Search

Recent developments in technology have accelerated the influence of artificial intelligence (AI) on our lives. The ability of generative-AI tools such as ChatGPT, Gemini and Claude to both 'write' and 'read' texts in a human-like manner means they are set to play an increasingly important role in the literacy lives of children, young people and adults. The National Literacy Trust is interested in exploring how such platforms might influence, and potentially redefine, what it means to be literate in the digital age. This summary report explores children, young people's and teachers' attitudes, behaviour and confidence around using generative AI to support literacy and learning. Findings are based on data from more than 50,000 children and young people taking part in the Annual Literacy Survey, with a focus on more than 15,000 young people aged 13 to 18 and 1,228 teachers from primary and secondary schools across the UK.

Recent developments in technology have accelerated the influence of artificial intelligence (AI) on our lives. The ability of generative-AI tools such as ChatGPT, Gemini and Claude to both 'write' (generate new texts) and 'read' (e.g. summarise texts) in a human-like manner means they are set to play an increasingly important role in the literacy lives of children, young people and adults. Anticipating this, the National Literacy Trust is exploring how such platforms might influence, and even redefine, what it means to be literate in the digital age. Based on data from the National Literacy Trust's Annual Literacy Survey, this report explores teachers' attitudes, behaviour and confidence around using AI to support literacy, as well as complementing similar research with children and young people carried out in 2023 and 2024.

Recent developments in technology have accelerated the influence of artificial intelligence (AI) on our lives. The National Literacy Trust is interested in exploring how such platforms might influence, and potentially redefine, what it means to be literate in the digital age. Based on data from more than 50,000 children and young people taking part in the National Literacy Trust's Annual Literacy Survey, and with a focus on more than 15,000 young people aged 13 to 18, this report explores young people's attitudes, behaviour and confidence around using generative AI to support literacy and learning. Primarily, findings show that awareness and use of these tools by children and young people has increased rapidly in the last year, with 3 in 4 young people aged 13 to 18 saying they'd used them in 2024. The survey also provides evidence that many young people are using generative AI to support literacy.

This report builds on National Literacy Trust's previous research from the 2024 Annual Literacy Survey to investigate how children and young people felt about writing in early 2024. It includes findings on how many enjoyed writing and how often they wrote in their free time, what motivated them to write, and what they wrote. This report is based on 76,131 responses from children and young people aged 5 to 18 in schools across the UK in early 2024 and explores responses by age, gender, socioeconomic background, and geographical region. Findings show that children and young people's enjoyment of writing, and frequency of writing in their free time, is at an unprecedented low. Increasing evidence of a long-term downward trend calls for urgent action to reconnect children and young people with writing that promotes connection with creativity, self-expression and mental wellbeing.

Weyl semimetals have a variety of intriguing physical properties, including topologically protected electronic states that coexist with conducting states. Possible exploitation of topologically protected states in a conducting material is promising for technological applications. Weyl semimetals that form in a non-centrosymmetric structure that also contain magnetic moments may host a variety of emergent phenomena that cannot be seen in magnetic, centrosymmetric Weyl materials. It can be difficult to distinguish definitively between a centrosymmetric structure and one of its non-centrosymmetric subgroups with standard powder X-ray diffractometers in cases where two atoms in the compound have nearly the same atomic number, as is the case for the candidate Weyl semimetal CeGaGe. In these cases, a careful single-crystal neutron diffraction experiment with high-angle reflections provides complimentary information to X-ray diffraction and definitively resolves any ambiguity between centrosymmetric and non-centrosymmetric crystal structures. Single-crystal neutron diffraction measurements on the candidate Weyl semimetal CeGaGe confirms that its structure is non-centrosymmetric, described by space group 109 $\left(I4_1md\right)$ rather than the centrosymmetric space group 141 $\left(I4_1/amd\right)$. There are many high-angle reflections in the data set that give clear, physically intuitive evidence that CeGaGe forms with $I4_1md$ symmetry since Bragg planes of these reflections can contain Ga with no Ge or vice versa whereas the Bragg planes for a structure with $I4_1/amd$ symmetry would have a mix of Ga and Ge. Further, in some crystals we have studied, there is clear evidence for a structural transition from body-centered $I4_1md$ symmetry to primitive $P4_3$ and/or $P4_1$ symmetry.

