Search

Large Language Models (LLMs) have shown remarkable capabilities, not only in generating human-like text, but also in acquiring knowledge. This highlights the need to go beyond the typical Natural Language Processing downstream benchmarks and asses the various aspects of LLMs including knowledge and reasoning. Numerous benchmarks have been developed to evaluate LLMs knowledge, but they predominantly focus on the English language. Given that many LLMs are multilingual, relying solely on benchmarking English knowledge is insufficient. To address this issue, we introduce AraSTEM, a new Arabic multiple-choice question dataset aimed at evaluating LLMs knowledge in STEM subjects. The dataset spans a range of topics at different levels which requires models to demonstrate a deep understanding of scientific Arabic in order to achieve high accuracy. Our findings show that publicly available models of varying sizes struggle with this dataset, and underscores the need for more localized language models. The dataset is freely accessible on Hugging Face.

Anomaly detection (AD) plays a crucial role in time series applications, primarily because time series data is employed across real-world scenarios. Detecting anomalies poses significant challenges since anomalies take diverse forms making them hard to pinpoint accurately. Previous research has explored different AD models, making specific assumptions with varying sensitivity toward particular anomaly types. To address this issue, we propose a novel model selection for unsupervised AD using a combination of time series forest (TSF) and reinforcement learning (RL) approaches that dynamically chooses an AD technique. Our approach allows for effective AD without explicitly depending on ground truth labels that are often scarce and expensive to obtain. Results from the real-time series dataset demonstrate that the proposed model selection approach outperforms all other AD models in terms of the F1 score metric. For the synthetic dataset, our proposed model surpasses all other AD models except for KNN, with an impressive F1 score of 0.989. The proposed model selection framework also exceeded the performance of GPT-4 when prompted to act as an anomaly detector on the synthetic dataset. Exploring different reward functions revealed that the original reward function in our proposed AD model selection approach yielded the best overall scores. We evaluated the performance of the six AD models on an additional three datasets, having global, local, and clustered anomalies respectively, showing that each AD model exhibited distinct performance depending on the type of anomalies. This emphasizes the significance of our proposed AD model selection framework, maintaining high performance across all datasets, and showcasing superior performance across different anomaly types.

Many global optimization algorithms of the memetic variety rely on some form of stochastic search, and yet they often lack a sound probabilistic basis. Without a recourse to the powerful tools of stochastic calculus, treading the fine balance between exploration and exploitation could be tricky. In this work, we propose an evolutionary algorithm (EA) comprising two types of additive updates. The primary update utilizes stochastic approximation to guide a population of randomly initialized particles towards an optimum. We incorporate derivative-free Robbins-Monro type gains in the first update so as to provide a directional guidance to the candidate solutions. The secondary update leverages stochastic conditioning to apply random perturbations for a controlled yet efficient exploration. Specifically, conceptualized on a change of measures, the perturbation strategy discourages two or more trajectories exploring the same region of the search space. Our optimization algorithm, dubbed as SABRES (Stochastic Approximation and Brownian Repulsion based Evolutionary Search), is applied to CEC-2022 benchmark functions on global optimization. Numerical results are indicative of the potentialities of SABRES in solving a variety of challenging multi-modal, non-separable, and asymmetrical benchmark functions.

This contribution presents an improved low-order 3D finite element formulation with hourglass stabilization using automatic differentiation (AD). Here, the former Q1STc formulation is enhanced by an approximation-free computation of the inverse Jacobian. To this end, AD tools automate the computation and allow a direct evaluation of the inverse Jacobian, bypassing the need for a Taylor series expansion. Thus, the enhanced version, Q1STc+, is introduced. Numerical examples are conducted to compare the performance of both element formulations for finite strain applications, with particular focus on distorted meshes. Moreover, the performance of the new element formulation for an elasto-plastic material is investigated. To validate the obtained results, a volumetric locking-free reference element based on scaled boundary parametrization is used. Both the implementation of the element routine Q1STc+ and the corresponding material subroutine are made accessible to the public at https://doi.org/10.5281/zenodo.14259791, Comment: 41 pages, 12 figures, 4 tables

