Search

The main challenges limiting the adoption of deep learning-based solutions in medical workflows are the availability of annotated data and the lack of interpretability of such systems. Concept Bottleneck Models (CBMs) tackle the latter by constraining the final disease prediction on a set of predefined and human-interpretable concepts. However, the increased interpretability achieved through these concept-based explanations implies a higher annotation burden. Moreover, if a new concept needs to be added, the whole system needs to be retrained. Inspired by the remarkable performance shown by Large Vision-Language Models (LVLMs) in few-shot settings, we propose a simple, yet effective, methodology, CBVLM, which tackles both of the aforementioned challenges. First, for each concept, we prompt the LVLM to answer if the concept is present in the input image. Then, we ask the LVLM to classify the image based on the previous concept predictions. Moreover, in both stages, we incorporate a retrieval module responsible for selecting the best examples for in-context learning. By grounding the final diagnosis on the predicted concepts, we ensure explainability, and by leveraging the few-shot capabilities of LVLMs, we drastically lower the annotation cost. We validate our approach with extensive experiments across four medical datasets and twelve LVLMs (both generic and medical) and show that CBVLM consistently outperforms CBMs and task-specific supervised methods without requiring any training and using just a few annotated examples. More information on our project page: https://cristianopatricio.github.io/CBVLM/., Comment: This work has been submitted to the IEEE for possible publication

For semiconductor device fabrication, Pulsed Laser Annealing (PLA) offers significant advantages over conventional thermal processes. Notably, it can provide ultrafast (~ns) and high temperature profiles ($>1000^\circ$C). When the maximum temperature exceeds the melting point, a solid-liquid phase transition is observed, immediately followed by rapid recrystallization. This unique annealing mechanism gives raises questions about dopant diffusion and residual defects, in not only in the recrystallized region, but also just below it. As power devices require micrometer-sized junctions, high laser energy densities are needed, which were proved to promote the incorporation of complex impurities from the surface and the creation of defects at the liquid/solid interface. This paper reports on the impact of laser annealing at high energy densities (up to 8.0 J/cm$^2$) on the leakage current, using Schottky and PN diodes, and DLTS measurements. Various laser annealing conditions were used: energy densities between 1.7 and 8.0 J/cm$^2$ with 1 to 10 pulses. Our results suggest that the liquid and solid solubility of vacancies in silicon are fixed by the maximum temperature reached, so to the energy density. Increasing the number of laser pulses allows, not only to reach this maximum vacancy concentration but also to promote their diffusion towards the surface. Concomitantly, the in-diffusion of complex impurities inside the melted region allows the coupling between both defect types to create trap centers, responsible for the degradation of the leakage current.

Per- and polyfluoroalkyl substances (PFAS), often referred as "forever chemicals," are pervasive environmental pollutants due to their resistance to degradation. Among these, perfluorooctanoic acid (PFOA) poses significant threats to human health, contaminating water sources globally. Here, we have demonstrated the potential of a novel electrochemical sensor based on two-dimensional (2D) aluminium-based multicomponent quasicrystals (2D-Al QC) for the ultrasensitive sub-picomolar level detection of PFOA. The 2D-Al QC-inked electrode was employed here to detect PFOA by differential pulse voltammetry (DPV). The limit of detection (LoD) achieved is 0.59 +/- 0.05 pM. The sensor was evaluated for selectivity with other interfering compounds, repeatability of cycles, and reproducibility for five similar electrodes with a deviation of 0.8 %. The stability of the sensor has also been analysed after ninety days ,which shows a minimal variation of 15%. Spectroscopic techniques and theoretical calculations were further utilized to understand the interaction between the 2D-Al QC and PFOA. The results demonstrate that the 2D-Al QC offers a promising platform for the rapid and sensitive detection of PFOA, potentially addressing current environmental monitoring challenges.

The Large Hadron Collider at CERN produces immense volumes of complex data from high-energy particle collisions, demanding sophisticated analytical techniques for effective interpretation. Neural Networks, including Graph Neural Networks, have shown promise in tasks such as event classification and object identification by representing collisions as graphs. However, while Graph Neural Networks excel in predictive accuracy, their "black box" nature often limits their interpretability, making it difficult to trust their decision-making processes. In this paper, we propose a novel approach that combines a Graph Transformer model with Mixture-of-Expert layers to achieve high predictive performance while embedding interpretability into the architecture. By leveraging attention maps and expert specialization, the model offers insights into its internal decision-making, linking predictions to physics-informed features. We evaluate the model on simulated events from the ATLAS experiment, focusing on distinguishing rare Supersymmetric signal events from Standard Model background. Our results highlight that the model achieves competitive classification accuracy while providing interpretable outputs that align with known physics, demonstrating its potential as a robust and transparent tool for high-energy physics data analysis. This approach underscores the importance of explainability in machine learning methods applied to high energy physics, offering a path toward greater trust in AI-driven discoveries.

Superradiance clouds of kinetically-mixed dark photons around spinning black holes can produce observable multi-messenger electromagnetic and gravitational wave signals. The cloud generates electric fields of up to a Teravolt-per-meter, which lead to a cascade production of charged particles, yielding a turbulent quasi-equilibrium plasma around the black hole, and resulting in electromagnetic fluxes ranging from supernova to pulsar-like luminosities. For stellar mass black holes, such systems resemble millisecond pulsars and are expected to emit pulsating radio waves and continuous gravitational waves (CWs) within the LIGO-Virgo-KAGRA (LVK) sensitivity band. We select 44 sources with approximately coincident frequencies or positive frequency drifts from existing pulsar catalogs as potential candidates of long-lasting superradiance clouds around old galactic black holes. For a subset of 34 sources that are well measured and have not been previously targeted, we perform the first search for CW emission in LVK data from the third observing run. We find no evidence of a CW signal and place 95% confidence level upper limits on the emitted strain amplitude. We interpret these results, together with limits from previous searches, in terms of the underlying dark photon theory by performing an analysis of the expected signals from superradiance clouds from galactic black holes. We find that, even for moderately spinning black holes, the absence of an observed CW signal disfavors a discrete set of dark photon masses between about $10^{-13}$ $\rm{eV}/c^2$ and $10^{-12}$ $\rm{eV}/c^2$ and kinetic mixing couplings in the range of $10^{-9}$-$10^{-7}$, subject to assumptions about the properties of the black hole population and the cloud's electromagnetic emission., Comment: 26 pages, 8 figures