We study a model of colored multiwebs, which generalizes the dimer model to allow each vertex to be adjacent to $n_v$ edges. These objects can be formulated as a random tiling of a graph with partial dimer covers. We examine the case of a cycle graph, and in particular we describe the local correlations of tiles in this setting., Comment: 20 pages, 4 figures

We present a detailed exploration of the most optically-luminous Little Red Dot ($L_{H\alpha}=10^{44}$erg/s, $L_V=10^{45}$erg/s, F444W=22AB) found to date. Located in the Abell 2744 field, source A744-45924 was observed by NIRSpec/PRISM with ultradeep spectroscopy reaching SNR$\sim$100pix$^{-1}$, high-resolution 3-4 micron NIRCam/Grism spectroscopy, and NIRCam Medium Band imaging. The NIRCam spectra reveal high rest-frame EW $W_{H\alpha,0,broad}>800$\r{A}, broad H$\alpha$ emission (FWHM$\sim$4500 km/s), on top of narrow, complex absorption. NIRSpec data show exceptionally strong rest-frame UV to NIR Fe II emission ($W_{FeII-UV,0}\sim$340\r{A}), N IV]$\lambda\lambda$1483,1486 and N III]$\lambda$1750, and broad NIR O I $\lambda$8446 emission. The spectra unambiguously demonstrate a broad-line region associated with an inferred $M_{BH}\sim10^9M_\odot$ supermassive black hole embedded in dense gas, which might explain a non-detection in ultradeep Chandra X-ray data (>$10\times$ underluminous relative to broad $L_{H\alpha}$). Strong UV Nitrogen lines suggest supersolar N/O ratios due to rapid star formation or intense radiation near the AGN. The continuum shows a clear Balmer break at rest-frame 3650\r{A}, which cannot be accounted for by an AGN power-law alone. A stellar population model produces an excellent fit with a reddened Balmer break and implying a massive ($M_*\sim8\times10^{10}M_\odot$), old $\sim$500 Myr, compact stellar core, among the densest stellar systems known ($\rho\sim3\times10^6M_\odot$/pc$^2$ for $R_{e,opt}=70\pm10$ pc), and AGN emission with extreme intrinsic EW $W_{H\alpha,0}\gg$1000\r{A}. However, although high $M_*$ and $M_{BH}$ are supported by evidence of an overdensity containing 40 galaxies at $z=4.41-4.51$, deep high-resolution spectroscopy is required to confirm stellar absorption and rule out that dense gas around the AGN causes the Balmer break instead., Comment: 28 pages,10 figures, submitted to ApJ

We propose a mathematical model for tumor invasion supported by angiogenesis and interactions with the surrounding tissue. For the model deduction we employ a multiscale approach starting from lower scales and obtaining by an informal parabolic upscaling a system of reaction-diffusion-taxis equations with a so-called 'taxis cascade', where one species is performing taxis towards a signal whose production/decay is controled by the other, for which it also serves as a tactic cue. We prove global existence and uniqueness of solutions to the obtained PDE-ODE system and perform numerical simulations to illustrate the behavior of solutions.

In this paper, we develop the first comprehensive tutorial on designing future 6G networks that synergistically integrate notions of resilience and criticality from the ground up. While resilience refers to the ability to absorb, adapt to, and recover from adversarial or challenging conditions, criticality indicates the degree of importance or urgency assigned to a particular service or component. Despite a spiking interest in designing resilient wireless networks, most prior works do not provide a unified resilience definition, nor harness the intricate interplay between resilience and criticality. In order to fill this gap, in this paper, we highlight the importance of a criticality-aware approach as a key enabler for providing reliable and resilient service functionality. Moreover, we delve into the unique challenges and opportunities of the envisioned 6G features pertaining to resilience and (mixed) criticality. After reviewing resilience definitions, we present a core resilience strategy, a unified resilience metric, different criteria for service criticality, and prioritization frameworks, that augment the 6G resilience prospects. Afterwards, we explore the opportunities presented by promising technologies that enable a resilient 6G network design from a radio access network protocol stack perspective. We briefly revisit state-of-the-art network architectures, establish a rough connection to the Open-RAN Alliance vision, and discuss opportunities, existing techniques, and promising enabling mechanisms for 6G at each layer. Finally, the article discusses important research directions and open problems concerning resilience and criticality in 6G., Comment: 23 pages, 7 figures. Submitted to IEEE for possible publication

