audienc
Perinatal exposure to the autism-linked metabolite p-Cresol has limited impact on early development in mice but lasting effects on adult social behavior
No full textInternational audienceEnvironmental exposures during pregnancy are increasingly recognized as significant risk factors for the offspring health. Perinatal exposure to environmental toxins, including p-Cresol, may disrupt essential physiological processes during critical developmental windows. The perinatal period is particularly critical for neurodevelopment, making the brain particularly vulnerable to external toxins that can impair neural circuitry and increase the risk of behavioral and neuropsychiatric disorders later in life. p-Cresol, a microbial metabolite produced by the human microbiota and through the bacterial decomposition of organic matter, is also released in exhaust fumes and widely used in chemical manufacturing. Elevated levels of p-Cresol have been consistently observed in patients with autism spectrum disorders (ASD), yet, its developmental toxicity and its impact on ASD-related behaviors remain to be determined. In this study, we exposed pregnant mice (C57BL/6J) to p-Cresol via drinking water from mid-gestation, through lactation, and until weaning, mimicking potential patterns of human exposure during pregnancy and early infancy. Our findings indicated that while perinatal p-Cresol exposure did not affect gestational outcomes, postnatal growth, developmental milestones, or sensorimotor reflexes, it resulted in marked social deficits and stereotyped behaviors in adult males and females offspring-which are core behavioral phenotypes associated with ASD. These results suggest that p-Cresol exposure selectively induces ASD-like behaviors, without broadly impairing early development. Our findings emphasize the need for future epidemiological studies to quantify p-Cresol exposure during pregnancy and early childhood, and to evaluate its potential association with social behavior impairments in children. This research provides a foundation for understanding how environmental factors, such as p-Cresol, may contribute to social impairments and neurodevelopmental disorders in the broader population.</div
Deep-Level Transient Spectroscopy evaluation of epilayers grown on bulk GaN substrates for vertical devices
No full textInternational audienceGaN-based lateral power devices such as High Electron Mobility Transistors are now under industrial development. However, vertical device structures are necessary to achieve higher breakdown capabilities and increase the compactness of GaN-based power devices. To ensure the quality of available substrates lattice matched with GaN, it is crucial to characterize epilayers in detail especially focusing on deep carrier traps. These traps can indeed compensate the drift layer doping, affect the leakage current, the switching losses and the breakdown voltage, impacting the global device performance. Deep-Level Transient Spectroscopy (DLTS) is a commonly used technique to characterize deep carrier traps. While numerous reports have identified different traps in GaN-on-GaN epilayers, few studies compare the quality of epilayers grown on different GaN substrates. Therefore, in this work, four different epilayers grown on different available substrates were characterized through DLTS analysis on identical Schottky diodes.The DLTS measurements were performed on 300 μm diameter contacts with two bench configurations; one for multiple measurements and another to reach 80 K. The analysis using the Fourier Transform b1 coefficient reveals three electron traps. The energy levels are around 0.09-0.15 eV for trap E1, 0.48-0.56 eV for trap E2, and 0.78-0.91 eV for trap E3. The first two traps are commonly observed in GaN-on-GaN epilayers as VN and FeGa point defects respectively. The third trap is not systematically observed and is often attributed to dislocations. By increasing the filling time pulse between 2×10-5 to 0.6 s, the E3 trap maximum follows logarithmically the time pulse width with a saturation around 0.1 s. This behaviour confirms the extended defect nature (dislocation-related) of E3 trap.The comparison between the Schottky diode devices reveals that the epilayer grown on the ammonothermal substrate showcases an E2 trap concentration lower by one to two orders of magnitude and the absence of E3 traps. The latter can be explained by the lower dislocation density (in the order of 104 cm-² vs 106-107 cm-² for other substrates) observed by cathodoluminescence on ammonothermal substrates. Therefore, this comparison highlights the higher quality of ammonothermal substrates
Perspicacité-AI: An LLM-Powered Platform for Accelerating Scientific Research and Education