We present a frequency tunable magneto-acoustic oscillator (MAO) operating in low-phase-noise and complex dynamical regimes based on a single composite YIG-GGG resonator. The magneto-acoustic resonator (MAR) is based on a YIG (yttrium iron garnet) layer epitaxially grown on a GGG (gadolinium gallium garnet) substrate. By optimizing the YIG thickness, we obtain a high magneto-elastic coupling of around 1 MHz between the ferromagnetic resonance (FMR) in YIG and high overtone acoustic resonances (HBARs) in the YIG-GGG structure in the 1-2 GHz frequency range. It allows to eliminate the need for pre-selectors and bulky circulators, thus simplifying the MAO design while maintaining the possibility to lock to HBAR YIG-GGG modes. With an adjustment in the loop over-amplification parameter, the MAO can be locked either only to high-Q magneto-acoustic HBARs or to both types of resonance including HBARs and the FMR mode of the YIG film. In a low-phase-noise regime, MAO generates only at certain values of the applied field and exhibits discrete frequency tunability with a 3.281 MHz step corresponding to the frequency separation between the adjacent HBAR modes in a YIG-GGG structure. In a complex regime where oscillation conditions expand to include both HBAR and FMR modes, MAO demonstrates continuous generation as the function of the applied field with variable phase noise parameters. Moreover, in low-phase-noise regime, MAO phase noise plot improves by 30 dB compared to the operational regime locked to the pure FMR in YIG which is in agreement with the measured FMR and HBAR Q-factors., Comment: Marie Sk{\l}odowska-Curie Actions, H2020-MSCA-ITN-2020; Project acronym SPEAR; Grant Agreement No. 955671

Underwater imagery often suffers from severe degradation that results in low visual quality and object detection performance. This work aims to evaluate state-of-the-art image enhancement models, investigate their impact on underwater object detection, and explore their potential to improve detection performance. To this end, we selected representative underwater image enhancement models covering major enhancement categories and applied them separately to two recent datasets: 1) the Real-World Underwater Object Detection Dataset (RUOD), and 2) the Challenging Underwater Plant Detection Dataset (CUPDD). Following this, we conducted qualitative and quantitative analyses on the enhanced images and developed a quality index (Q-index) to compare the quality distribution of the original and enhanced images. Subsequently, we compared the performance of several YOLO-NAS detection models that are separately trained and tested on the original and enhanced image sets. Then, we performed a correlation study to examine the relationship between enhancement metrics and detection performance. We also analyzed the inference results from the trained detectors presenting cases where enhancement increased the detection performance as well as cases where enhancement revealed missed objects by human annotators. This study suggests that although enhancement generally deteriorates the detection performance, it can still be harnessed in some cases for increased detection performance and more accurate human annotation.

As globular clusters (GCs) orbit the Milky Way, their stars are tidally stripped forming tidal tails that follow the orbit of the clusters around the Galaxy. The morphology of these tails is complex and shows correlations with the phase of the orbit and the orbital angular velocity, especially for GCs on eccentric orbits. Here, we focus on two GCs, NGC 1261 and NGC 1904, that have potentially been accreted alongside Gaia-Enceladus and that have shown signatures of having, in addition of tidal tails, structures formed by distributions of extra-tidal stars that are misaligned with the general direction of the clusters' respective orbits. To provide an explanation for the formation of these structures, we make use of spectroscopic measurements from the Southern Stellar Stream Spectroscopic Survey ($S^5$) as well as proper motion measurements from Gaia's third data release (DR3), and apply a Bayesian mixture modeling approach to isolate high-probability member stars. We recover extra-tidal features similar to those found in Shipp et al. (2018) surrounding each cluster. We conduct N-body simulations and compare the expected distribution and variation in the dynamical parameters along the orbit with those of our potential member sample. Furthermore, we use Dark Energy Camera (DECam) photometry to inspect the distribution of the member stars in the color-magnitude diagram (CMD). We find that the potential members agree reasonably with the N-body simulations and that the majority of them follow a simple stellar population-like distribution in the CMD which is characteristic of GCs. In the case of NGC 1904, we clearly detect the tidal debris escaping the inner and outer Lagrange points which are expected to be prominent when at or close to the apocenter of its orbit. Our analysis allows for further exploration of other GCs in the Milky Way that exhibit similar extra-tidal features.

