Search

The stellar age and mass of galaxies have been suggested as the primary determinants for the dynamical state of galaxies, with environment seemingly playing no or only a very minor role. We use a sample of 77 galaxies at intermediate redshift (z~0.3) in the Middle-Ages Galaxies Properties with Integral field spectroscopy (MAGPI) Survey to study the subtle impact of environment on galaxy dynamics. We use a combination of statistical techniques (simple and partial correlations and principal component analysis) to isolate the contribution of environment on galaxy dynamics, while explicitly accounting for known factors such as stellar age, star formation histories and stellar masses. We consider these dynamical parameters: high-order kinematics of the line-of-sight velocity distribution (parametrised by the Gauss-Hermite coefficients $h_3$ and $h_4$), kinematic asymmetries $V_{\rm asym}$ derived using kinemetry and the observational spin parameter proxy $\lambda_{R_e}$. Of these, the mean $h_4$ is the only parameter found to have a significant correlation with environment as parametrised by group dynamical mass. This correlation exists even after accounting for age and stellar mass trends. Finally, we confirm that variations in the spin parameter $\lambda_{R_e}$ are most strongly (anti-)correlated with age as seen in local studies, and show that this dependence is well-established by z~0.3., Comment: 17 pages, 13 figures, 3 tables, accepted for publication in PASA

Since 2021, unprecedented increases in energy costs, rapid increases in food costs and substantial increases in the cost of housing via higher rent/mortgage costs have driven a dramatic rise in the cost of living across England. Despite inflation having fallen from its peak in winter 2022, costs remain high compared to incomes and cost pressures are expected to persist for years to come. These cost-of-living pressures have had a profound impact on pupils and their families, alongside directly impacting on the expenditures of schools. Drawing on an online survey of over 1200 teachers and senior leaders in mainstream schools in March 2024, this research builds on previous research into the impact of the cost of living and explores how cost-of-living pressures, alongside wider financial pressures, are impacting schools a year on. This research aims to establish: (1) has the level of need among pupils changed since last year?; (2) what support are schools providing to pupils and how has this changed over the last year?; and (3) what is the ongoing impact of cost pressures, including cost of living increases, on schools' provision and financial positions? The findings highlight the scale of the pressures impacting families and schools which has entrenched high levels of need among pupils. As a result, schools are continuing to take unprecedented steps to meet need. The level of unmet need among pupils, coupled with the significant resourcing pressures affecting schools, is likely to impact on pupil attainment outcomes in both the short and long term and widen longstanding gaps between disadvantaged pupils and their more advantaged peers.

Neutrino quantum kinetics is a rapidly evolving field in computational astrophysics, with a primary focus on collective neutrino oscillations in core-collapse supernovae and post-merger phases of binary neutron star mergers. In recent years, there has been considerable debate concerning resolution dependence in numerical simulations. In this paper, we conduct a comprehensive resolution study in both angular- and spatial directions by using two independent schemes of quantum kinetic neutrino transport: finite volume and pseudospectral methods. We complement our discussion by linear stability analysis including inhomogeneous modes. Our result suggests that decreasing spatial resolutions underestimates the growth of flavor instability, and then leads to wrong asymptotic states of flavor conversions, which potentially has a critical impact on astrophysical consequences. We further delve into numerical results of low resolution simulations, that reveals the underlying mechanism responsible for numerical artifacts caused by insufficient resolutions. This study settles the debate on requirements of resolutions and serves as a guideline for numerical modeling of quantum kinetic neutrino transport., Comment: 11 pages, 10 figures

Fast radio bursts (FRBs) are very energetic pulses of unknown physical origin. These can be used to study the intergalactic medium (IGM) thanks to their dispersion measure (DM). The DM has several contributions that can be measured (or estimated), including the contribution from the host galaxy itself, DM_host. In this work, we empirically estimate DM_host for a sample of 12 galaxy hosts, using a direct method based solely on the properties of the host galaxies themselves (DM_host_dir). We use VLT/MUSE observations of the FRB hosts for estimating DM_host_dir. The method relies on estimating the DM contribution of both the FRB host galaxy's interstellar medium and its halo separately. For comparison purposes, we also provide an alternative indirect method to estimate DM_host based on the Macquart relation (DM_host_mq). We find an average = 80+/-11 pc/cc with a standard deviation of 38 pc/cc (in the rest-frame) based on our direct method, with a systematic uncertainty of 30%. We report positive correlations between DM_host and both the stellar masses and the star-formation rates of their host galaxies. In contrast, we do not find any strong correlation between DM_host and neither redshift nor the projected distances to the FRB hosts centers. Finally, we do not find any strong correlation between DM_host_dir and DM_host_mq, although their average values are consistent. Our reported correlations could be used to improve the priors used in establishing DM_host for individual FRBs. Similarly, such correlations and the lack of a strong redshift evolution can be used to constrain models for the progenitor of FRBs. However, the lack of a DM_host_dir and DM_host_mq correlation indicates that there may still be contributions to the DM of FRBs not included in our modeling, e.g. large DMs from the FRB progenitor and/or intervening large-scale structures not accounted for in DM_host_mq., Comment: Accepted for publication in Astronomy & Astrophysics on 20 January, 2025. Abstract has been abridged

Several ground- and space-based searches have increased the number of known exoplanets to nearly 6000. While most are highly unlike our planet, the first rocky worlds in the Habitable Zone (HZ) provide the most intriguing targets for the search for life in the cosmos. As the detections increase, it is critical to probe the HZ as well as its limits empirically using known exoplanets. However, there is not yet a list of rocky worlds that observers can use to probe the limits of surface habitability. We analyze all known exoplanets and identify 67 rocky worlds in the empirical HZ and 38 in a narrower 3D-model HZ. We compare their demographics with the full catalog of exoplanets, analyze their characteristics, give HZ limits for each system, and prioritize targets for observation. We address another critical dimension to this exploration and compile missing stellar age estimates for these candidates. To probe the limits of habitability, we identify several transiting and non-transiting rocky planets that can provide constraints on the inner and outer edge of the HZ, explore how eccentricity influences habitability, and identify the oldest HZ worlds as well as exoplanets with similar flux to modern Earth's. The resulting list of rocky exoplanet targets in the HZ will allow observers to shape and optimize their search strategies with space- and ground-based telescopes -- such as the James Webb Space Telescope, the Extremely Large Telescope, the Habitable Worlds Observatory, and LIFE -- and to design new observing strategies and instruments to explore these intriguing worlds, addressing the question of the limits of surface habitability on exoplanets., Comment: 9 pages, 6 figures, submitted to MNRAS, shared first authorship between A. Bohl, L. Lawrence, and G. Lowry, data will be available on zenodo upon publication