We develop an Integral Transformation Method (ITM) for the study of suitable optimal control and differential game models. This allows for a solution to such dynamic problems to be found through solving a family of optimization problems parametrized by time. The method is quite flexible, and it can be used in several economic applications where the state equation and the objective functional are linear in a state variable. We illustrate the ITM in the context of a two-country integrated assessment climate model. We characterize emissions, consumption, transfers, and welfare by computing the Nash equilibria of the associated dynamic game. We then compare them to efficiency benchmarks. Further, we apply the ITM in a robust control setup, where we investigate how (deep) uncertainty affects climate outcomes.

The future mobile network has the complex mission of distributing available radio resources among various applications with different requirements. The radio access network slicing enables the creation of different logical networks by isolating and using dedicated resources for each group of applications. In this scenario, the radio resource scheduling (RRS) is responsible for distributing the radio resources available among the slices to fulfill their service-level agreement (SLA) requirements, prioritizing critical slices while minimizing the number of intent violations. Moreover, ensuring that the RRS can deal with a high diversity of network scenarios is essential. Several recent papers present advances in machine learning-based RRS. However, the scenarios and slice variety are restricted, which inhibits solid conclusions about the generalization capabilities of the models after deployment in real networks. This paper proposes an intent-based RRS using multi-agent reinforcement learning in a radio access network (RAN) slicing context. The proposed method protects high-priority slices when the available radio resources cannot fulfill all the slices. It uses transfer learning to reduce the number of training steps required. The proposed method and baselines are evaluated in different network scenarios that comprehend combinations of different slice types, channel trajectories, number of active slices and users' equipment (UEs), and UE characteristics. The proposed method outperformed the baselines in protecting slices with higher priority, obtaining an improvement of 40% and, when considering all the slices, obtaining an improvement of 20% in relation to the baselines. The results show that by using transfer learning, the required number of training steps could be reduced by a factor of eight without hurting performance.

A significant debate has emerged in response to a paper written by Steven Piantadosi (Piantadosi, 2023) and uploaded to the LingBuzz platform, the open archive for generative linguistics. Piantadosi's dismissal of Chomsky's approach is ruthless, but generative linguists deserve it. In this paper, I will adopt three idealized perspectives -- computational, theoretical, and experimental -- to focus on two fundamental issues that lend partial support to Piantadosi's critique: (a) the evidence challenging the Poverty of Stimulus (PoS) hypothesis and (b) the notion of simplicity as conceived within mainstream Minimalism. In conclusion, I argue that, to reclaim a central role in language studies, generative linguistics -- representing a prototypical theoretical perspective on language -- needs a serious update leading to (i) more precise, consistent, and complete formalizations of foundational intuitions and (ii) the establishment and utilization of a standardized dataset of crucial empirical evidence to evaluate the theory's adequacy. On the other hand, ignoring the formal perspective leads to major drawbacks in both computational and experimental approaches. Neither descriptive nor explanatory adequacy can be easily achieved without the precise formulation of general principles that can be challenged empirically.

We investigate the dynamics of multi-mode optical systems driven by two-photon processes and subject to non-local losses, incorporating quantum noise at the Gaussian level. Our findings show that the statistics retrieved from a single Gaussian quantum trajectory exhibits emergent thermal equilibrium governed by an Ising Hamiltonian, encoded in the dissipative coupling between modes. The system's effective temperature is set by the driving strength relative to the oscillation threshold. Given the ultra-short time scales typical of all-optical devices, our study demonstrates that such multi-mode optical systems can operate as ultra-fast Boltzmann samplers, paving the way towards the realization of efficient hardware for combinatorial optimization, with promising applications in machine learning and beyond., Comment: Manuscript 6 pages, 4 figures. Supplementary material 3 pages, 2 figures

In a Universe with nearly-Gaussian initial curvature perturbations, the abundance of primordial black holes can be derived from the curvature power spectrum. When the latter is enhanced within a narrow range around a characteristic scale, the resulting mass function has a single distinct peak, corresponding to Schwarzschild radii set by the horizon entry time of that scale. In contrast, we show (both numerically and by providing an analytic estimation) that a broad enhancement - such as a plateau bounded by infrared and ultraviolet scales - produces a bimodal mass function, with a primary peak close to the infrared scale. We find that the typical initial gravitational potential (compaction function), conditioned on meeting the threshold for critical collapse, is generated by a thin spherical shell with infrared radius and a thickness comparable to the ultraviolet scale. This suggests a higher-than-expected abundance of PBH originating from Type II initial fluctuations. Our results significantly impact overproduction bounds on the amplitude of the power spectrum, and tighten the viable mass range for primordial black holes as dark matter., Comment: 7 pages, 3 figures

We theoretically investigate transport affected by cavity-mediated electron hopping in multi-terminal quantum Hall bars, quantum point contacts, and Aharonov-Bohm interferometers. Beyond determining conductances and resistances, we analyze spatially resolved current distributions and local density of states. Our study reveals how cavity-mediated inter-edge scattering impacts quantum magnetotransport in finite-size systems and how the cavity-mediated hopping significantly alters electron quantum interference effects., Comment: 12 pages, 11 figures

Describing the processes involved in analyzing data from electrophysiology experiments to investigate the function of neural systems is inherently challenging. On the one hand, data can be analyzed by distinct methods that serve a similar purpose, such as different algorithms to estimate the spectral power content of a measured time series. On the other hand, different software codes can implement the same algorithm for the analysis while adopting different names to identify functions and parameters. Having reproducibility in mind, with these ambiguities the outcomes of the analysis are difficult to report, e.g., in the methods section of a manuscript or on a platform for scientific findings. Here, we illustrate how using an ontology to describe the analysis process can assist in improving clarity, rigour and comprehensibility by complementing, simplifying and classifying the details of the implementation. We implemented the Neuroelectrophysiology Analysis Ontology (NEAO) to define a unified vocabulary and to standardize the descriptions of the processes involved in analyzing data from neuroelectrophysiology experiments. Real-world examples demonstrate how the NEAO can be employed to annotate provenance information describing an analysis process. Based on such provenance, we detail how it can be used to query various types of information (e.g., using knowledge graphs) that enable researchers to find, understand and reuse prior analysis results.