In the face of increasing demand for zero-touch networks to automate network management and operations, two pivotal concepts have emerged: "Learn to Slice" (L2S) and "Slice to Learn" (S2L). L2S involves leveraging Artificial intelligence (AI) techniques to optimize network slicing for general services, while S2L centers on tailoring network slices to meet the specific needs of various AI services. The complexity of optimizing and automating S2L surpasses that of L2S due to intricate AI services' requirements, such as handling uncontrollable parameters, learning in adversarial conditions, and achieving long-term performance goals. This paper aims to automate and optimize S2L by integrating the two concepts of L2S and S2L by using an intelligent slicing agent to solve S2L. Indeed, we choose two candidate slicing agents, namely the Exploration and Exploitation (EXP3) and Deep Q-Network (DQN) from the Online Convex Optimization (OCO) and Deep Reinforcement Learning (DRL) frameworks, and compare them. Our evaluation involves a series of carefully designed experiments that offer valuable insights into the strengths and limitations of EXP3 and DQN in slicing for AI services, thereby contributing to the advancement of zero-touch network capabilities., Comment: 9 pages, 2 figures and 2 tables magazine paper

Surgical scene segmentation is essential for enhancing surgical precision, yet it is frequently compromised by the scarcity and imbalance of available data. To address these challenges, semantic image synthesis methods based on generative adversarial networks and diffusion models have been developed. However, these models often yield non-diverse images and fail to capture small, critical tissue classes, limiting their effectiveness. In response, we propose the Class-Aware Semantic Diffusion Model (CASDM), a novel approach which utilizes segmentation maps as conditions for image synthesis to tackle data scarcity and imbalance. Novel class-aware mean squared error and class-aware self-perceptual loss functions have been defined to prioritize critical, less visible classes, thereby enhancing image quality and relevance. Furthermore, to our knowledge, we are the first to generate multi-class segmentation maps using text prompts in a novel fashion to specify their contents. These maps are then used by CASDM to generate surgical scene images, enhancing datasets for training and validating segmentation models. Our evaluation, which assesses both image quality and downstream segmentation performance, demonstrates the strong effectiveness and generalisability of CASDM in producing realistic image-map pairs, significantly advancing surgical scene segmentation across diverse and challenging datasets.

This paper introduces a comprehensive framework for Post-Disaster Search and Rescue (PDSR), aiming to optimize search and rescue operations leveraging Unmanned Aerial Vehicles (UAVs). The primary goal is to improve the precision and availability of sensing capabilities, particularly in various catastrophic scenarios. Central to this concept is the rapid deployment of UAV swarms equipped with diverse sensing, communication, and intelligence capabilities, functioning as an integrated system that incorporates multiple technologies and approaches for efficient detection of individuals buried beneath rubble or debris following a disaster. Within this framework, we propose architectural solution and address associated challenges to ensure optimal performance in real-world disaster scenarios. The proposed framework aims to achieve complete coverage of damaged areas significantly faster than traditional methods using a multi-tier swarm architecture. Furthermore, integrating multi-modal sensing data with machine learning for data fusion could enhance detection accuracy, ensuring precise identification of survivors., Comment: This paper is currently under review at IEEE IoT Magazine

The forthcoming 6G networks will embrace a new realm of AI-driven services that requires innovative network slicing strategies, namely slicing for AI, which involves the creation of customized network slices to meet Quality of service (QoS) requirements of diverse AI services. This poses challenges due to time-varying dynamics of users' behavior and mobile networks. Thus, this paper proposes an online learning framework to optimize the allocation of computational and communication resources to AI services, while considering their unique key performance indicators (KPIs), such as accuracy, latency, and cost. We define a problem of optimizing the total accuracy while balancing conflicting KPIs, prove its NP-hardness, and propose an online learning framework for solving it in dynamic environments. We present a basic online solution and two variations employing a pre-learning elimination method for reducing the decision space to expedite the learning. Furthermore, we propose a biased decision space subset selection by incorporating prior knowledge to enhance the learning speed without compromising performance and present two alternatives of handling the selected subset. Our results depict the efficiency of the proposed solutions in converging to the optimal decisions, while reducing decision space and improving time complexity.