We investigate the pseudospectrum of a Schwarzschild-like spacetime within the framework of black hole perturbation theory to analyze a counterintuitive assertion regarding the instability of quasinormal modes. Recent findings suggest that random perturbations to the effective potential associated with gravitational waves may enhance the stability of the underlying wave operator, thereby yielding a stable spectrum of randomly displaced quasinormal modes. Given the unphysical nature of such random perturbations, this work examines these findings within a spacetime that inherently exhibits a perturbed quasinormal spectrum. We find that, in contrast to the QNM spectrum of the Schwarzschild spacetime under random perturbations, the quasinormal spectrum of a Schwarzschild-like black hole deformed through a physically motivated implementation of the Rezzolla-Zhidenko parametrization is unstable. In particular, we show that the pseudospectra of these Schwarzschild-like black holes do not display the typical features associated with wave operators that yield stable quasinormal spectra. We corroborate our findings by computing the quasinormal spectra when additional (ad-hoc) deformations are added to the effective potential of the Rezzolla-Zhidenko black hole. We also argue that when multiple perturbation sources are present, identifying the origin of the instability may be difficult., Comment: v1:12 pages, 4 figures

Text-to-SQL prompt strategies based on Large Language Models (LLMs) achieve remarkable performance on well-known benchmarks. However, when applied to real-world databases, their performance is significantly less than for these benchmarks, especially for Natural Language (NL) questions requiring complex filters and joins to be processed. This paper then proposes a strategy to compile NL questions into SQL queries that incorporates a dynamic few-shot examples strategy and leverages the services provided by a database keyword search (KwS) platform. The paper details how the precision and recall of the schema-linking process are improved with the help of the examples provided and the keyword-matching service that the KwS platform offers. Then, it shows how the KwS platform can be used to synthesize a view that captures the joins required to process an input NL question and thereby simplify the SQL query compilation step. The paper includes experiments with a real-world relational database to assess the performance of the proposed strategy. The experiments suggest that the strategy achieves an accuracy on the real-world relational database that surpasses state-of-the-art approaches. The paper concludes by discussing the results obtained.

Fixed-node diffusion quantum Monte Carlo (FN-DMC) is a widely-trusted many-body method for solving the Schr\"{o}dinger equation, known for its reliable predictions of material and molecular properties. Furthermore, its excellent scalability with system complexity and near-perfect utilization of computational power makes FN-DMC ideally positioned to leverage new advances in computing to address increasingly complex scientific problems. Even though the method is widely used as a computational gold standard, reproducibility across the numerous FN-DMC code implementations has yet to be demonstrated. This difficulty stems from the diverse array of DMC algorithms and trial wave functions, compounded by the method's inherent stochastic nature. This study represents a community-wide effort to address the titular question, affirming that: Yes, FN-DMC is reproducible (when handled with care). Using the water-methane dimer as the canonical test case, we compare results from eleven different FN-DMC codes and show that the approximations to treat the non-locality of pseudopotentials are the primary source of the discrepancies between them. In particular, we demonstrate that, for the same choice of determinantal component in the trial wave function, reliable and reproducible predictions can be achieved by employing the T-move (TM), the determinant locality approximation (DLA), or the determinant T-move (DTM) schemes, while the older locality approximation (LA) leads to considerable variability in results. This work lays the foundation to establish accurate and reproducible FN-DMC estimates for all future studies across applications in materials science, physics, chemistry, and biology.

We present measurements of large-scale cosmic microwave background (CMB) E-mode polarization from the Cosmology Large Angular Scale Surveyor (CLASS) 90 GHz data. Using 115 det-yr of observations collected through 2024 with a variable-delay polarization modulator, we achieved a polarization sensitivity of $78\,\mathrm{\mu K\,arcmin}$, comparable to Planck at similar frequencies (100 and 143 GHz). The analysis demonstrates effective mitigation of systematic errors and addresses challenges to large-angular-scale power recovery posed by time-domain filtering in maximum-likelihood map-making. A novel implementation of the pixel-space transfer matrix is introduced, which enables efficient filtering simulations and bias correction in the power spectrum using the quadratic cross-spectrum estimator. Overall, we achieved an unbiased time-domain filtering correction to recover the largest angular scale polarization, with the only power deficit, arising from map-making non-linearity, being characterized as less than $3\%$. Through cross-correlation with Planck, we detected the cosmic reionization at $99.4\%$ significance and measured the reionization optical depth $\tau=0.053^{+0.018}_{-0.019}$, marking the first ground-based attempt at such a measurement. At intermediate angular scales ($\ell>30$), our results, both independently and in cross-correlation with Planck, remain fully consistent with Planck's measurements., Comment: 24 pages, 19 figures, 3 tables; submitted to ApJ

Lean R&D has been used at PUC-Rio to foster industry-academia collaboration in innovation projects across multiple sectors. This industrial experience paper describes recent experiences and evaluation results from applying Lean R&D in partnership with Petrobras in the oil and gas sector and Americanas in retail. The findings highlight Lean R&D's effectiveness in transforming ideas into meaningful business outcomes. Based on responses from 57 participants - including team members, managers, and sponsors - the assessment indicates that stakeholders find the structured phases of Lean R&D well-suited to innovation projects and endorse the approach. Although acknowledging that successful collaboration relies on various factors, this industrial experience positions Lean R&D as a promising framework for industry-academia projects focused on achieving rapid, impactful results for industry partners.