Adequately generating and evaluating prediction models based on supervised machine learning (ML) is often challenging, especially for less experienced users in applied research areas. Special attention is required in settings where the model generation process involves hyperparameter tuning, i.e. data-driven optimization of different types of hyperparameters to improve the predictive performance of the resulting model. Discussions about tuning typically focus on the hyperparameters of the ML algorithm (e.g., the minimum number of observations in each terminal node for a tree-based algorithm). In this context, it is often neglected that hyperparameters also exist for the preprocessing steps that are applied to the data before it is provided to the algorithm (e.g., how to handle missing feature values in the data). As a consequence, users experimenting with different preprocessing options to improve model performance may be unaware that this constitutes a form of hyperparameter tuning - albeit informal and unsystematic - and thus may fail to report or account for this optimization. To illuminate this issue, this paper reviews and empirically illustrates different procedures for generating and evaluating prediction models, explicitly addressing the different ways algorithm and preprocessing hyperparameters are typically handled by applied ML users. By highlighting potential pitfalls, especially those that may lead to exaggerated performance claims, this review aims to further improve the quality of predictive modeling in ML applications.

Scanning Electron Microscopy (SEM) experiments provide detailed insights into material microstructures, enabling high-resolution imaging as well as crystallographic analysis through advanced techniques like Electron Backscatter Diffraction (EBSD). However, the interaction of the high-energy electron beam with the material can lead to localized heating, which may significantly impact specimen integrity, especially in applications requiring prolonged beam exposure, for instance when mapping the crystal structure using EBSD. This study examines electron-beam-induced heating effects on a model metal sample (iron), directly measuring the locally deposited electron beam energy with a MEMS-based heating device and validating these measurements through simulations, including Monte Carlo and Finite Element methods. The analysis focuses on the effects of various experimental parameters such as acceleration voltage (from 5 to 30 kV), beam current (from 0.17 nA to 22 nA), dwell time (from 1$\mu$s to 1ms) and sample tilt (0{\deg} to 70{\deg}). The findings reveal that local sample temperatures can increase by up to 70 {\deg}C during EBSD experiments, primarily affected by the choice in beam current and acceleration voltage, with beam current having the most significant impact.

JWST observations have unveiled faint active galactic nuclei (AGN) at high-redshift that provide insights on the formation of supermassive black holes (SMBHs) and their coevolution with galaxies. However, disentangling stellar from AGN light in these sources is challenging. Here, we use an empirical approach to infer the average stellar mass of 6 faint broad line (BL) Halpha emitters at z = 4 - 5 with BH masses ~ 6 (4 - 15)x10^6 Msun, with a method independent of their spectral energy distribution (SED). We use the deep JWST/NIRcam grism survey ALT to measure the over-densities around BL-Halpha emitters and around a spectroscopic reference sample of ~300 galaxies. In our reference sample, we find that Mpc-scale over-density correlates with stellar mass, while pair counts are flat below ~50 kpc due to satellites. Their large-scale environments suggest that BL-Halpha emitters are hosted by galaxies with stellar masses ~5x10^7 Msun, ~40 times lower than those inferred from galaxy-only SED fits. Adding measurements around more luminous z~6 AGNs, we find tentative correlations between line width, BH mass and the over-density, suggestive of a steep BH to halo mass relation. The main implications are (1) when BH masses are taken at face value, we confirm extremely high BH to stellar mass ratios of ~10 %, (2) the low stellar mass galaxies hosting growing SMBHs are in tension with typical hydrodynamical simulations, except those without feedback, (3) a 1 % duty cycle implied by the host mass hints at super-Eddington accretion, which may imply over-estimated SMBH masses, (4) the masses are at odds with a high stellar density interpretation of the line broadening, (5) our results imply a diversity of galaxy masses, environments and SEDs among AGN samples, depending on their luminosity., Comment: 19 pages, 15 figures. Submitted to ApJ. Main results in Figures 12, 13, 14. Comments welcome