Fairness-aware Machine Learning (FairML) applications are often characterized by complex social objectives and legal requirements, frequently involving multiple, potentially conflicting notions of fairness. Despite the well-known Impossibility Theorem of Fairness and extensive theoretical research on the statistical and socio-technical trade-offs between fairness metrics, many FairML tools still optimize or constrain for a single fairness objective. However, this one-sided optimization can inadvertently lead to violations of other relevant notions of fairness. In this socio-technical and empirical study, we frame fairness as a many-objective (MaO) problem by treating fairness metrics as conflicting objectives. We introduce ManyFairHPO, a human-in-the-loop, fairness-aware model selection framework that enables practitioners to effectively navigate complex and nuanced fairness objective landscapes. ManyFairHPO aids in the identification, evaluation, and balancing of fairness metric conflicts and their related social consequences, leading to more informed and socially responsible model-selection decisions. Through a comprehensive empirical evaluation and a case study on the Law School Admissions problem, we demonstrate the effectiveness of ManyFairHPO in balancing multiple fairness objectives, mitigating risks such as self-fulfilling prophecies, and providing interpretable insights to guide stakeholders in making fairness-aware modeling decisions.

Thousands of individuals succumb annually to leukemia alone. This study explores the application of image processing and deep learning techniques for detecting Acute Lymphoblastic Leukemia (ALL), a severe form of blood cancer responsible for numerous annual fatalities. As artificial intelligence technologies advance, the research investigates the reliability of these methods in real-world scenarios. The study focuses on recent developments in ALL detection, particularly using the latest YOLO series models, to distinguish between malignant and benign white blood cells and to identify different stages of ALL, including early stages. Additionally, the models are capable of detecting hematogones, which are often misclassified as ALL. By utilizing advanced deep learning models like YOLOv8 and YOLOv11, the study achieves high accuracy rates reaching 98.8%, demonstrating the effectiveness of these algorithms across multiple datasets and various real-world situations., Comment: 4 pages, 6 figures, 3 tables

The computation of off-diagonal blocks of matrix functions $f(T)$, where $T$ is block triangular, poses a challenging problem in scientific computing. We present a novel algorithm that exploits the structure of block triangular matrices, generalizing the algorithm of Al-Mohy and Higham for computing the Fr\'echet derivative of the matrix exponential. This work has significant applications in fields such as exponential integrators for solving systems of first-order differential equations, Hamiltonian linear systems in control theory, and option pricing in finance. Our approach introduces a linear operator that maps off-diagonal blocks of $T$ into their counterparts in $f(T)$. By studying the algebraic properties of the operator, we establish a comprehensive computational framework, paving the way to extend existing Fr\'echet derivative algorithms of matrix functions to more general settings. For the matrix exponential, in particular, the algorithm employs the scaling and squaring method with diagonal Pad\'e approximants to $\exp(x)$, with parameters chosen based on a rigorous backward error analysis, which notably does not depend on the norm of the off-diagonal blocks. The numerical experiment demonstrates that our algorithm surpasses existing algorithms in terms of accuracy and efficiency, making it highly valuable for a wide range of applications.