The problem of sampling edge-colorings of graphs with maximum degree $\Delta$ has received considerable attention and efficient algorithms are available when the number of colors is large enough with respect to $\Delta$. Vizing's theorem guarantees the existence of a $(\Delta+1)$-edge-coloring, raising the natural question of how to efficiently sample such edge-colorings. In this paper, we take an initial step toward addressing this question. Building on the approach of Dotan, Linial, and Peled, we analyze a randomized algorithm for generating random proper $(\Delta+1)$-edge-colorings, which in particular provides an algorithmic interpretation of Vizing's theorem. The idea is to start from an arbitrary non-proper edge-coloring with the desired number of colors and at each step, recolor one edge uniformly at random provided it does not increase the number of conflicting edges (a potential function will count the number of pairs of adjacent edges of the same color). We show that the algorithm almost surely produces a proper $(\Delta+1)$-edge-coloring and propose several conjectures regarding its efficiency and the uniformity of the sampled colorings., Comment: 11 pages, 7 figures

This paper combines two significant areas of political science research: measuring individual ideological position and cohesion. Although both approaches help analyze legislative behaviors, no unified model currently integrates these dimensions. To fill this gap, the paper proposes a methodology called B-Call that combines ideological positioning with voting cohesion, treating votes as random variables. The model is empirically validated using roll-call data from the United States, Brazil, and Chile legislatures, which represent diverse legislative dynamics. The analysis aims to capture the complexities of voting and legislative behaviors, resulting in a two-dimensional indicator. This study addresses gaps in current legislative voting models, particularly in contexts with limited party control., Comment: 23 pages

We present Fanar, a platform for Arabic-centric multimodal generative AI systems, that supports language, speech and image generation tasks. At the heart of Fanar are Fanar Star and Fanar Prime, two highly capable Arabic Large Language Models (LLMs) that are best in the class on well established benchmarks for similar sized models. Fanar Star is a 7B (billion) parameter model that was trained from scratch on nearly 1 trillion clean and deduplicated Arabic, English and Code tokens. Fanar Prime is a 9B parameter model continually trained on the Gemma-2 9B base model on the same 1 trillion token set. Both models are concurrently deployed and designed to address different types of prompts transparently routed through a custom-built orchestrator. The Fanar platform provides many other capabilities including a customized Islamic Retrieval Augmented Generation (RAG) system for handling religious prompts, a Recency RAG for summarizing information about current or recent events that have occurred after the pre-training data cut-off date. The platform provides additional cognitive capabilities including in-house bilingual speech recognition that supports multiple Arabic dialects, voice and image generation that is fine-tuned to better reflect regional characteristics. Finally, Fanar provides an attribution service that can be used to verify the authenticity of fact based generated content. The design, development, and implementation of Fanar was entirely undertaken at Hamad Bin Khalifa University's Qatar Computing Research Institute (QCRI) and was sponsored by Qatar's Ministry of Communications and Information Technology to enable sovereign AI technology development.

This paper investigates the coexistence of two competing species on random geometric graphs (RGGs) in continuous time. The species grow by occupying vacant sites according to Richardson's model, while simultaneously competing for occupied sites under the dynamics of the voter model. Coexistence is defined as the event in which both species occupy at least one site simultaneously at any given time. We prove that coexistence occurs with strictly positive annealed probability by applying results from moderate deviations in first-passage percolation and random walk theory, with a focus on specific regions of the space., Comment: 16 pages, 5 figures

The 3rd Workshop on Maritime Computer Vision (MaCVi) 2025 addresses maritime computer vision for Unmanned Surface Vehicles (USV) and underwater. This report offers a comprehensive overview of the findings from the challenges. We provide both statistical and qualitative analyses, evaluating trends from over 700 submissions. All datasets, evaluation code, and the leaderboard are available to the public at https://macvi.org/workshop/macvi25., Comment: Part of the MaCVi 2025 workshop

We study the complexity of the Virtual Network Embedding Problem (VNE), which is the combinatorial core of several telecommunication problems related to the implementation of virtualization technologies, such as Network Slicing. VNE is to find an optimal assignment of virtual demands to physical resources, encompassing simultaneous placement and routing decisions. The problem is known to be strongly NP-hard, even when the virtual network is a uniform path, but is polynomial in some practical cases. This article aims to draw a cohesive frontier between easy and hard instances for VNE. For this purpose, we consider uniform demands to focus on structural aspects, rather than packing ones. To this end, specific topologies are studied for both virtual and physical networks that arise in practice, such as trees, cycles, wheels and cliques. Some polynomial greedy or dynamic programming algorithms are proposed, when the physical network is a tree or a cycle, whereas other close cases are shown NP-hard.

The physical and orbital parameters of Trans-Neptunian Objects (TNOs) provide valuable information about the Solar System's formation and evolution. In particular, the characterization of binaries provides insights into the formation mechanisms that may be playing a role at such large distances from the Sun. Studies show two distinct populations, and (38628) Huya occupies an intermediate position between the unequal-size binaries and those with components of roughly equal sizes. In this work, we predicted and observed three stellar occultation events by Huya. Huya and its satellite - S/2012 (38628) 1 - were detected during occultations in March 2021 and again in June 2023. Additionally, an attempt to detect Huya in February 2023 resulted in an additional single-chord detection of the secondary. A spherical body with a minimum diameter of D = 165 km can explain the three single-chord observations and provide a lower limit for the satellite size. The astrometry of Huya's system, as derived from the occultations and supplemented by observations from the Hubble Space Telescope and Keck Observatory, provided constraints on the satellite orbit and the mass of the system. Therefore, assuming the secondary is in an equatorial orbit around the primary, the limb fitting was constrained by the satellite orbit position angle. The system density, calculated by summing the most precise measurement of Huya's volume to the spherical satellite average volume, is $\rho_{1}$ = 1073 $\pm$ 66 kg m$^{-3}$. The density that the object would have assuming a Maclaurin equilibrium shape with a rotational period of 6.725 $\pm$ 0.01 hours is $\rho_{2}$ = 768 $\pm$ 42 kg m$^{-3}$. This difference rules out the Maclaurin equilibrium assumption for the main body shape., Comment: Manuscript accepted by Planetary Science Journal (PSJ)