Our peculiar velocity imprints a dipole on galaxy density maps derived from redshift surveys. The dipole gives rise to an oscillatory signal in the multipole moments of the observed power spectrum which we indicate as the finger-of-the-observer (FOTO) effect. Using a suite of large mock catalogues mimicking ongoing and future $\textrm{H}\alpha$- and $\textrm{H}\scriptstyle\mathrm{I}$-selected surveys, we demonstrate that the oscillatory features can be measured with a signal-to-noise ratio of up to 7 (depending on the sky area coverage and provided that observational systematics are kept under control on large scales). We also show that the FOTO effect cannot be erased by correcting the individual galaxy redshifts. On the contrary, by leveraging the power of the redshift corrections, we propose a novel method to determine both the magnitude and the direction of our peculiar velocity. After applying this technique to our mock catalogues, we conclude that it can be used to either test the kinematic interpretation of the temperature dipole in the cosmic microwave background or to extract cosmological information such as the matter density parameter and the equation of state of dark energy., Comment: 44 pages, 17 figures

While the accurate description of redox reactions remains a challenge for first-principles calculations, it has been shown that extended Hubbard functionals (DFT+U+V) can provide a reliable approach, mitigating self-interaction errors, in materials with strongly localized d or f electrons. Here, we first show that DFT+U+V molecular dynamics is capable to follow the adiabatic evolution of oxidation states over time, using representative Li-ion cathode materials. In turn, this allows to develop redox-aware machine-learned potentials. We show that considering atoms with different oxidation states (as accurately predicted by DFT+U+V) as distinct species in the training leads to potentials that are able to identify the correct ground state and pattern of oxidation states for redox elements present. This can be achieved, e.g., through a systematic combinatorial search for the lowest energy configuration or with stochastic methods. This brings the advantages of machine-learned potentials to key technological applications (e.g., rechargeable batteries), which require an accurate description of the evolution of redox states.

Autonomous driving needs good roads, but 85% of Brazilian roads have damages that deep learning models may not regard as most semantic segmentation datasets for autonomous driving are high-resolution images of well-maintained urban roads. A representative dataset for emerging countries consists of low-resolution images of poorly maintained roads and includes labels of damage classes; in this scenario, three challenges arise: objects with few pixels, objects with undefined shapes, and highly underrepresented classes. To tackle these challenges, this work proposes the Performance Increment Strategy for Semantic Segmentation (PISSS) as a methodology of 14 training experiments to boost performance. With PISSS, we reached state-of-the-art results of 79.8 and 68.8 mIoU on the Road Traversing Knowledge (RTK) and Technik Autonomer Systeme 500 (TAS500) test sets, respectively. Furthermore, we also offer an analysis of DeepLabV3+ pitfalls for small object segmentation.

Misaligned circumbinary disks will produce dust traffic jams during alignment or anti-alignment to the binary orbital plane. We conduct a hydrodynamical simulation of an initially misaligned circumbinary disk undergoing polar alignment with multiple dust species. Due to differential precession between the gas and dust components, multiple dust traffic jams are produced within the disk during polar alignment. The radial locations of the dust traffic jams depend on the Stokes number of the grains, which depends on grain size. We compute the dust temperature structure using post-processing radiative transfer to produce continuum images at cm-wavelengths. Multiple distinct rings emerge in the continuum images, corresponding to the dust traffic jams. The angular resolution of upcoming observations from SKA and ngVLA will be sufficient to detect centimeter-sized grains in protoplanetary disks and resolve the widths of dust traffic jams. Therefore, dust traffic jams resulting from the differential precession of gas and dust in misaligned circumbinary disks will be a prime target for more extended wavelength observations., Comment: 12 pages, 5 figures, accepted for publication in ApJL

We report the discovery of a rare isolated group of five dwarf galaxies located at z = 0.0086 ($D$ = 36 Mpc). All member galaxies are star-forming, blue, and gas-rich with $g-r$ indices ranging from 0.2 to 0.6 mag, and two of them show signs of ongoing mutual interaction. The most massive member of the group has a stellar mass that is half of the Small Magellanic Cloud stellar mass, and the median stellar mass of the group members is 7.87 $\times$ 10$^{7}$ M$_{\odot}$. The derived total dynamical mass of the group is $M_{\rm dyn}$ = 6.02$\times$10$^{10}$ M$_{\odot}$, whereas its total baryonic mass (stellar + HI) is 2.6$\times$10$^{9}$ M$_{\odot}$, which gives us the dynamical to baryonic mass ratio of 23. Interestingly, all galaxies found in the group are aligned along a straight line in the plane of the sky. The observed spatial extent of the member galaxies is 154 kpc, and their relative line-of-sight velocity span is within 75 km s$^{-1}$. Using the spatially resolved optical spectra provided by DESI EDR, we find that three group members share a common rotational direction. With these unique properties of the group and its member galaxies, we discuss the possible importance of such a system in the formation and evolution of dwarf galaxy groups and in testing the theory of large-scale structure formation., Comment: Accepted for publication in ApJL