Peg-in-hole assembly in unknown environments is a challenging task due to onboard sensor errors, which result in uncertainty and variations in task parameters such as the hole position and orientation. Meta Reinforcement Learning (Meta RL) has been proposed to mitigate this problem as it learns how to quickly adapt to new tasks with different parameters. However, previous approaches either depend on a sample-inefficient procedure or human demonstrations to perform the task in the real world. Our work modifies the data used by the Meta RL agent and uses simple features that can be easily measured in the real world even with an uncalibrated camera. We further adapt the Meta RL agent to use data from a force/torque sensor, instead of the camera, to perform the assembly, using a small amount of training data. Finally, we propose a fine-tuning method that consistently and safely adapts to out-of-distribution tasks with parameters that differ by a factor of 10 from the training tasks. Our results demonstrate that the proposed data modification significantly enhances the training and adaptation efficiency and enables the agent to achieve 100% success in tasks with different hole positions and orientations. Experiments on a real robot confirm that both camera- and force/torque sensor-equipped agents achieve 100% success in tasks with unknown hole positions, matching their simulation performance and validating the approach's robustness and applicability. Compared to the previous work with sample-inefficient adaptation, our proposed methods are 10 times more sample-efficient in the real-world tasks., Comment: 8 pages, 9 figures, submitted to ICRA 2025

By exciting a series of $1\mathrm{s}^{2}\, ^{1}\mathrm{S}_{0} \to 1\mathrm{s}n\mathrm{p}\, ^{1}\mathrm{P}_{1}$ transitions in helium-like nitrogen ions with linearly polarized monochromatic soft X-rays at the Elettra facility, we found a change in the angular distribution of the fluorescence sensitive to the principal quantum number $n$. In particular it is observed that the ratio of emission in directions parallel and perpendicular to the polarization of incident radiation increases with higher $n$. We find this $n$-dependence to be a manifestation of the Hanle effect, which served as a practical tool for lifetime determinations of optical transitions since its discovery in 1924. In contrast to traditional Hanle effect experiments, in which one varies the magnetic field and considers a particular excited state, we demonstrate a 'soft X-ray Hanle effect' which arises in a static magnetic field but for a series of excited states. By comparing experimental data with theoretical predictions, we were able to determine lifetimes ranging from hundreds of femtoseconds to tens of picoseconds of the $1\mathrm{s}n\mathrm{p}\, ^{1}\mathrm{P}_{1}$ levels, which find excellent agreement with atomic-structure calculations. We argue that dedicated soft X-ray measurements could yield lifetime data that is beyond current experimental reach and cannot yet be predicted with sufficient accuracy., Comment: 7 pages, 4 figures

Tactile sensing represents a crucial technique that can enhance the performance of robotic manipulators in various tasks. This work presents a novel bioinspired neuromorphic vision-based tactile sensor that uses an event-based camera to quickly capture and convey information about the interactions between robotic manipulators and their environment. The camera in the sensor observes the deformation of a flexible skin manufactured from a cheap and accessible 3D printed material, whereas a 3D printed rigid casing houses the components of the sensor together. The sensor is tested in a grasping stage classification task involving several objects using a data-driven learning-based approach. The results show that the proposed approach enables the sensor to detect pressing and slip incidents within a speed of 2 ms. The fast tactile perception properties of the proposed sensor makes it an ideal candidate for safe grasping of different objects in industries that involve high-speed pick-and-place operations., Comment: 9 pages, 10 figures, journal

High-dimensional action spaces remain a challenge for dynamic algorithm configuration (DAC). Interdependencies and varying importance between action dimensions are further known key characteristics of DAC problems. We argue that these Coupled Action Dimensions with Importance Differences (CANDID) represent aspects of the DAC problem that are not yet fully explored. To address this gap, we introduce a new white-box benchmark within the DACBench suite that simulates the properties of CANDID. Further, we propose sequential policies as an effective strategy for managing these properties. Such policies factorize the action space and mitigate exponential growth by learning a policy per action dimension. At the same time, these policies accommodate the interdependence of action dimensions by fostering implicit coordination. We show this in an experimental study of value-based policies on our new benchmark. This study demonstrates that sequential policies significantly outperform independent learning of factorized policies in CANDID action spaces. In addition, they overcome the scalability limitations associated with learning a single policy across all action dimensions. The code used for our experiments is available under https://github.com/PhilippBordne/candidDAC., Comment: 5 pages main paper, 11 pages references and appendix, 9 figures, to be published in: Proceedings of the Third International Conference on Automated Machine Learning (AutoML 2024), Workshop Track