Group theory has been used in machine learning to provide a theoretically grounded approach for incorporating known symmetry transformations in tasks from robotics to protein modeling. In these applications, equivariant neural networks use known symmetry groups with predefined representations to learn over geometric input data. We propose MatrixNet, a neural network architecture that learns matrix representations of group element inputs instead of using predefined representations. MatrixNet achieves higher sample efficiency and generalization over several standard baselines in prediction tasks over the several finite groups and the Artin braid group. We also show that MatrixNet respects group relations allowing generalization to group elements of greater word length than in the training set., Comment: NeurIPS 2024

We establish new convergence rates for the moment-sum-of-squares (Moment-SOS) relaxations for the Generalized Moment Problem (GMP). These bounds, which adapt to the geometry of the underlying semi-algebraic set, apply to both the convergence of optima, and to the convergence in Hausdorff distance between the relaxation feasibility set and the GMP feasibility set. This research extends previous works limited to specific problems in polynomial optimization, volume computation and optimal control. We complement our theoretical analysis with an application: minimal rank symmetric tensor decomposition. In the examples, we formulate the problem as a GMP, solve using Moment-SOS relaxation, and apply the theoretical results to observe a convergence rate of the relaxations.

Graph APIs are capable of flexibly retrieving or manipulating graph-structured data over the web. This rather novel type of APIs presents new challenges when it comes to properly securing the APIs against the usual web application security risks, e.g., broken access control. A prominent security testing approach is taint analysis, which traces tainted, i.e., security-relevant, data from sources (where tainted data is inserted) to sinks (where the use of tainted data may lead to a security risk), over the information flow in an application. We present a first systematic approach to static and dynamic taint analysis for Graph APIs focusing on broken access control. The approach comprises the following. We taint nodes in the Graph API if they represent data requiring specific privileges in order to be retrieved or manipulated, and identify API calls which are related to sources and sinks. Then, we statically analyze whether tainted information flow between API source and sink calls occurs. To this end, we model the API calls using graph transformation rules. We subsequently use critical pair analysis to automatically analyze potential dependencies between rules representing source calls and rules representing sink calls. We distinguish direct from indirect tainted information flow and argue under which conditions the CPA is able to detect not only direct, but also indirect tainted flow. The static taint analysis (i) identifies flows that need to be further reviewed, since tainted nodes may be created by an API call and used or manipulated by another API call later without having the necessary privileges, and (ii) can be used to systematically design dynamic security tests for broken access control. The dynamic taint analysis checks if potential broken access control risks detected during the static taint analysis really occur. We apply the approach to a part of the GitHub GraphQL API., Comment: Intermediate preprint for submission to ICGT 24 Special Issue in LMCS

The Simons Observatory (SO) is a ground-based cosmic microwave background (CMB) survey experiment that currently consists of three 0.42m small-aperture telescopes (SATs) and one 6m large-aperture telescope (LAT), located at an elevation of 5200m in the Atacama Desert in Chile. At the LAT's focal plane, SO will install >62,000 transition-edge sensor detectors across 13 optics tubes (OTs) within the Large Aperture Telescope Receiver (LATR), the largest cryogenic camera ever built to observe the CMB. Here we report on the validation of the LATR in the laboratory and the subsequent dark testing and validation within the LAT. We show that the LATR meets cryogenic, optical, and detector specifications required for high-sensitivity measurements of the CMB. At the time of writing, the LATR is installed in the LAT with six OTs (corresponding to >31,000 detectors), and the LAT mirrors and remaining seven OTs are undergoing development.

MPS-VG, a Virtual Grating (VG) model for the Lagrangian mesh-free Moving Particle Semi-implicit (MPS) method is proposed for replacing conventional particle-based solid modeling of gratings with a set of thin inclined slats. Unlike most approaches for perforated wave energy dampening devices, in which the flow through the device is simplified by pressure loss or damping effects without flow deflection, MPS-VG models the angular deviation caused by hydrodynamic impact on inclined slats. Both accuracy and computational performance of the model were checked through a simulation of wet dam break scenarios with the grating structures placed horizontally or vertically. The results were compared with those from fully particle-based modeling. MPS-VG correctly predicted complex wave-structure interactions using a relatively low-resolution model and significantly reduced processing time and memory storage compared to conventional particle-based MPS modeling. The evaluation of the performance of the gratings with inclined slats as wave energy dampers revealed the horizontal gratings outperformed the vertical ones. Therefore, qualitative and quantitative agreements strengthened the potential of MPS-VG as a practical and computationally efficient tool for the study of multi-scale phenomena of wave impacts on grating with inclined slats., Comment: 39, 26 figures

The increasing complexity and volume of network data demand effective analysis approaches, with visual exploration proving particularly beneficial. Immersive technologies, such as augmented reality, virtual reality, and large display walls, have enabled the emerging field of immersive analytics, offering new opportunities to enhance user engagement, spatial awareness, and problem-solving. A growing body of work explores immersive environments for network visualisation, ranging from design studies to fully integrated applications across various domains. Despite these advancements, the field remains fragmented, with diverse methodologies, hardware setups, and evaluation criteria, often lacking clear connections to prior work. This fragmentation complicates the comparability and generalisability of findings. To address this, we present a structured survey of visual network analysis in immersive environments. We systematically categorise and analyse existing approaches, revealing connections and coverage within the design space. By synthesising findings of experiments and evaluating current applications, we identify key achievements, challenges, and research gaps. Additionally, we provide an interactive online resource for exploring and updating results, aiming to guide researchers and practitioners in advancing the field. This work provides a comprehensive overview of the research landscape and proposes actionable insights to foster innovation in immersive network analysis.