The [C II] fine-structure line at 157.74 $\mu$m is one of the brightest far-infrared emission lines and an important probe of galaxy properties like the star formation rate (SFR) and the molecular gas mass ($M_{\mathrm{mol}}$). Using high-resolution numerical simulations, we test the reliability of the [C II] line as a tracer of $M_{\mathrm{mol}}$ in high-redshift galaxies and investigate secondary dependences of the [C II]-$M_{\mathrm{mol}}$ relation on the SFR and metallicity. We investigate the time evolution of the [C II] luminosity function (LF) and the relative spatial extent of [C II] emission and star formation. We post-process galaxies from the MARIGOLD simulations at redshifts $3 \le z \leq 7$ to obtain their [C II] emission. These simulations were performed with the sub-grid chemistry model, HYACINTH, to track the non-equilibrium abundances of $\mathrm{H_2}$, $\mathrm{CO}$, $\rm C$ and $\mathrm{C^+}$ on the fly. Based on a statistical sample of galaxies at these redshifts, we investigate correlations between the [C II] line luminosity, L([C II]), and the SFR, the $M_{\mathrm{mol}}$, the total gas mass and the metal mass in gas phase ($M_{\mathrm{metal}}$). We find that accounting for secondary dependencies in the L([C II])-$M_{\mathrm{mol}}$ relation improves the $M_{\mathrm{mol}}$ prediction by a factor of 2.3. The [C II] emission in our simulated galaxies shows the tightest correlation with $M_{\mathrm{metal}}$. About 20% (10%) of our simulated galaxies at $z=5$ ($z=4$) have [C II] emission extending $\geq 2$ times farther than the star formation activity. The [C II] LF evolves rapidly and is always well approximated by a double power law that does not show an exponential cutoff at the bright end. We record a 600-fold increase in the number density of L([C II]) $\sim 10^9 \, \mathrm{L_{\odot}}$ emitters in 1.4 Gyr., Comment: 25 pages, 20 figures. Submitted to A&A. Comments are welcome

The main challenges hindering the adoption of deep learning-based systems in clinical settings are the scarcity of annotated data and the lack of interpretability and trust in these systems. Concept Bottleneck Models (CBMs) offer inherent interpretability by constraining the final disease prediction on a set of human-understandable concepts. However, this inherent interpretability comes at the cost of greater annotation burden. Additionally, adding new concepts requires retraining the entire system. In this work, we introduce a novel two-step methodology that addresses both of these challenges. By simulating the two stages of a CBM, we utilize a pretrained Vision Language Model (VLM) to automatically predict clinical concepts, and a Large Language Model (LLM) to generate disease diagnoses based on the predicted concepts. We validate our approach on three skin lesion datasets, demonstrating that it outperforms traditional CBMs and state-of-the-art explainable methods, all without requiring any training and utilizing only a few annotated examples. The code is available at https://github.com/CristianoPatricio/2-step-concept-based-skin-diagnosis., Comment: Preprint submitted for review

We present an improved method for vetoing candidates of continuous gravitational-wave sources during all-sky searches utilizing the Frequency Hough pipeline. This approach leverages linear correlations between source parameters induced by the Earth Doppler effect, which can be effectively identified through the Hough Transform. Candidates that do not align with these patterns are considered spurious and can thus be vetoed, enhancing the depth and statistical significance of follow-up analyses. Additionally, we provide a comprehensive explanation of the method calibration, which intrinsically linked to the total duration of the observing run. On average, the procedure successfully vetoes $56\%$ of candidates. To assess the method performance, we conducted a Monte-Carlo simulation injecting fake continuous-wave signals into data from the third observing run of the LIGO detectors. This analysis allowed us to infer strain amplitude upper limits at a $90\%$ confidence level. We found that the optimal sensitivity is $h_0^{90\%} = 3.62^{+0.23}_{-0.22}\times 10^{-26}$ in the [128, 200] Hz band, which is within the most sensible frequency band of the LIGO detectors., Comment: 13 pages, 9 figures, 5 tables

By introducing the small noise expansion techniques, we show that the fully nonlinear (non-Markovian) stochastic inflationary system, may be re-cast in terms of an infinite set of Wiener processes (stochastic equations with white noises). As a byproduct, we show that the Starobinsky test field approximation might only provide information about the linear regime of cosmological perturbations and scalar-feld non-Gaussianities might only appear at next to leading order in slow-roll parameters., Comment: 24 pages 2 appendices

Security and RAS are two non-functional requirements under focus for current systems developed for the computing continuum. Due to the increased number of interconnected computer systems across the continuum, security becomes especially pervasive at all levels, from the smallest edge device to the high-performance cloud at the other end. Similarly, RAS (Reliability, Availability, and Serviceability) ensures the robustness of a system towards hardware defects. Namely, making them reliable, with high availability and design for easy service. In this paper and as a result of the Vitamin-V EU project, the authors detail the comprehensive approach to malware and hardware attack detection; as well as, the RAS features envisioned for future systems across the computing continuum.

Hydroelectricity accounted for roughly 66% of the total generation in Brazil in 2023 and addressed most of the intermittency of wind and solar generation. Thus, one of the most important steps in the operation planning of this country is the forecast of the natural inflow energy (NIE) time series, an approximation of the energetic value of the water inflows. To manage water resources over time, the Brazilian system operator performs long-term forecasts for the NIE to assess the water values through long-term hydrothermal planning models, which are then used to define the short-term merit order in day-ahead scheduling. Therefore, monitoring optimistic bias in NIE forecasts is crucial to prevent an optimistic view of future system conditions and subsequent riskier storage policies. In this article, we investigate and showcase strong evidence of an optimistic bias in the official NIE forecasts, with predicted values consistently exceeding the observations over the past 12 years in the two main subsystems (Southeast and Northeast). Rolling window out-of-sample tests conducted with real data demonstrate that the official forecast model exhibits a statistically significant bias of 6%, 13%, 18%, and 23% for 1, 6, 12, and 24 steps ahead in the Southeast subsystem, and 19%, 57%, 80%, and 108% in the Northeast.