No full textInternational audienceRecently, emerging tools such as LLM have gained attention in scientific research and education. However, their limitations, notably outdated knowledge and untrackable provenance, hinder their ability to keep pace with rapid advancements in scientific fields, like metabolomics. Some of these challenges can be addressed by coupling LLM with Retrieval-Augmented Generation (RAG) techniques, which combine relevant data retrieval with generative AI. Here, we developed Perspicacité-AI, a free and open source agentic literature assistant with advanced RAG capabilities designed for scientific research and education. Perspicacité-AI required the development of the Bibt2KB python package, which seamlessly transforms open-access bibliographic references into a structured and searchable Facebook AI Similarity Search (FAISS) knowledge base (KB). The generated KB is used by the pipeline’s framework to deploy an AI-assistant enhanced by RAG, ensuring that each question asked is grounded in relevant literature and documents. The framework features different advanced modes of document search and question-answering, including Perspicacité-Profound, an iterative, structured reasoning workflow suitable for answering complex scientific questions with the trackable facts and sources. Furthermore, the pipeline supports major open source and proprietary LLM providers, broadening the access for different users. Perspicacité-AI also incorporates a novel reranking function for document selection, Sigmoid Weighted RRF (SW-RRF). The benchmarking of SW-RRF showed means of up to 4.5% improvement over the traditional Reciprocal Reranking Function (RRF) for all tested metrics: Recall, Precision, F1, Mean Reciprocal Rank (MRR), Average Precision (AP) and Normalized Discounted Cumulative Gain (NDCG) at 1, 3, 5 and 10 documents retrieved at no computational expenses.Initial applications of Perspicacité-AI in computational metabolomics [metaboguide.holobiomicslab.eu] demonstrated effective chatbot responses with accurate citations, and while still ongoing, initial evaluations of the pipeline show promising results, not only mitigating some of the limitations of standalone LLMs but empowering scientific literature with generative AI
Robert Badinter, militant d'une Justice pénale internationale
No full textInternational audienceLa journée d'étude en hommage à Robert Badinter a célébré la mémoire d'un homme profondément engagé en faveur des droits fondamentaux. Marqué par la Shoah, il a combattu l'impunité des auteurs de crimes contre l'humanité ou de génocides et refusé les lois d'amnistie et les complaisances politiques. Ces auteurs d'actes odieux doivent être jugés au nom des valeurs universelles des droits de l'homme. Il a ainsi oeuvré pour la création de tribunaux pénaux internationaux, dont celui pour l'ex-Yougoslavie, et pour celle d'une juridiction permanente : la Cour pénale internationale. Défenseur lucide de cette dernière, il en reconnaissait les limites : lenteur, coût et absence de ratification par certaines puissances. Malgré ces imperfections, il y voyait une avancée majeure. Son ultime combat fut consacré à la mise en accusation de Vladimir Poutine pour les crimes perpétrés en Ukraine et à la possibilité de le traduire devant une juridiction pénale internationale.</div
Where do I go? Decoding temporal neural dynamics of scene processing and visuospatial memory interactions using convolutional neural networks
No full textInternational audienceVisual scene perception enables rapid interpretation of the surrounding environment by integrating multiple visual features related to task demands and context, which is essential for goal-directed behavior. In the present work, we investigated the temporal neural dynamics underlying the interaction between the processing of bottom-up visual features and top-down contextual knowledge during scene perception. We asked whether newly acquired spatial knowledge would immediately modulate the early neural responses involved in the extraction of navigational affordances available (i.e., the number of open doors). For this purpose, we analyzed electroencephalographic data from 30 participants performing interleaved blocks of a scene memory task and a visuospatial memory task in which we manipulated the number of navigational affordances available. We used convolutional neural networks coupled with gradient-weighted class activation mapping to assess the main electroencephalographic channels and time points contributing to the classification performances. The results indicated an early temporal window of integration in occipitoparietal activity (50–250 ms post stimulus) for several aspects of visual perception, including scene color and number of affordances, as well as for spatial memory content. Moreover, a convolutional neural network trained to detect affordances in the scene memory task failed to generalize to detect the same affordances after participants learned spatial information about goal position within the scene. Taken together, these results reveal an early common window of integration for scene and visuospatial memory information, with a specific and immediate top-down influence of newly acquired spatial knowledge on early neural correlates of scene perception