A flow control framework based on linear stability analysis is proposed focusing on reducing the aerodynamic drag due to dynamic stall through a finite-window temporal actuation. The methodology is applied on a periodically plunging SD7003 airfoil.Finite-time Lyapunov exponent (FTLE) fields reveal a saddle point near the airfoil leading edge, where a shear layer forms and feeds a dynamic stall vortex (DSV). A local stability analysis conducted at this saddle point identifies a Kelvin-Helmholtz instability, and the most unstable eigenvalue frequencies remain constant when the variation in the effective angle of attack is minimal. The findings from the FTLE fields and the stability analysis are used to inform the position and finite duty cycle of a periodic blowing and suction actuation applied in a wall-resolved large eddy simulation (LES). The present framework reduces the actuation duty cycle by 77.5% during the airfoil plunging motion, while maintaining the same performance as a continuous actuation throughout the entire cycle. The LES results demonstrate that disturbances from the stability-analysis-informed actuation modify the leading-edge dynamics, preventing the formation of the coherent DSV and significantly reducing the drag., Comment: 23 pages, 10 figures

Compact, star-forming galaxies with high star formation rate surface densities ($\Sigma_{\text{SFR}}$) are often efficient Lyman continuum (LyC) emitters at $z\leq 4.5$, likely as intense stellar feedback creates low-density channels that allow photons to escape. Irregular or disturbed morphologies, such as those resulting from mergers, can also facilitate LyC escape by creating anisotropic gas distributions. We investigate the influence of galaxy morphology on LyC production and escape at redshifts $5 \leq z \leq 7$ using observations from various \textit{James Webb Space Telescope} (JWST) surveys. Our sample consists of 436 sources, which are predominantly low-mass ($\sim 10^{8.15} M_\odot$), star-forming galaxies with ionizing photon efficiency ($\xi_{\rm ion}$) values consistent with canonical expectations. Since direct measurements of $f_{\rm esc}$ are not possible during the Epoch of Reionization (EoR), we predict $f_{\rm esc}$ for high-redshift galaxies by applying survival analysis to a subsample of LyC emitters from the Low-Redshift Lyman Continuum Survey (LzLCS), selected to be direct analogs of reionization-era galaxies. We find that these galaxies exhibit on average modest predicted escape fractions ($\sim 0.04$). Additionally, we assess the correlation between morphological features and LyC emission. Our findings indicate that neither $\xi_{\rm ion}$ nor the predicted $f_{\rm esc}$ values show a significant correlation with the presence of merger signatures. This suggests that in low-mass galaxies at $z \geq 5$, strong morphological disturbances are not the primary mechanism driving LyC emission and leakage. Instead, compactness and star formation activity likely play a more pivotal role in regulating LyC escape., Comment: Submitted to A&A

Increasingly, recommender systems are tasked with improving users' long-term satisfaction. In this context, we study a content exploration task, which we formalize as a bandit problem with delayed rewards. There is an apparent trade-off in choosing the learning signal: waiting for the full reward to become available might take several weeks, slowing the rate of learning, whereas using short-term proxy rewards reflects the actual long-term goal only imperfectly. First, we develop a predictive model of delayed rewards that incorporates all information obtained to date. Rewards as well as shorter-term surrogate outcomes are combined through a Bayesian filter to obtain a probabilistic belief. Second, we devise a bandit algorithm that quickly learns to identify content aligned with long-term success using this new predictive model. We prove a regret bound for our algorithm that depends on the \textit{Value of Progressive Feedback}, an information theoretic metric that captures the quality of short-term leading indicators that are observed prior to the long-term reward. We apply our approach to a podcast recommendation problem, where we seek to recommend shows that users engage with repeatedly over two months. We empirically validate that our approach significantly outperforms methods that optimize for short-term proxies or rely solely on delayed rewards, as demonstrated by an A/B test in a recommendation system that serves hundreds of millions of users.

In recent years, the importance of well-documented metadata has been discussed increasingly in many research fields. Making all metadata generated during scientific research available in a findable, accessible, interoperable, and reusable (FAIR) manner remains a significant challenge for researchers across fields. Scientific communities are agreeing to achieve this by making all data available in a semantically annotated knowledge graph using semantic web technologies. Most current approaches do not gather metadata in a consistent and community-agreed standardized way, and there are insufficient tools to support the process of turning them into a knowledge graph. We present an example solution in which the creation of a schema and ontology are placed at the beginning of the scientific process which is then - using the electronic laboratory notebook framework Herbie - turned into a bespoke data collection platform to facilitate validation and semantic annotation of the metadata immediately during an experiment. Using the example of synchrotron radiation-based nano computed tomography measurements, we present a holistic approach which can capture the complex metadata of such research instruments in a flexible and straightforward manner. Different instrument setups of this beamline can be considered, allowing a user-friendly experience. We show how Herbie turns all semantic documents into an accessible user interface, where all data entered automatically fulfills all requirements of being FAIR, and present how data can be directly extracted via competency questions without requiring familiarity with the fine-grained structure of the knowledge graph., Comment: 21 pages, 13 figures, 5 tables. Fabian Kirchner and Florian Wieland have contributed equally to the manuscript. Corresponding authors: fabian.kirchner@hereon.de, catriona.eschke@hereon.de

We introduce a scissors congruence $K$-theory spectrum which lifts the equivariant scissors congruence groups for compact $G$-manifolds with boundary, and we show that on $\pi_0$ this is the source of a spectrum level lift of the Burnside ring valued equivariant Euler characteristic of a compact $G$-manifold. We also show that the equivariant scissors congruence groups for varying subgroups assemble into a Mackey functor, which is a shadow of a conjectural higher genuine equivariant structure., Comment: 20 pages, 1 figure

We present VIRAC2, a catalogue of positions, proper motions, parallaxes and $Z$, $Y$, $J$, $H$, and $K_s$ near-infrared photometric time series of 545 346 537 unique stars. The catalogue is based on a point spread function fitting reduction of nearly a decade of VISTA VVV and VVVX images, which cover $560~{\rm deg}^2$ of the Southern Galactic plane and bulge. The catalogue is complete at the $>90$ per cent level for $11