We study how stronger noise can enhance the entanglement in inhomogeneously monitored quantum systems. We consider a free fermions model composed of two coupled chains - a system chain and an ancilla chain, each subject to its own different noise - and explore the dynamics of entanglement within the system chain under different noise intensities. Our results demonstrate that, contrary to the detrimental effects typically associated with noise, certain regimes of noise on the ancilla can significantly enhance entanglement within the system. Numerical simulations demonstrate the robustness of such entanglement enhancement across various system sizes and noise parameters. This enhancement is found to be highly dependent on the hopping strength in the ancilla, suggesting that the interplay between unitary dynamics and noise can be tuned to optimize the entanglement of a quantum system., Comment: 17 pages, 9 figures

We investigate the formation of dust traffic jams in polar-aligning circumbinary discs. In our first paper, we found as the circumbinary disc evolves towards a polar configuration perpendicular to the binary orbital plane, the differential precession between the gas and dust components leads to multiple dust traffic jams. These dust traffic jams evolve to form a coherent dust ring. In part two, we use 3D smoothed particle hydrodynamical simulations of gas and dust to model an initially highly misaligned circumbinary disc around the 99 Herculis (99 Her) binary system. Our results reveal that the formation of these dust rings is observed across various disc parameters, including the disc aspect ratio, viscosity, surface density power law index, and temperature power law index. The dust traffic jams are long-lived and persist even when the disc is fully aligned polar. The midplane dust-to-gas ratio within the rings can surpass unity, which may be a favourable environment for planetesimal formation. Using 2D inviscid shearing box calculations with parameters from our 3D simulations, we find streaming instability modes with significant growth rates. The streaming instability growth timescale is less than the tilt oscillation timescale during the alignment process. Therefore, the dust ring will survive once the gas disc aligns polar, suggesting that the streaming instability may aid in forming polar planets around 99 Her., Comment: 13 pages, 11 figures, Accepted for publication in MNRAS

Recent experimental and computational studies have demonstrated that nanoconfinement profoundly alters the phase behavior of water, facilitating complex phase transitions at pressures and temperatures far lower than typically observed in bulk systems. When combined with adsorption, nanoconfinement substantially enhances water uptake, primarily due to condensation occurring at the onset of the isotherm curve-a phenomenon intimately related to the facilitated formation of hydrogen bond networks. In this study, we adopt a dual approach to investigate water confined within infinite graphene slits. Our Molecular Dynamics simulations reveal hysteresis across all investigated temperatures. Unlike in finite slits, where hysteresis arises due to surface tension effects at the edges, in the case of infinite slits, the hysteresis is the result of a genuine phase transition at the nanoscale. We analyze the spatial and orientational arrangements of the water molecules, demonstrating how the graphene surface promotes the formation of a hydrogen bond network in the adjacent water layers. The remarkably low pressure required for water uptake in this nano-environment is explained at the mean-field level using a simple interacting lattice model. This is attributed to the exponential dependence of the critical pressure on the adsorbate-adsorbent interaction., Comment: 14 pages, 13 figures, with supplementary information

We introduce the first method of uncertainty quantification in the domain of Kolmogorov-Arnold Networks, specifically focusing on (Higher Order) ReLUKANs to enhance computational efficiency given the computational demands of Bayesian methods. The method we propose is general in nature, providing access to both epistemic and aleatoric uncertainties. It is also capable of generalization to other various basis functions. We validate our method through a series of closure tests, including simple one-dimensional functions and application to the domain of (Stochastic) Partial Differential Equations. Referring to the latter, we demonstrate the method's ability to correctly identify functional dependencies introduced through the inclusion of a stochastic term. The code supporting this work can be found at https://github.com/wmdataphys/Bayesian-HR-KAN, Comment: 13 pages, 7 Figures

An all-fiber dual Mach-Zehnder interferometer (MZI) based on an acoustically modulated hollow-core fiber (HCF) is experimentally demonstrated for the first time. By attaching an acoustic driver in between the fixed ends of an HCF, we fabricated two acousto-optic modulators (AOMs) with distinct driver positions, allowing for synchronizing two in-line MZIs inside the HCF. The first MZI is set by two acoustic long-period gratings separated by a second MZI formed at the fiber and driver attaching region. We show that this setup enables frequency-tuning of the coupling between the fundamental and higher-order modes in the HCF. Additionally, we simulate and analyze the HCF modal couplings and MZIs' modulated spectra under distinct device parameters using the transfer matrix method. The new AOM-MZI enables tuning of the MZIs free spectral range by adjusting 1 Hz of the electrical frequency, which is promising to modulate multiwavelength filters, sensors and fiber lasers.

Effective point cloud processing is crucial to LiDARbased autonomous driving systems. The capability to understand features at multiple scales is required for object detection of intelligent vehicles, where road users may appear in different sizes. Recent methods focus on the design of the feature aggregation operators, which collect features at different scales from the encoder backbone and assign them to the points of interest. While efforts are made into the aggregation modules, the importance of how to fuse these multi-scale features has been overlooked. This leads to insufficient feature communication across scales. To address this issue, this paper proposes the Point Pyramid RCNN (POP-RCNN), a feature pyramid-based framework for 3D object detection on point clouds. POP-RCNN consists of a Point Pyramid Feature Enhancement (PPFE) module to establish connections across spatial scales and semantic depths for information exchange. The PPFE module effectively fuses multi-scale features for rich information without the increased complexity in feature aggregation. To remedy the impact of inconsistent point densities, a point density confidence module is deployed. This design integration enables the use of a lightweight feature aggregator, and the emphasis on both shallow and deep semantics, realising a detection framework for 3D object detection. With great adaptability, the proposed method can be applied to a variety of existing frameworks to increase feature richness, especially for long-distance detection. By adopting the PPFE in the voxel-based and point-voxel-based baselines, experimental results on KITTI and Waymo Open Dataset show that the proposed method achieves remarkable performance even with limited computational headroom., Comment: 12 pages

