Scientific Publications of the University of Toulouse II Le Mirail
Not a member yet
92205 research outputs found
Sort by
Les décors figuratifs sur support céramique au 6e millénaire en Méditerranée occidentale : l’exemple de Basi (Serra- di-Ferro, Corse du Sud)
International audienc
Bumblebees locate goals in 3D with absolute height estimation from ventral optic flow
International audienceWhen foraging, flying insects such as bees are often required to change their flight altitude from close to the ground to above the height of vegetation to reach their nest or a food source. Although the mechanisms of navigating towards a goal in two dimensions are well investigated, the explicit use of height as a source for navigation in three dimensions remains mostly unknown. Our study aimed to unravel which strategies bumblebees use for height estimation and whether they rely on global or local cues. We expanded a 2D goal localization paradigm, where a goal location is indicated by cylindrical landmarks, to the third dimension using spherical landmarks to indicate a feeder's position in 3D and examined the search pattern of bumblebees. Additionally, we assessed the ability of bees to estimate the height of a feeder based on local landmarks and global references such as the ground. The search distribution for a feeder's position in 3D was less spatially concentrated compared with 2D. Assessing the height estimation ability of the bees, we found they could estimate a feeder's height using the ground as a reference. However, the feeder needed to be sufficiently close to the ground for the bees to choose correctly. When bumblebees are faced with the challenge of foraging in a 3D environment where the height of a food source and landmark cues are important, they demonstrate the ability to learn and return to a specific flower height. This suggests bumblebees rely on ventral optic flow for goal height estimation
Interindividual variability in flower pickiness by foraging bumblebees
International audiencePollinators navigate complex and heterogeneous “flower markets”, where floral resources vary in quality, availability, and spatial distribution. Bumblebees, as generalist foragers, visit numerous flowers during their foraging bouts, yet the factors influencing their flower choices and the individual differences in foraging behaviour remain poorly understood. Here, we tested how bees adjust their foraging in response to different reward structures. Bombus terrestris workers completed three foraging bouts in two artificial flower environments: one simulating a favourable environment with patches alternating high- and low-quality flowers (40% vs. 20% w/w sucrose solution), and the other a challenging environment with high-quality flowers alongside unrewarded ones (40% w/w sucrose solution vs. plain water). We hypothesised that bees would improve their foraging efficiency in both environments, but more rapidly in the more extreme one, where the greater reward difference creates stronger pressure to learn quickly. In both conditions, bees increased their sucrose intake per unit time over bouts. We also observed consistent differences in flower selectivity among individuals: in the favourable environment, bees that first visited high-quality flowers focused on them and avoided low-quality ones (became “picky”), while bees that first visited low-quality flowers kept visiting both types. Despite these differences, bees across environments and pickiness levels all reached similar sucrose intake rates by the third foraging bout, either by becoming more selective, collecting more sucrose solution, or reducing time spent foraging. These findings highlight the adaptability of bee foraging and suggest that early flower experiences may contribute to lasting individual differences in foraging behaviour. Significance statement Bumblebees are highly efficient pollinators of wild plants and commercial crops, yet their foraging behaviour varies notably among individuals. Understanding why bees differ in their foraging decisions is crucial, as it affects how they collect resources. We show that individual Bombus terrestris workers rapidly adapt to flower patches with different reward qualities. Remarkably, a bee’s first flower visits have lasting effects: individuals that started with high-quality flowers consistently favoured them and avoided low-quality ones, while those that started with low-quality flowers continued to visit both types over time. Despite these differences, bees reached similar nectar intake rates after just three foraging trips, regardless of how selective they were. Our findings show that bees can quickly adjust their foraging behaviour, and that early experiences play a key role in how they exploit resources across different environments
Advancing Amharic Information Retrieval: A Comparative Analysis of Traditional, Neural, and Transformer-Based Models
International audienc
QPP++ 2025: Query Performance Prediction and its Applications in the Era of Large Language Models
International audienceQuery performance prediction (QPP) is a key task in information retrieval (IR) and has been studied for over a decade. The task of QPP is defined as estimating search effectiveness without human relevance judgments. In this workshop, we aim to bring together researchers and practitioners from academia and industry to discuss new perspectives on QPP. Amongst the limitations in the existing QPP literature, we can mention little work has focused on (i) predicting the performance of newly emerged large language model (LLM)-based retrievers/re-rankers or of generative AI systems, (ii) leveraging LLM to model QPP, (iii) investigating concrete applications of QPP, (iv) exploring QPP in the context of multi-modal content, and (v) Exploring multilingual QPP. Those are examples of topics that we encourage authors to contribute to in this workshop
Generating Heterogeneous Multi-dimensional Data : A Comparative Study
Allocation of personnel and material resources is highly sensible in the case of firefighter interventions. This allocation relies on simulations to experiment with various scenarios. The main objective of this allocation is the global optimization of the firefighters response. Data generation is then mandotory to study various scenariosIn this study, we propose to compare different data generation methods. Methods such as Random Sampling, Tabular Variational Autoencoders, standard Generative Adversarial Networks, Conditional Tabular Generative Adversarial Networks and Diffusion Probabilistic Models are examined to ascertain their efficacy in capturing the intricacies of firefighter interventions. Traditional evaluation metrics often fall short in capturing the nuanced requirements of synthetic datasets for real-world scenarios. To address this gap, an evaluation of synthetic data quality is conducted using a combination of domain-specific metrics tailored to the firefighting domain and standard measures such as the Wasserstein distance. Domainspecific metrics include response time distribution, spatialtemporal distribution of interventions, and accidents representation. These metrics are designed to assess data variability, the preservation of fine and complex correlations and anomalies such as event with a very low occurrence, the conformity with the initial statistical distribution and the operational relevance of the synthetic data. The distribution has the particularity of being highly unbalanced, none of the variables following a Gaussian distribution, adding complexity to the data generation process.</div