This paper investigates the formation, populations, and evolutionary paths of UltraLuminous X-ray Sources (ULXs) within Globular Clusters (GCs). ULXs, characterised by their extreme X-ray luminosities, present a challenge to our understanding of accretion physics and compact object formation. While previous studies have largely focused on field populations, this research examines the unique environment of GCs, where dynamical interactions play a significant role. Using the MOCCA Monte Carlo code, we explore how dynamics influences ULX populations within these dense stellar clusters. Our findings reveal that dynamical processes, such as binary hardening and exchanges, can both facilitate and impede ULX formation in GCs. The study explores the impact of parameters including the initial binary fraction, tidal filling, and multiple stellar populations on the evolution of ULXs. We find that non-tidally filling clusters exhibit significantly larger ULX populations compared to tidally filling ones. The results indicate that the apparent scarcity of ULXs in GCs may be related to the older stellar populations of GCs relative to the field. Furthermore, the study identifies a population of "escaper" ULXs, which originate in GCs but are ejected and emit X-rays outside the cluster. These escapers may significantly contribute to the observed field ULX population., Comment: 13 pages, 3 figures, submitted to A&A

The CIELO project introduces a novel set of chemo-dynamical zoom-in simulations designed to simultaneously resolve galaxies and their nearby environments. The initial conditions include a diverse range of cosmic structures, such as local groups, filaments, voids, and walls, allowing for a detailed exploration of galaxies within the broader context of the cosmic web. This study presents the initial conditions and characterizes the global properties of CIELO galaxies and their environments. It focuses on galaxies with stellar masses ranging from log [8,11] solar masses and examines key scaling relations, including the mass-size relation, the Tully-Fisher relation, and the mass-metallicity relation for both stars and star-forming gas. The DisPerSe algorithm was used to determine the positions of CIELO galaxies within the cosmic web, with a specific focus on the Pehuen haloes. The selection of local group volumes was guided by criteria based on the relative positions and velocities of the two primary galaxies. The Pehuen regions were chosen to map walls, filaments, and voids. Synthetic images in the SDSS i, r, and g bands were generated using the SKIRT radiative transfer code. Additionally, a dynamical decomposition was performed to classify galaxy morphologies into bulge, disc, and stellar halo components (abridged)., Comment: 13 pages, 12 figures, submitted. Comments are welcomed

In Self-Supervised Learning (SSL), pre-training and evaluation are resource intensive. In the speech domain, current indicators of the quality of SSL models during pre-training, such as the loss, do not correlate well with downstream performance. Consequently, it is often difficult to gauge the final downstream performance in a cost efficient manner during pre-training. In this work, we propose unsupervised efficient methods that give insights into the quality of the pre-training of SSL speech models, namely, measuring the cluster quality and rank of the embeddings of the SSL model. Results show that measures of cluster quality and rank correlate better with downstream performance than the pre-training loss with only one hour of unlabeled audio, reducing the need for GPU hours and labeled data in SSL model evaluation.

Neural network verification is a new and rapidly developing field of research. So far, the main priority has been establishing efficient verification algorithms and tools, while proper support from the programming language perspective has been considered secondary or unimportant. Yet, there is mounting evidence that insights from the programming language community may make a difference in the future development of this domain. In this paper, we formulate neural network verification challenges as programming language challenges and suggest possible future solutions., Comment: Accepted at ESOP 2025, European Symposium on Programming Languages

We introduce a novel subgrid-scale model for shear flows, exploiting the spatial intermittency and the scale separation between large-scale flows and coherent small-scale structures. The model is highly sparse, focusing exclusively on the most intense structures, which are represented by vortons dynamically regularized quasi-singularities that experience rapid distortion from the large-scale shear. The vortons, in turn, influence the large-scale flow through the sub-grid stress tensor. The model displays an interesting transition between two distinct regimes: (i) a laminar regime, where dissipation is entirely attributed to the large-scale flow, and the vortons dynamics is essentially diffusive, and (ii) a turbulent regime, in which most of the dissipation arises from the vortons. These regimes correspond to different scalings of dissipation and the Grashof number as functions of the Reynolds number, with power-law relationships that resemble those observed in classical turbulence.

We address the longstanding challenge in quantum statistical mechanics of reconciling unitary dynamics with irreversible relaxation. In classical chaos, the unitary evolution operator develops Ruelle-Pollicott (RP) resonances inside the unit circle in the continuum limit, leading to mixing. In contrast, the quantum theory of many-body RP resonances and their link to irreversibility remain underdeveloped. We relate the spectral form factor to the sum of autocorrelation functions and, in generic many-body systems without conservation laws, argue that all quantum RP resonances converge inside the unit disk. While we conjecture this picture to be general, we analytically prove the emergence of irreversibility in the random phase model (RPM), a paradigmatic Floquet quantum circuit model, in the limit of large local Hilbert space dimension. To this end, we couple it to local environments and compute the exact time evolution of autocorrelation functions, the dissipative form factor, and out-of-time-order correlation functions. Although valid for any dissipation strength, we then focus on weak dissipation to clarify the origin of irreversibility in unitary systems. When the dissipationless limit is taken after the thermodynamic limit, the unitary quantum map develops an infinite tower of RP resonances -- chaotic systems display so-called anomalous relaxation. We identify the exact RP resonances in the RPM and prove that the same RP resonances are obtained from operator truncation. We also show that the OTOC in the RPM can undergo a two-stage relaxation and that during the second stage, the approach to the stationary value is again controlled by the leading RP resonance. Finally, we demonstrate how conservation laws, many-body localization, and nonlocal interactions merge the leading RP resonance into the unit circle, thereby suppressing anomalous relaxation., Comment: 24 pages, 7 figures

Large language models (LLMs) are advanced AI systems applied across various domains, including NLP, information retrieval, and recommendation systems. Despite their adaptability and efficiency, LLMs have not been extensively explored for signal processing tasks, particularly in the domain of global navigation satellite system (GNSS) interference monitoring. GNSS interference monitoring is essential to ensure the reliability of vehicle localization on roads, a critical requirement for numerous applications. However, GNSS-based positioning is vulnerable to interference from jamming devices, which can compromise its accuracy. The primary objective is to identify, classify, and mitigate these interferences. Interpreting GNSS snapshots and the associated interferences presents significant challenges due to the inherent complexity, including multipath effects, diverse interference types, varying sensor characteristics, and satellite constellations. In this paper, we extract features from a large GNSS dataset and employ LLaVA to retrieve relevant information from an extensive knowledge base. We employ prompt engineering to interpret the interferences and environmental factors, and utilize t-SNE to analyze the feature embeddings. Our findings demonstrate that the proposed method is capable of visual and logical reasoning within the GNSS context. Furthermore, our pipeline outperforms state-of-the-art machine learning models in interference classification tasks.