Machine Learning systems are increasingly prevalent across healthcare, law enforcement, and finance but often operate on historical data, which may carry biases against certain demographic groups. Causal and counterfactual fairness provides an intuitive way to define fairness that closely aligns with legal standards. Despite its theoretical benefits, counterfactual fairness comes with several practical limitations, largely related to the reliance on domain knowledge and approximate causal discovery techniques in constructing a causal model. In this study, we take a fresh perspective on counterfactually fair prediction, building upon recent work in in context learning (ICL) and prior fitted networks (PFNs) to learn a transformer called FairPFN. This model is pretrained using synthetic fairness data to eliminate the causal effects of protected attributes directly from observational data, removing the requirement of access to the correct causal model in practice. In our experiments, we thoroughly assess the effectiveness of FairPFN in eliminating the causal impact of protected attributes on a series of synthetic case studies and real world datasets. Our findings pave the way for a new and promising research area: transformers for causal and counterfactual fairness.

Our objectives are to map the filamentary network around the Fornax-Eridanus Complex and probe the influence of the local environment on galaxy morphology. We employ the novel machine-learning tool, 1-DREAM (1-Dimensional, Recovery, Extraction, and Analysis of Manifolds) to detect and model filaments around the Fornax cluster. We then use the morphology-density relation of galaxies to examine the variation in the galaxies' morphology with respect to their distance from the central axis of the detected filaments. We detect 27 filaments that vary in length and galaxy-number density around the Fornax-Eridanus Complex. These filaments showcase a variety of environments; some filaments encompass groups/clusters, while others are only inhabited by galaxies in pristine filamentary environments. We also reveal a well-known structure -- the Fornax Wall, that passes through the Dorado group, Fornax cluster, and Eridanus supergroup. Regarding the morphology of galaxies, we find that early-type galaxies (ETGs) populate high-density filaments and high-density regions of the Fornax Wall. Furthermore, the fraction of ETGs decreases as the distance to the filament spine increases. Of the total galaxy population in filaments, ~7% are ETGs and ~24% are late-type galaxies (LTGs) located in pristine environments of filaments, while ~27% are ETGs and ~42% are LTGs in groups/clusters within filaments. This study reveals the Cosmic Web around the Fornax Cluster and asserts that filamentary environments are heterogeneous in nature. When investigating the role of the environment on galaxy morphology, it is essential to consider both, the local number-density and a galaxy's proximity to the filament spine. Within this framework, we ascribe the observed morphological segregation in the Fornax Wall to pre-processing of galaxies within groups embedded in it., Comment: Accepted for publication in A&A. 21 pages with 15 figures

A novel method to reconstruct the energy of hadronic showers in the CMS High Granularity Calorimeter (HGCAL) is presented. The HGCAL is a sampling calorimeter with very fine transverse and longitudinal granularity. The active media are silicon sensors and scintillator tiles readout by SiPMs and the absorbers are a combination of lead and Cu/CuW in the electromagnetic section, and steel in the hadronic section. The shower reconstruction method is based on graph neural networks and it makes use of a dynamic reduction network architecture. It is shown that the algorithm is able to capture and mitigate the main effects that normally hinder the reconstruction of hadronic showers using classical reconstruction methods, by compensating for fluctuations in the multiplicity, energy, and spatial distributions of the shower's constituents. The performance of the algorithm is evaluated using test beam data collected in 2018 prototype of the CMS HGCAL accompanied by a section of the CALICE AHCAL prototype. The capability of the method to mitigate the impact of energy leakage from the calorimeter is also demonstrated.