In this paper, we investigate the presence of additive bias in Large Language Models (LLMs), drawing a parallel to the cognitive bias observed in humans where individuals tend to favor additive over subtractive changes. Using a series of controlled experiments, we tested various LLMs, including GPT-3.5 Turbo, Claude 3.5 Sonnet, Mistral, Math$\Sigma$tral, and Llama 3.1, on tasks designed to measure their propensity for additive versus subtractive modifications. Our findings demonstrate a significant preference for additive changes across all tested models. For example, in a palindrome creation task, Llama 3.1 favored adding letters 97.85% of the time over removing them. Similarly, in a Lego tower balancing task, GPT-3.5 Turbo chose to add a brick 76.38% of the time rather than remove one. In a text summarization task, Mistral 7B produced longer summaries in 59.40% to 75.10% of cases when asked to improve its own or others' writing. These results indicate that, similar to humans, LLMs exhibit a marked additive bias, which might have implications when LLMs are used on a large scale. Addittive bias might increase resource use and environmental impact, leading to higher economic costs due to overconsumption and waste. This bias should be considered in the development and application of LLMs to ensure balanced and efficient problem-solving approaches., Comment: 25 pages, 8 figures

This study conducts a bibliometric review of scientific literature on the European Union Emissions Trading System (EU ETS) from 2004 to 2024, using research articles from the Scopus database. Using the Bibliometrix R package, we analyze publication trends, key themes, influential authors, and prominent journals related to the EU ETS. Our results indicate a notable increase in research activity over the past two decades, particularly during significant policy changes and economic events affecting carbon markets. Key research focuses include carbon pricing, market volatility, and economic impacts, highlighting a shift toward financial analysis and policy implications. Thematic mapping shows cap-and-trade systems, and carbon leakage as central topics linking various research areas. Additionally, we observe key areas where further research could be beneficial, such as expanding non-parametric methodologies, deepening the exploration of macroeconomic factors, and enhancing the examination of financial market connections. Moreover, we highlight recent and innovative papers that contribute new insights, showcasing emerging trends and cutting-edge approaches within the field. This review provides insights for researchers and policymakers, highlighting the evolving landscape of EU ETS research and its relevance to global climate strategies., Comment: 22 pages, 13 figures, 4 tables

The canonical theory for planet formation in circumstellar disks proposes that planets are grown from initially much smaller seeds. The long-considered alternative theory proposes that giant protoplanets can be formed directly from collapsing fragments of vast spiral arms induced by gravitational instability -- if the disk is gravitationally unstable. For this to be possible, the disk must be massive compared to the central star: a disk-to-star mass ratio of 1/10 is widely held as the rough threshold for triggering gravitational instability, inciting significant non-Keplerian dynamics and generating prominent spiral arms. While estimating disk masses has historically been challenging, the motion of the gas can reveal the presence of gravitational instability through its effect on the disk velocity structure. Here we present kinematic evidence of gravitational instability in the disk around AB Aurigae, using deep observations of 13CO and C18O line emission with the Atacama Large Millimeter/submillimeter Array (ALMA). The observed kinematic signals strongly resemble predictions from simulations and analytic modelling. From quantitative comparisons, we infer a disk mass of up to 1/3 the stellar mass enclosed within 1" to 5" on the sky., Comment: Published in Nature. 4 main figures, 9 extended data figures, 1 extended data table. Published version at https://www.nature.com/articles/s41586-024-07877-0. Observational data products available at https://doi.org/10.11570/24.0087 with reduction guide & scripts at https://jjspeedie.github.io/guide.2021.1.00690.S. Simulated data products available at https://doi.org/10.5281/zenodo.11668694

Pair-instability supernovae (PISNe) have crucial implications for many astrophysical topics, including the search for very massive stars, the black hole mass spectrum, and galaxy chemical enrichment. To this end, we need to understand where PISNe are across cosmic time, and what are their favourable galactic environments. We present a new determination of the PISN rate as a function of redshift, obtained by combining up-to-date stellar evolution tracks from the PARSEC and FRANEC codes, with an up-to-date semi-empirical determination of the star formation rate and metallicity evolution of star-forming galaxies throughout cosmic history. We find the PISN rate to exhibit a huge dependence on the model assumptions, including the criterion to identify stars unstable to pair production, and the upper limit of the stellar initial mass function. Remarkably, the interplay between the maximum metallicity at which stars explode as PISNe, and the dispersion of the galaxy metallicity distribution, dominates the uncertainties, causing a $\sim$ seven-orders-of-magnitude PISN rate range. Furthermore, we show a comparison with the core-collapse supernova rate, and study the properties of the favourable PISN host galaxies. According to our results, the main contribution to the PISN rate comes from metallicities between $\sim 10^{-3}$ and $10^{-2}$, against the common assumption that views very-low-metallicity, Population III stars as exclusive or dominant PISN progenitors. The strong dependencies we find offer the opportunity to constrain stellar and galaxy evolution models based on possible future (or the lack of) PISN observations., Comment: Accepted for publication on MNRAS, 23 pages, 25 figures, 4 tables

The fragile-to-strong glass transition is a fascinating phenomenon that still presents many theoretical and experimental challenges. A major one is how to tune the fragility of a glass-forming liquid. Here, we study a two-dimensional (2D) system composed of vortices in a superconducting film, which effectively behaves as a 2D glass-forming liquid. We show that the kinetic fragility in this system can be experimentally varied by tuning a single parameter: the external magnetic field $H$ applied transversely to the film. This conclusion is supported by the direct comparison between the analysis of experimental measurements in an amorphous MoGe superconducting film and Monte Carlo simulations in a disordered XY model, that captures the universality class of the two-step melting transition. We show that by increasing disorder strength a fragile-to-strong transition is induced, in close similarity with the experimental findings in a magnetic field. Our numerical results shed light on the evolution of the dynamical heterogeneity from a fragile to strong glass, as due to the subtle interplay between caging effects arising from hexatic order and strong random pinning., Comment: 26 pages, 6 figures