We present the detection and characterization of TOI-4994b (TIC 277128619b), a warm Saturn-sized planet discovered by the NASA Transiting Exoplanet Survey Satellite (TESS). TOI-4994b transits a G-type star (V = 12.6 mag) with a mass, radius, and effective temperature of $M_{\star} =1.005^{+0.064}_{-0.061} M_{\odot}$, $R_{\star} = 1.055^{+0.040}_{-0.037} R_{\odot}$, and $T_{\rm eff} = 5640 \pm 110$ K. We obtained follow-up ground-based photometry from the Las Cumbres Observatory (LCO) and the Antarctic Search for Transiting ExoPlanets (ASTEP) telescopes, and we confirmed the planetary nature of TOI-4994b with multiple radial velocity observations from the PFS, CHIRON, HARPS, FEROS, and CORALIE instruments. From a global fit to the photometry and radial velocities, we determine that TOI-4994b is in a 21.5-day, eccentric orbit ($e = 0.32 \pm 0.04$) and has a mass of $M_{P}= 0.280^{+0.037}_{-0.034} M_{J}$, a radius of $R_{P}= 0.762^{+0.030}_{-0.027}R_{J}$, and a Saturn-like bulk density of $\rho_{p} = 0.78^{+0.16}_{-0.14}$ $\rm g/cm^3$. We find that TOI-4994 is a potentially viable candidate for follow-up stellar obliquity measurements. TOI-4994b joins the small sample of warm Saturn analogs and thus sheds light on our understanding of these rare and unique worlds., Comment: 17 pages, 10 figures. Accepted to AJ

3D open-vocabulary scene graph methods are a promising map representation for embodied agents, however many current approaches are computationally expensive. In this paper, we reexamine the critical design choices established in previous works to optimize both efficiency and performance. We propose a general scene graph framework and conduct three studies that focus on image pre-processing, feature fusion, and feature selection. Our findings reveal that commonly used image pre-processing techniques provide minimal performance improvement while tripling computation (on a per object view basis). We also show that averaging feature labels across different views significantly degrades performance. We study alternative feature selection strategies that enhance performance without adding unnecessary computational costs. Based on our findings, we introduce a computationally balanced approach for 3D point cloud segmentation with per-object features. The approach matches state-of-the-art classification accuracy while achieving a threefold reduction in computation.

A central problem in the study of open quantum systems is the characterisation of non-Markovian processes, where an environment retains memory of its interaction with the system. A key distinction is whether or not this memory can be simulated classically, as this can lead to efficient modelling and noise mitigation. Powerful tools have been developed recently within the process matrix formalism, a framework that conveniently characterises all multi-time correlations through a sequence of measurements. This leads to a detailed classification of classical and quantum-memory processes and provides operational procedures to distinguish between them. However, these results leave open the question of what type of system-environment interactions lead to classical memory. More generally, process-matrix methods lack a direct connection to joint system-environment evolution, a cornerstone of open-system modelling. In this work, we characterise Hamiltonian and circuit-based models of system-environment interactions leading to classical memory. We show that general time-dependent Hamiltonians with product eigenstates, and where the environment's eigenstates form a time-independent, orthonormal basis, always produce a particular type of classical memory: probabilistic mixtures of unitary processes. Additionally, we show that the most general type of classical-memory processes can be generated by a quantum circuit in which system and environment interact through a specific class of controlled unitaries. Our results establish the first strong link between process-matrix methods and traditional Hamiltonian-based approaches to open quantum systems., Comment: 13 pages, 6 figures

As mission- and safety-critical wireless applications grow in complexity and diversity, next-generation wireless systems must meet increasingly stringent and multifaceted requirements. These systems demand resilience along with enhanced intelligence and adaptability to ensure reliable communication under diverse conditions. This paper proposes an event-based multi-stage resilience framework, offering a guideline for efficiently integrating a combination of error mitigation techniques. The framework is applied to a case study focusing on uplink transmission of mixed-criticality data in the presence of random link blockages. The proposed scheme combines multiple blockage mitigation strategies - rate-splitting multiple access (RSMA), one-sided access point cooperation, and central decoding - within an event-driven algorithm. Each method, increasing in effectiveness and complexity, is activated sequentially to systematically overcome blockages. We model a mixed-criticality queuing system and formulate two transmit power allocation problems, one for separate decoding and one for central decoding, to ensure queue stability and fairness. Simulations evaluate the delay performance under varying blockage durations and examine the cost tradeoffs among resilience mechanisms within the proposed framework. The results suggest that passive robustness strategies effectively handle frequent short-term fluctuations, while more complex adaptation becomes germane for rare and prolonged blockages. Additionally, the results emphasize the importance of criticality-awareness for resilient communication design., Comment: 13 pages, 8 figures, submitted to IEEE Transactions on Wireless Communications