This work presents FreeSVC, a promising multilingual singing voice conversion approach that leverages an enhanced VITS model with Speaker-invariant Clustering (SPIN) for better content representation and the State-of-the-Art (SOTA) speaker encoder ECAPA2. FreeSVC incorporates trainable language embeddings to handle multiple languages and employs an advanced speaker encoder to disentangle speaker characteristics from linguistic content. Designed for zero-shot learning, FreeSVC enables cross-lingual singing voice conversion without extensive language-specific training. We demonstrate that a multilingual content extractor is crucial for optimal cross-language conversion. Our source code and models are publicly available.

To ensure resilience against the unavoidable noise in quantum computers, quantum information needs to be encoded using an error-correcting code, and circuits must have a particular structure to be fault-tolerant. Compilation of fault-tolerant quantum circuits is thus inherently different from the non-fault-tolerant case. However, automated fault-tolerant compilation methods are widely underexplored, and most known constructions are obtained manually for specific codes only. In this work, we focus on the problem of automatically synthesizing fault-tolerant circuits for the deterministic initialization of an encoded state for a broad class of quantum codes that are realizable on current and near-term hardware. To this end, we utilize methods based on techniques from classical circuit design, such as satisfiability solving, resulting in tools for the synthesis of (optimal) fault-tolerant state preparation circuits for near-term quantum codes. We demonstrate the correct fault-tolerant behavior of the synthesized circuits using circuit-level noise simulations. We provide all routines as open-source software as part of the Munich Quantum Toolkit (MQT) at https://github.com/cda-tum/mqt-qecc., Comment: 7 pages, 4 figures, accepted at DATE 2025

Incoherent Diffraction Imaging - IDI - is a diffraction-based imaging technique that has been recently proposed to exploit the partial coherence of incoherently scattered light to retrieve structural information from the scattering centers. Similar to the stellar intensity interferometry of Hanbury Brown and Twiss, the signal builds up on the second-order spatial correlations of the emitted light. The complex spatial distribution of the target is thereby encoded in the spatial intensity fluctuations of the scattered light. The first experimental realisations of this imaging technique have been realised using the fluorescence excited by an ultra-short X-ray pulse at Free Electron Laser (FEL) facilities. Here, we propose an alternative set-up based on a table-top Pseudo-Thermal Light Source. This set-up allows us to explore IDI under a wide range of physically relevant conditions as well as to benchmark numerical and analytical models currently used to determine the imaging capabilities of this technique.

This study presents new insights into gluon transverse momentum distributions through nonextensive statistical mechanics, addressing their implications for QCD phenomenology. The saturation physics and scaling laws present in high energy collision data are investigated as a consequence of gluon distribution modification at high density regime. The analysis explores how these modifications influence observables across different collision systems, such as proton-proton, proton-nucleus, and relativistic heavy-ion collisions. Both high and low $p_T$ regions are successfully described in hadron production., Comment: 17 pages, 6 Figures. Contribution to MDPI Physics Special Issue "Complexity in High Energy and Statistical Physics"

Grokking, the sudden generalization that occurs after prolonged overfitting, is a surprising phenomenon challenging our understanding of deep learning. Although significant progress has been made in understanding grokking, the reasons behind the delayed generalization and its dependence on regularization remain unclear. In this work, we argue that without regularization, grokking tasks push models to the edge of numerical stability, introducing floating point errors in the Softmax function, which we refer to as Softmax Collapse (SC). We demonstrate that SC prevents grokking and that mitigating SC enables grokking without regularization. Investigating the root cause of SC, we find that beyond the point of overfitting, the gradients strongly align with what we call the na\"ive loss minimization (NLM) direction. This component of the gradient does not alter the model's predictions but decreases the loss by scaling the logits, typically by scaling the weights along their current direction. We show that this scaling of the logits explains the delay in generalization characteristic of grokking and eventually leads to SC, halting further learning. To validate our hypotheses, we introduce two key contributions that address the challenges in grokking tasks: StableMax, a new activation function that prevents SC and enables grokking without regularization, and $\perp$Grad, a training algorithm that promotes quick generalization in grokking tasks by preventing NLM altogether. These contributions provide new insights into grokking, elucidating its delayed generalization, reliance on regularization, and the effectiveness of existing grokking-inducing methods. Code for this paper is available at https://github.com/LucasPrietoAl/grokking-at-the-edge-of-numerical-stability.

We present the Cosmic Evolution Early Release Science (CEERS) Survey, a 77.2 hour Director's Discretionary Early Release Science Program. CEERS demonstrates, tests, and validates efficient extragalactic surveys using coordinated, overlapping parallel observations with the JWST instrument suite, including NIRCam and MIRI imaging, NIRSpec low (R~100) and medium (R~1000) resolution spectroscopy, and NIRCam slitless grism (R~1500) spectroscopy. CEERS targets the Hubble Space Telescope-observed region of the Extended Groth Strip (EGS) field, supported by a rich set of multiwavelength data. CEERS facilitated immediate community science in both of the extragalactic core JWST science drivers ``First Light" and ``Galaxy Assembly," including: 1) The discovery and characterization of large samples of galaxies at z >~ 10 from ~90 arcmin^2 of NIRCam imaging, constraining their abundance and physical nature; 2) Deep spectra of >1000 galaxies, including dozens of galaxies at 63; and 4) Characterizing galaxy mid-IR emission with MIRI to study dust-obscured star-formation and supermassive black hole growth at z~1-3. As a legacy product for the community, the CEERS team has provided several data releases, accompanied by detailed notes on the data reduction procedures and notebooks to aid in reproducibility. In addition to an overview of the survey and quality of the data, we provide science highlights from the first two years with CEERS data., Comment: 38 pages, 13 figures, 6 tables