Étude réflexive auprès d’étudiants en kinésithérapie : quand la simulation en santé éclaire le raisonnement clinique
No full textClinical reasoning is an essential skill in physiotherapy, involving analysis, decision-making, and adaptation to clinical situations. In this context, simulation training represents an innovative educational method. This study explores how students perceive their own clinical reasoning through the method of simple self-confrontation. Six final-year physiotherapy students from the Nice Physiotherapy Institute took part in a clinical simulation within one of three areas of practice: each simulation was recorded and followed by a debriefing in the form of a simple self-confrontation interview, during which the student commented on their performance. A thematic and enunciative analysis was conducted based on the full transcriptions of these simulations. The results reveal varying levels of structured clinical reasoning among the students. Some adopted an explicit analytical approach, while others relied on more intuitive and fragmented reasoning. The self-confrontation method encouraged the emergence of metacognitive reflections and improved verbalization of clinical choices. Simulation, when combined with simple self-confrontation, appears to be a relevant tool to clarify and enhance clinical reasoning. Its integration into physiotherapy training could offer an opportunity to structure the learning of clinical reasoning and foster a reflective attitude throughout the professional development process.Le raisonnement clinique est une compétence essentielle en kinésithérapie, mobilisant analyse, prise de décision et adaptation aux situations cliniques. Dans ce contexte, la simulation en santé représente une méthode pédagogique innovante.Cette étude explore la manière dont les étudiants perçoivent leur raisonnement clinique à travers la méthode d'auto-confrontation simple. Six étudiants en dernière année de formation en Masso-Kinésithérapie (MK) à l'Institut de Formation en Masso-Kinésithérapie (IFMK) de Nice ont participé à une simulation clinique dans l'un des trois champs d'enseignement. Chaque simulation filmée a été suivie d'un débriefing sous forme d'auto-confrontation simple, où l'étudiant commentait sa prestation. Une analyse thématique et énonciative a été réalisée à partir des transcriptions. Les résultats révèlent des niveaux variés de structuration du raisonnement clinique : certains étudiants adoptent une démarche analytique explicite, tandis que d'autres mobilisent un raisonnement plus intuitif et fragmenté. L'auto-confrontation a favorisé l'émergence de réflexions métacognitives et une meilleure verbalisation des choix cliniques. La simulation, couplée à l'auto-confrontation simple, apparaît comme un levier pertinent pour expliciter et enrichir le raisonnement clinique. Son intégration dans la formation initiale pourrait constituer une opportunité pour structurer l'apprentissage du raisonnement clinique, encourager une posture réflexive tout au long du parcours de professionnalisation
Continuous limits of large plant-pollinator random networks and some applications
No full textInternational audienceWe study a stochastic individual-based model of interacting plant and pollinator species through a bipartite graph: each species is a node of the graph, an edge representing interactions between a pair of species. The dynamics of the system depends on the between- and within-species interactions: pollination by insects increases plant reproduction rate but has a cost which can increase plant death rate, depending on the densities of pollinators. Pollinators reproduction is increased by the resources harvested on plants. Each species is characterized by a trait corresponding to its degree of generalism. This trait determines the structure of the interaction graph and the quantities of resources exchanged between species. Our model includes in particular nested or modular networks. Deterministic approximations of the stochastic measure-valued process by systems of ordinary differential equations or integro-differential equations are established and studied, when the population is large or when the graph is dense and can be replaced with a graphon. The long-time behaviors of these limits are studied and central limit theorems are established to quantify the difference between the discrete stochastic individual-based model and the deterministic approximations. Finally, studying the continuous limits of the interaction network and the resulting PDEs, we show that nested plant-pollinator communities are expected to collapse towards a coexistence between a single pair of species of plants and pollinators