Liquid composites moulding is an important manufacturing technology for fibre reinforced composites, due to its cost-effectiveness. Challenges lie in the optimisation of the process due to the lack of understanding of key characteristic of textile fabrics - permeability. The problem of computing the permeability coefficient can be modelled as the well-known Stokes-Brinkman equation, which introduces a heterogeneous parameter $\beta$ distinguishing macropore regions and fibre-bundle regions. In the present work, we train a Fourier neural operator to learn the nonlinear map from the heterogeneous coefficient $\beta$ to the velocity field $u$, and recover the corresponding macroscopic permeability $K$. This is a challenging inverse problem since both the input and output fields span several order of magnitudes, we introduce different regularization techniques for the loss function and perform a quantitative comparison between them., Comment: 23 pages, 7 figures

The separation of colloidal particles is of great importance in many fields, such as chemistry, medicine, and chemical engineering. However, separating particles based on their surface physico-chemical properties remains challenging. Diffusiophoresis and diffusioosmosis have been used for the continuous separation of colloids based on particle size and zeta potential. Here, we demonstrate that diffusiophoresis and diffusioosmosis in electrolyte streams enable the separation of carboxylate polystyrene particles with similar sizes and zeta potentials but distinct surface concentrations of carboxyl groups. A double junction microfluidic device, fed with low and high electrolyte concentration streams, was used to create steady-state salt concentration gradients. As the streams mix along the channel, the particles are exposed to environments with varying salinity levels, and their dynamics are influenced by the sensitivity of their zeta potential to the local salt concentration. Typically, for carboxylate polystyrene particles, the magnitude of the electrophoretically measured zeta potential decreases with increasing salt concentrations. However, surface conductance effects, which depend on the surface concentration of the carboxyl groups, may lead to an opposite sensitivity of the zeta potential to salt. Therefore, under selected salt conditions, similarly sized polystyrene colloids with comparable zeta potentials in the low concentration stream, but different surface concentrations of carboxyl groups, can be separated in a continuous flow process when they exhibit opposite zeta potential sensitivities to salt. This mechanism has discipline-spanning potential for the continuous separation of colloids distinguished solely by surface physico-chemical properties that influence their zeta potential sensitivities, like roughness, permeability, heterogeneity, and chemical composition., Comment: 12 pages, 6 figures

Activity and autonomous motion are fundamental aspects of many living and engineering systems. Here, the scale of biological agents covers a wide range, from nanomotors, cytoskeleton, and cells, to insects, fish, birds, and people. Inspired by biological active systems, various types of autonomous synthetic nano- and micromachines have been designed, which provide the basis for multifunctional, highly responsive, intelligent active materials. A major challenge for understanding and designing active matter is their inherent non-equilibrium nature due to persistent energy consumption, which invalidates equilibrium concepts such as free energy, detailed balance, and time-reversal symmetry. Furthermore, interactions in ensembles of active agents are often non-additive and non-reciprocal. An important aspect of biological agents is their ability to sense the environment, process this information, and adjust their motion accordingly. It is an important goal for the engineering of micro-robotic systems to achieve similar functionality. With many fundamental properties of motile active matter now reasonably well understood and under control, the ground is prepared for the study of physical aspects and mechanisms of motion in complex environments, of the behavior of systems with new physical features like chirality, of the development of novel micromachines and microbots, of the emergent collective behavior and swarming of intelligent self-propelled particles, and of particular features of microbial systems. The vast complexity of phenomena and mechanisms involved in the self-organization and dynamics of motile active matter poses major challenges, which can only be addressed by a truly interdisciplinary effort involving scientists from biology, chemistry, ecology, engineering, mathematics, and physics.