The mass accretion process controls pre-main-sequence evolution, although its intrinsic instability has yet to be fully understood, especially towards the protostellar stage. In this work, we have undertaken a thorough examination of the mid-infrared variability of Spitzer-selected YSOs in the Cygnus-X star-forming region over the last decade, using the NEOWISE time series. This work compares two groups of young stars: embedded Class I objects, and the more evolved flat-spectrum/Class II sources. We report on 48 candidate eruptive variables within these groups, including 14 with characteristics that resemble the photometric behaviour of FUors. We also include an additional 20 YSOs, which are of a less certain categorisation. We find the candidate FUors to be an order of magnitude more common among the younger Class I systems than more evolved objects. A large number of the identified short-duration eruptive YSOs display mid-infrared colour behaviour that is redder-when-brighter, which contrasts with optically bright outbursts seen in YSOs. Finally, we note the unusual long-term rising behaviours of four Class I YSOs, with rise timescales longer than five years, which is far slower than 6-12 month timescale for the majority of optically discovered FUors. Additionally, our broader investigation of MIR variability for embedded class I YSOs shows that there is a higher incidence of high amplitude variability for these stars, than is seen in class II sources. This holds true for all variable class I YSOs, not just the eruptive sources., Comment: 20 Pages, 24 Figures and 3 Tables

For a list-assignment $L$, the reconfiguration graph $C_L(G)$ of a graph $G$ is the graph whose vertices are proper $L$-colorings of $G$ and whose edges link two colorings that differ on only one vertex. If $|L(v)| \ge d(v) + 2$ for every vertex of $G$, it is known that $C_L(G)$ is connected. In this case, Cambie et al. investigated the diameter of $C_L(G)$. They conjectured that $diam(C_L(G)) \le n(G) + \mu(G)$ with $\mu(G)$ the size of a maximum matching of $G$ and proved several results towards this conjecture. We answer to two of their open problems by proving the conjecture for two classes of graphs, namely subcubic graphs and complete multipartite graphs., Comment: 12 pages, 7 figures

In recent years, copositive programming has received significant attention for its ability to model hard problems in both discrete and continuous optimization. Several relaxations of copositive programs based on semidefinite programming (SDP) have been proposed in the literature, meant to provide tractable bounds. However, while these SDP-based relaxations are amenable to the ellipsoid algorithm and interior point methods, it is not immediately obvious that they can be solved in polynomial time (even approximately). In this paper, we consider the sum-of-squares (SOS) hierarchies of relaxations for copositive programs introduced by Parrilo (2000), de Klerk & Pasechnik (2002) and Pe\~na, Vera & Zuluaga (2006), which can be formulated as SDPs. We establish sufficient conditions that guarantee the polynomial-time computability (up to fixed precision) of these relaxations. These conditions are satisfied by copositive programs that represent standard quadratic programs and their reciprocals. As an application, we show that the SOS bounds for the (weighted) stability number of a graph can be computed efficiently. Additionally, we provide pathological examples of copositive programs (that do not satisfy the sufficient conditions) whose SOS relaxations admit only feasible solutions of doubly-exponential size.

We present a theoretical analysis of nearly monochromatic light propagation through a gas of two-level atoms using the Heisenberg-Langevin equation method. Our focus is on the evolution of the photon annihilation operator and its impact on the second-order correlation function, $g^{(2)}({\tau})$, with particular emphasis on photon antibunching behavior. The model accounts for both open and closed atomic system approximations, including Doppler broadening and the influence of pump field detuning. We derive expressions that reproduce known results from scattering theory and extend the analysis to complex systems, such as inhomogeneously broadened media. The theoretical predictions are compared with experimental data from a waveguide QED platform, which show good agreement and thereby demonstrate the power of our approach. Future work will explore extensions to even more complex systems and other quantum light characteristics for practical applications., Comment: 19 pages, 10 figures

Large Language Models (LLMs) have demonstrated remarkable performance in various NLP tasks, including semantic parsing, which trans lates natural language into formal code representations. However, the reverse process, translating code into natural language, termed semantic captioning, has received less attention. This task is becoming increasingly important as LLMs are integrated into platforms for code generation, security analysis, and educational purposes. In this paper, we focus on the captioning of SQL query (SQL2Text) to address the critical need for understanding and explaining SQL queries in an era where LLM-generated code poses potential security risks. We repurpose Text2SQL datasets for SQL2Text by introducing an iterative ICL prompt using GPT-4o to generate multiple additional utterances, which enhances the robustness of the datasets for the reverse task. We conduct our experiments using in-context learning (ICL) based on different sample selection methods, emphasizing smaller, more computationally efficient LLMs. Our findings demonstrate that leveraging the inherent graph properties of SQL for ICL sample selection significantly outperforms random selection by up to 39% on BLEU score and provides better results than alternative methods. Dataset and codes are published: \url{https://github.com/aliwister/ast-icl}., Comment: Accepted to COLING'25

Rapid advances in Natural Language Processing (NLP) have revolutionized many fields, including healthcare. However, these advances raise significant privacy concerns, especially when models specialized on sensitive data can memorize and then expose and regurgitate confidential information. This paper presents a privacy-by-design language modeling approach to address the problem of language models anonymization, and thus promote their sharing. Specifically, we propose both a Masking Language Modeling (MLM) methodology to specialize a BERT-like language model, and a Causal Language Modeling (CLM) methodology to specialize a GPT-like model that avoids the model from memorizing direct and indirect identifying information present in the training data. We have comprehensively evaluated our approaches using medical datasets and compared them against different baselines. Our results indicate that by avoiding memorizing both direct and indirect identifiers during model specialization, our masking and causal language modeling schemes offer the best tradeoff for maintaining high privacy while retaining high utility.