The evolution of spatial patterns of brain activity during development and their role in the functional specialization of brain networks
International audienceThe development of functionally specialized brain networks allows the emergence and improvement of cognitive abilities. This specialization emerges in part from the interactions between brain regions through changes in the temporal organization of their brain activity. In addition, it is now well assumed that the functioning of mature brain networks also depends on the spatial organization of their brain activity. However, how this spatial organization is formed during development remains unknown, nor do we know whether it contributes to the functional specialization of brain networks. This study attempts to answer these questions by examining the developmental changes in the spatial patterns of brain activity of the functional brain network involved in the theory of mind (capacity to understand others' mental states). Using fMRI dataset from the cross-sectional study of Richardson et al. (2018), our findings show for the first time that the brain regions of this network show increasingly similar spatial patterns of activity. Furthermore, this increasing spatial similarity is associated with improved children's performance in a theory-of-mind task. We propose that the increasing similarity between spatial brain activity patterns during development has important implications for understanding the functional specialization of brain networks
Stated skein modules of 3-manifolds and TQFT
37 pagesInternational audienceWe study the behaviour of the Kauffman bracket skein modules of 3-manifolds under gluing along surfaces. For this purpose we extend the notion of Kauffman bracket skein modules to -manifolds with marking consisting of open intervals and circles in the boundary. The new module is called the stated skein module. The first main results concern non-injectivity of certain natural maps defined when forming connected sums along a sphere or along a closed disk. These maps are injective for surfaces, or for generic quantum parameter, but we show that in general they are not injective when the quantum parameter is a root of 1. The result applies to the classical skein modules as well. A particular interesting result is that when the quantum parameter is a root of 1, the empty skein is zero in a connected sum where each constituent manifold has non-empty marking. We also prove various non injectivity results for the Chebyshev-Frobenius map and the natural map induced by the deletion of marked balls. We then consider the general case of gluing along a surface, showing that the stated skein module can be interpreted as a monoidal symmetric functor from a category of "decorated cobordisms" to a Morita category of algebras and their bimodules. We apply this result to deduce several properties of stated skein modules as a Van-Kampen like theorem as well as a computation through Heegaard decompositions and a relation to Hochshild homology for trivial circle bundles over surfaces
Le pouvoir discrétionnaire en pratique(s)
International audienceWhat does discretionary power signify? How does it differ from arbitrariness? In what practices does it manifest? What do these practices reveal about the state of democracy? And what are the most effective methodological tools for studying it?Based on recent and original research, and drawing inspiration from urgent contemporary issues—ranging from the national use of states of emergency to informal negotiations within local government—the contributions in this book illuminate the various dimensions of discretionary power. They reveal, on one hand, the interactions and organizational contexts that shape the practices of those who wield it, and, on the other hand, its sociopolitical and moral consequences for access to rights and social justice. This book invites readers to critically examine the role of legal constraints on power within so-called democratic systems, at a time when the rule of law appears to be under siege from all directions.De quoi le pouvoir discrétionnaire est-il le nom ? En quoi se distingue-t-il de l’arbitraire ? En quelles pratiques s’incarne-t-il ? Que nous disent-elles de l’état de la démocratie ? Quels sont les outils méthodologiques les plus pertinents pour l’étudier ?En se fondant sur des enquêtes récentes et originales, en s’inspirant d’une actualité brûlante allant de l’usage de l’état d’urgence au niveau national aux négociations informelles au sein d’une mairie, les contributions réunies dans cet ouvrage éclairent les différentes facettes du discrétionnaire en montrant, d’une part, les interactions et les contextes organisationnels qui façonnent les pratiques de celles et ceux qui en font usage, et, d’autre part, leurs conséquences sociopolitiques et morales en matière d’accès aux droits et de justice sociale. Ce livre est une invitation à questionner la place des bornes juridiques du pouvoir dans les systèmes dits démocratiques, à l’heure où l’état de droit semble attaqué de toute part
Sums of squares certificates for polynomial moment inequalities
26 pagesInternational audienceThis paper introduces and develops the algebraic framework of moment polynomials, which are polynomial expressions in commuting variables and their formal mixed moments. Their positivity and optimization over probability measures supported on semialgebraic sets and subject to moment polynomial constraints is investigated. A positive solution to Hilbert's 17th problem for pseudo-moments is given. On the other hand, moment polynomials positive on actual measures are shown to be sums of squares and formal moments of squares up to arbitrarily small perturbation of their coefficients. When only measures supported on a bounded semialgebraic set are considered, a stronger algebraic certificate for moment polynomial positivity is derived. This result gives rise to a converging hierarchy of semidefinite programs for moment polynomial optimization. Finally, as an application, two nonlinear Bell inequalities from quantum physics are settled