Process mining extracts valuable insights from event data to help organizations improve their business processes, which is essential for their growth and success. By leveraging process mining techniques, organizations gain a comprehensive understanding of their processes' execution, enabling the discovery of process models, detection of deviations, identification of bottlenecks, and assessment of performance. Compliance checking, a specific area within conformance checking, ensures that the organizational activities adhere to prescribed process models and regulations. Linear Temporal Logic over finite traces ($LTL_{f}$ ) is commonly used for conformance checking, but it may not capture all temporal aspects accurately. This paper proposes Metric Temporal Logic over finite traces ($MTL_{f}$ ) to define explicit time-related constraints effectively in addition to the implicit time-ordering covered by $LTL_f$. Therefore, it provides a universal formal approach to capture compliance rules. Moreover, we define a minimal set of generic $MTL_f$ formulas and show that they are capable of capturing all the common patterns for compliance rules. As compliance validation is largely driven by the data model used to represent the event logs, we provide a mapping from $MTL_f$ to the common data models we found in the literature to encode event logs, namely, the relational and the graph models. A comprehensive study comparing various data models and an empirical evaluation across real-life event logs demonstrates the effectiveness of the proposed approach.

Optical heterodyne detection is a powerful technique for characterizing a wide range of physical excitations. Here, we use two types of optical heterodyne detection techniques (fundamental and parametric pumping) to microscopically characterize the high-frequency auto-oscillations of single and multiple nano-constriction spin Hall nano-oscillators (SHNOs). To validate the technique and demonstrate its robustness, we study SHNOs made from two different material stacks, NiFe/Pt and W/CoFeB/MgO, and investigate the influence of both the RF injection power and the laser power on the measurements, comparing the optical results to conventional electrical measurements. To demonstrate the key features of direct, non-invasive, submicron, spatial, and phase-resolved characterization of the SHNO magnetodynamics, we map out the auto-oscillation magnitude and phase of two phase-binarized SHNOs used in Ising Machines. This proof-of-concept platform establishes a strong foundation for further extensions, contributing to the ongoing development of crucial characterization techniques for emerging computing technologies based on spintronics devices, Comment: 7 pages

The Euclid ERO programme targeted the Perseus cluster of galaxies, gathering deep data in the central region of the cluster over 0.7 square degree, corresponding to approximately 0.25 r_200. The data set reaches a point-source depth of IE=28.0 (YE, JE, HE = 25.3) AB magnitudes at 5 sigma with a 0.16" and 0.48" FWHM, and a surface brightness limit of 30.1 (29.2) mag per square arcsec. The exceptional depth and spatial resolution of this wide-field multi-band data enable the simultaneous detection and characterisation of both bright and low surface brightness galaxies, along with their globular cluster systems, from the optical to the NIR. This study advances beyond previous analyses of the cluster and enables a range of scientific investigations summarised here. We derive the luminosity and stellar mass functions (LF and SMF) of the Perseus cluster in the Euclid IE band, thanks to supplementary u,g,r,i,z and Halpha data from the CFHT. We adopt a catalogue of 1100 dwarf galaxies, detailed in the corresponding ERO paper. We identify all other sources in the Euclid images and obtain accurate photometric measurements using AutoProf or AstroPhot for 138 bright cluster galaxies, and SourceExtractor for half a million compact sources. Cluster membership for the bright sample is determined by calculating photometric redshifts with Phosphoros. Our LF and SMF are the deepest recorded for the Perseus cluster, highlighting the groundbreaking capabilities of the Euclid telescope. Both the LF and SMF fit a Schechter plus Gaussian model. The LF features a dip at M(IE)=-19 and a faint-end slope of alpha_S = -1.2 to -1.3. The SMF displays a low-mass-end slope of alpha_S = -1.2 to -1.35. These observed slopes are flatter than those predicted for dark matter halos in cosmological simulations, offering significant insights for models of galaxy formation and evolution., Comment: Submitted to A&A, 44 pages, 35 figures, Part of the A&A special issue `Euclid on Sky', which contains Euclid key reference papers and first results from the Euclid Early Release Observations