Objective: This paper introduces a novel PET imaging methodology called 3-dimensional positron imaging (3D{\pi}), which integrates total-body (TB) coverage, time-of-flight (TOF) technology, ultra-low dose imaging capabilities, and ultra-fast readout electronics inspired by emerging technology from the DarkSide collaboration. Approach: The study evaluates the performance of 3D{\pi} using Monte Carlo simulations based on NEMA NU 2-2018 protocols. The methodology employs a homogenous, monolithic scintillator composed of liquid argon (LAr) doped with xenon (Xe) with silicon photomultipliers (SiPM) operating at cryogenic temperatures. Main results: Significant enhancements in system performance are observed, with the 3D{\pi} system achieving a noise equivalent count rate (NECR) of 3.2 Mcps which is approximately two times higher than uEXPLORER's peak NECR (1.5 Mcps) at 17.3 (kBq/mL). Spatial resolution measurements show an average FWHM of 2.7 mm across both axial positions. The system exhibits superior sensitivity, with values reaching 373 kcps/MBq with a line source at the center of the field of view. Additionally, 3D{\pi} achieves a TOF resolution of 151 ps at 5.3 kBq/mL, highlighting its potential to produce high-quality images with reduced noise levels. Significance: The study underscores the potential of 3D{\pi} in improving PET imaging performance, offering the potential for shorter scan times and reduced radiation exposure for patients. The Xe-doped LAr offers advantages such as fast scintillation, enhanced light yield, and cost-effectiveness. Future research will focus on optimizing system geometry and further refining reconstruction algorithms to exploit the strengths of 3D{\pi} for clinical applications., Comment: 8 pages, 8 figures, 6 tables

Redshift-space distortions present a significant challenge in building models for the three-point correlation function (3PCF). We compare two possible lines of attack: the streaming model and standard perturbation theory (SPT). The two approaches differ in their treatment of the non-linear mapping from real to redshift space: SPT expands this mapping perturbatively, while the streaming model retains its non-linear form but relies on simplifying assumptions about the probability density function (PDF) of line-of-sight velocity differences between pairs or triplets of tracers. To assess the quality of the predictions and the validity of the assumptions of these models, we measure the monopole of the matter 3PCF and the first two moments of the pair- and triplewise velocity PDF from a suite of N-body simulations. We also evaluate the large-scale limit of the streaming model and determine under which conditions it aligns to SPT. On scales $>10\,h^{-1}\mathrm{Mpc}$, we find that the streaming model for the 3PCF monopole is dominated by the first two velocity moments, making the exact shape of the PDF irrelevant. This model can match the accuracy of a Stage-IV galaxy survey, if the velocity moments are measured directly from the simulations. However, replacing the measurements with perturbative expressions to leading order generates large errors already on scales of $60-70 h^{-1}\mathrm{Mpc}$. This is the main drawback of the streaming model. Conversely, the SPT model for the 3PCF cannot account for the significant velocity dispersion that is present at all scales, and consequently provides predictions with limited accuracy. We demonstrate that this issue can be addressed by isolating the large-scale limit of the dispersion, which leads to typical Fingers-of-God damping functions. Overall, the SPT model with a damping function provides the best deal in terms of accuracy and computing time.

G-quadruplexes are non-canonical DNA structures rather ubiquitous in human genome, which are thought to play a crucial role in the development of 85-90 % of cancers. Here, we present a novel coarse-grained approach in modeling G-quadruplexes which accounts for their structural flexibility. We apply it to study the polymeric properties of G-quadruplex multimers, with and without crowder particles, to mimic in-vivo conditions. We find that, contrary to some suggestions found in the literature, long G-quadruplex multimers are rather flexible polymeric macromolecules, with a local persistence length comparable to monomer size, exhibiting chain stiffness variation profile consistent with a real polymer in good solvent. Moreover, in a crowded environment (up to 10% volume fraction), we report that G-quadruplex multimers exhibit an increased propensity for coiling, with a corresponding decrease in the measured chain stiffness. Accurately accounting for the polymeric properties of G4 multimers is crucial for understanding their interactions with anticancer G4-targeting drugs, thereby significantly enhancing the design and effectiveness of these drugs.

Data Science (DS) has become a cornerstone for modern software, enabling data-driven decisions to improve companies services. Following modern software development practices, data scientists use third-party libraries to support their tasks. As the APIs provided by these tools often require an extensive list of arguments to be set up, data scientists rely on default values to simplify their usage. It turns out that these default values can change over time, leading to a specific type of breaking change, defined as Default Argument Breaking Change (DABC). This work reveals 93 DABCs in three Python libraries frequently used in Data Science tasks -- Scikit Learn, NumPy, and Pandas -- studying their potential impact on more than 500K client applications. We find out that the occurrence of DABCs varies significantly depending on the library; 35% of Scikit Learn clients are affected, while only 0.13% of NumPy clients are impacted. The main reason for introducing DABCs is to enhance API maintainability, but they often change the function's behavior. We discuss the importance of managing DABCs in third-party DS libraries and provide insights for developers to mitigate the potential impact of these changes in their applications., Comment: 38 pages, 3 figures, 13 tables

In this work, we obtain estimates for the upper bound of gaps between consecutive eigenvalues for the eigenvalue problem of a class of second-order elliptic differential operators in divergent form, with Dirichlet boundary conditions, in a limited domain of n-dimensional Euclidean space. This class of operators includes the well-known Laplacian and the square Cheng-Yau operator. For the Laplacian case, our estimate coincides with that obtained by D. Chen, T. Zheng, and H. Yang, which is the best possible in terms of the order of the eigenvalues. For pinched Cartan-Hadamard manifolds the estimates were made in particular cases of this operator.