The Euclid ERO showcase Euclid's capabilities in advance of its main mission, targeting 17 astronomical objects, from galaxy clusters, nearby galaxies, globular clusters, to star-forming regions. A total of 24 hours observing time was allocated in the early months of operation, engaging the scientific community through an early public data release. We describe the development of the ERO pipeline to create visually compelling images while simultaneously meeting the scientific demands within months of launch, leveraging a pragmatic, data-driven development strategy. The pipeline's key requirements are to preserve the image quality and to provide flux calibration and photometry for compact and extended sources. The pipeline's five pillars are: removal of instrumental signatures; astrometric calibration; photometric calibration; image stacking; and the production of science-ready catalogues for both the VIS and NISP instruments. We report a PSF with a full width at half maximum of 0.16" in the optical and 0.49" in the three NIR bands. Our VIS mean absolute flux calibration is accurate to about 1%, and 10% for NISP due to a limited calibration set; both instruments have considerable colour terms. The median depth is 25.3 and 23.2 AB mag with a SNR of 10 for galaxies, and 27.1 and 24.5 AB mag at an SNR of 5 for point sources for VIS and NISP, respectively. Euclid's ability to observe diffuse emission is exceptional due to its extended PSF nearly matching a pure diffraction halo, the best ever achieved by a wide-field, high-resolution imaging telescope. Euclid offers unparalleled capabilities for exploring the LSB Universe across all scales, also opening a new observational window in the NIR. Median surface-brightness levels of 29.9 and 28.3 AB mag per square arcsec are achieved for VIS and NISP, respectively, for detecting a 10 arcsec x 10 arcsec extended feature at the 1 sigma level., Comment: Submitted to A&A, 44 pages, 36 figures - Part of the A&A special issue `Euclid on Sky', which contains Euclid key reference papers and first results from the Euclid Early Release Observations

We have learnt in the last decades that the majority of galaxies belong to high density regions interconnected in a sponge-like fashion. This large-scale structure is characterised by clusters, filaments, walls, where most galaxies concentrate, but also under-dense regions, called voids. The void regions and the galaxies within represent an ideal place for the study of galaxy formation and evolution as they are largely unaffected by the complex physical processes that transform galaxies in high-density environments. These void galaxies can hold the key as well to answer current challenges to the $\Lambda$CDM paradigm. The Calar Alto Void Integral-field Treasury surveY (CAVITY) is a Legacy project approved by the Calar Alto Observatory to obtain spatially resolved spectroscopic information of $\sim300$ void galaxies in the Local Universe (0.005 < z < 0.050) covering from -17.0 to -21.5 in $\rm r$ band absolute magnitude. It officially started in January 2021 and has been awarded 110 useful dark observing nights at the 3.5 m telescope using the PMAS spectrograph. Complementary follow-up projects including deep optical imaging, integrated, as well as resolved CO data, and integrated HI spectra, have joint the PMAS observations and naturally complete the scientific aim of characterising galaxies in cosmic voids. The extension data has been denominated CAVITY+. The data will be available to the whole community in different data releases, the first of which is planned for July 2024, and it will provide the community with PMAS data cubes for around 100 void galaxies through a user friendly, and well documented, database platform. We present here the survey, sample selection, data reduction, quality control schemes, science goals, and some examples of the scientific power of the CAVITY and CAVITY+ data., Comment: 19 pages, 11 figures. Accepted for publication in A&A - Replacement after A&A minor language edition

The rapid advancement in artificial intelligence and natural language processing has led to the development of large-scale datasets aimed at benchmarking the performance of machine learning models. Herein, we introduce 'RetChemQA,' a comprehensive benchmark dataset designed to evaluate the capabilities of such models in the domain of reticular chemistry. This dataset includes both single-hop and multi-hop question-answer pairs, encompassing approximately 45,000 Q&As for each type. The questions have been extracted from an extensive corpus of literature containing about 2,530 research papers from publishers including NAS, ACS, RSC, Elsevier, and Nature Publishing Group, among others. The dataset has been generated using OpenAI's GPT-4 Turbo, a cutting-edge model known for its exceptional language understanding and generation capabilities. In addition to the Q&A dataset, we also release a dataset of synthesis conditions extracted from the corpus of literature used in this study. The aim of RetChemQA is to provide a robust platform for the development and evaluation of advanced machine learning algorithms, particularly for the reticular chemistry community. The dataset is structured to reflect the complexities and nuances of real-world scientific discourse, thereby enabling nuanced performance assessments across a variety of tasks. The dataset is available at the following link: https://github.com/nakulrampal/RetChemQA