We develop the theory of groupoid graded semisimple rings. Our rings are neither unital nor one-sided artinian. Instead, they exhibit a strong version of having local units and being locally artinian, and we call them $\Gamma_0$-artinian. One of our main results is a groupoid graded version of the Wedderburn-Artin Theorem, where we characterize groupoid graded semisimple rings as direct sums of graded simple $\Gamma_0$-artinian rings and we exhibit the structure of this latter class of rings. In this direction, we also prove a groupoid graded version of Jacobson-Chevalley density theorem. We need to define and study properties of groupoid gradings on matrix rings (possibly of infinite size) over groupoid graded rings, and specially over groupoid graded division rings. Because of that, we study groupoid graded division rings and their graded modules. We consider a natural notion of freeness for groupoid graded modules that, when specialized to group graded rings, gives the usual one, and show that for a groupoid graded division ring all graded modules are free (in this sense). Contrary to the group graded case, there are groupoid graded rings for which all graded modules are free according to our definition, but they are not graded division rings. We exhibit an easy example of this kind of rings and characterize such class among groupoid graded semisimple rings. We also relate groupoid graded semisimple rings with the notion of semisimple category defined by B. Mitchell. For that, we show the link between functors from a preadditive category to abelian groups and graded modules over the groupoid graded ring associated to this category, generalizing a result of P. Gabriel. We characterize simple artinian categories and categories for which every functor from them to abelian groups is free in the sense of B. Mitchell., Comment: 87 pages

In pathology, accurate and efficient analysis of Hematoxylin and Eosin (H\&E) slides is crucial for timely and effective cancer diagnosis. Although many deep learning solutions for nuclei instance segmentation and classification exist in the literature, they often entail high computational costs and resource requirements, thus limiting their practical usage in medical applications. To address this issue, we introduce a novel convolutional neural network, NuLite, a U-Net-like architecture designed explicitly on Fast-ViT, a state-of-the-art (SOTA) lightweight CNN. We obtained three versions of our model, NuLite-S, NuLite-M, and NuLite-H, trained on the PanNuke dataset. The experimental results prove that our models equal CellViT (SOTA) in terms of panoptic quality and detection. However, our lightest model, NuLite-S, is 40 times smaller in terms of parameters and about 8 times smaller in terms of GFlops, while our heaviest model is 17 times smaller in terms of parameters and about 7 times smaller in terms of GFlops. Moreover, our model is up to about 8 times faster than CellViT. Lastly, to prove the effectiveness of our solution, we provide a robust comparison of external datasets, namely CoNseP, MoNuSeg, and GlySAC. Our model is publicly available at https://github.com/CosmoIknosLab/NuLite

Underground gas storage is a versatile tool for managing energy resources and addressing pressing environmental concerns. While natural gas is stored in geological formations since the beginning of the 20th century, hydrogen has recently been considered as a potential candidate toward a more flexible and sustainable energy infrastructure. Furthermore, these formations can also be used to sequester environmentally harmful gases such as CO2. When such operations are implemented in faulted basins, however, safety concerns may arise due to the possible reactivation of pre-existing faults, which could result in (micro)-seismicity events. In the Netherlands, it has been recently noted that fault reactivation can occur "unexpectedly" during the life of an underground gas storage (UGS) site, even when stress conditions are not expected to cause a failure. The present two-part work aims to develop a modeling framework to investigate the physical mechanisms causing such occurrences and define a safe operational bandwidth for pore pressure variation for UGS operations in the faulted reservoirs of the Rotliegend formation, the Netherlands. In this follow-up paper, we investigate in detail the mechanisms and crucial factors that result in fault reactivation at various stages of a UGS. The mathematical and numerical model described in Part I is used, also accounting for the effect of geochemical dissolution on reservoir and caprock weakening. TThe study investigates the risks of fault activation caused by the storage of different fluids for various purposes, such as long-term CO2 sequestration, CH4 and N2 injection and extraction cycles, and N2 permanent storage. The results show how geomechanical properties and reservoir operating conditions may increase the risk of fault reactivation at various UGS stages. Finally, operational guidelines for improving secure storage operations are presented., Comment: 31 pages, 24 figures, 4 tables

Gravitational waves from sub-solar mass primordial black holes could be detected in LIGO, Virgo and KAGRA data. Here, we apply a method originally designed to look for rapidly spinning-down neutron stars, the generalized frequency-Hough transform, to search for planetary-mass primordial black holes using data from the first half of the third observing run of advanced LIGO. In this companion paper to arXiv:2402.19468, in which the main results of our search are presented, we delve into the details of the search methodology, the choices we have made regarding the parameter space to explore, the follow-up procedure we use to confirm or reject possible candidates returned in our search, and a comparison of our analytic procedure of generating upper limits to those obtained through injections., Comment: 13 pages, 10 figures -- comments welcome! A portion of this article draws from the appendix of arXiv:2402.19468v1, which is not present in the latest version (v2)

Irradiations at Ultra High Dose Rate (UHDR) regimes, exceeding 40 Gy/s in single fractions lasting less than 200 ms, have shown an equivalent antitumor effect compared to conventional radio-therapy with reduced harm to normal tissues. This work details the hardware and software modi-fications implemented to deliver 10 MeV UHDR electron beams with a Linear Accelerator Elekta SL 18 MV and the beam characteristics obtained. GafChromic EBT XD films and an Advanced Markus chamber were used for the dosimetry characterization, while a silicon sensor assessed the machine's beam pulses stability and repeatability. Dose per pulse, average dose rate and instantaneous dose rate in the pulse were evaluated for four experimental settings, varying the source-to-surface dis-tance and the beam collimation, i.e. with and without the use of a cylindrical applicator. Results showed dose per pulse from 0.6 Gy to a few tens of Gy and average dose rate up to 300 Gy/s. The obtained results demonstrate the possibility to perform in-vitro radiobiology experiments and test of new technologies for beam monitoring and dosimetry at the upgraded LINAC, thus contributing to the electron UHDR research field.

This paper studies the evolution of economic activities using a continuous time-space aggregation-diffusion model, which encompasses competing effects of agglomeration and congestion. To bring the model to the real data, a novel discretization technique over time and space is introduced. This technique effectively disentangles spatial effects into pure topography, agglomeration, repulsion, and diffusion forces, which is crucial for developing robust econometric methods in spatial economics. Our empirical analysis of personal income across Italian municipalities from 2008 to 2019 validates the model's primary predictions and demonstrates superior performance compared to the most common spatial econometric models in the literature.