Content moderation and toxicity classification represent critical tasks with significant social implications. However, studies have shown that major classification models exhibit tendencies to magnify or reduce biases and potentially overlook or disadvantage certain marginalized groups within their classification processes. Researchers suggest that the positionality of annotators influences the gold standard labels in which the models learned from propagate annotators' bias. To further investigate the impact of annotator positionality, we delve into fine-tuning BERTweet and HateBERT on the dataset while using topic-modeling strategies for content moderation. The results indicate that fine-tuning the models on specific topics results in a notable improvement in the F1 score of the models when compared to the predictions generated by other prominent classification models such as GPT-4, PerspectiveAPI, and RewireAPI. These findings further reveal that the state-of-the-art large language models exhibit significant limitations in accurately detecting and interpreting text toxicity contrasted with earlier methodologies. Code is available at https://github.com/aheldis/Toxicity-Classification.git.

The growth of semiconductor technology is unprecedented, with profound transformational consequences for society. This includes feeding an over-reliance on digital solutions to systemic problems such as climate change ('techno-solutionism'). Such technologies come at a cost: environmental, social and material. We unpack topics arising from "The True Cost of ICT: From Materiality to Techno-Solutionism (TCICT)", a workshop held at the International ICT for Sustainability (ICT4S) conference 2024 in Stockholm, Sweden -- exploring, as a matter of global climate injustice, the drivers and material dependencies of these technologies. We point to the importance of addressing ICT's impacts as a system, rather than purely in terms of efficiency and energy use. We conclude by calling to build a community of like-minded and critical colleagues to address the intersectional climate impacts of the semiconductor industry and the techno-solutionism it embodies., Comment: This position paper and extended abstract is accepted for presentation at LOCO '24: 1st International Workshop on Low Carbon Computing, 2024-12-03, in Glasgow, UK

The UV continuum slope of galaxies, $\beta$, is a powerful diagnostic. Understanding the redshift evolution of $\beta$ and its dependence on key galaxy properties can shed light on the evolution of galaxy physical properties over cosmic time. In this study, we present $\beta$ measurements for 295 spectroscopically confirmed galaxies at $5.5 9.5$ to begin to redden, deviating from the trend observed at $z < 9.5$. By producing stacked spectra in bins of redshift and $\beta$, we derive trends between $\beta$ and dust attenuation, metallicity, ionization parameter, and stellar age indicators directly from spectra, finding a lack of dust attenuation to be the dominant driver of bluer $\beta$ values. We further report six galaxies with $\beta<-3.0$, which show a range of spectroscopic properties and signs of significant LyC photon leakage. Finally, we show that the redder $\beta$ values at $z > 9.5$ may require rapid build-up of dust reservoirs in the very early Universe or a significant contribution from the nebular continuum emission to the observed UV spectra, with the nebular continuum fraction depending on the gas temperatures and densities. Our modeling shows that in the absence of dust, nebular emission at $T > 15,000$ K can reproduce the range of $\beta$ that we see in our sample. Higher gas temperatures driven by hot, massive stars can boost the fraction of nebular continuum emission, potentially explaining the observed $\beta$ values as well as bright UV magnitudes seen across galaxies at $z > 10$., Comment: submitted to MNRAS, 22 pages including 1 appendix, 12 Figures (including 1 in appendix), comments welcome!

We introduce the notions of weighted extremal K{\"a}hler twins together with the related notion of extremal Sasaki twins. In the K\"ahler setting this leads to a generalization of the twinning phenomenon appearing among LeBrun's strongly Hermitian solutions to the Einstein-Maxwell equations on the first Hirzebruch surface \cite{Leb16} to weighted extremal metrics on Hirzebruch surfaces in general. We discover that many twins appear and that this can be viewed in the Sasaki setting as a case where we have more than one extremal ray in the Sasaki cone even when we do not allow changes within the isotopy class. We also study extremal Sasaki twins directly in the Sasaki setting with a main focus on the toric Sasaki case., Comment: 35 pages, 3 figures

Understanding and predicting the structure and evolution of coronal mass ejections (CMEs) in the heliosphere remains one of the most sought-after goals in heliophysics and space weather research. A powerful tool for improving current knowledge and capabilities consists of multi-spacecraft observations of the same event, which take place when two or more spacecraft fortuitously find themselves in the path of a single CME. Multi-probe events can not only supply useful data to evaluate the large-scale of CMEs from 1D in-situ trajectories, but also provide additional constraints and validation opportunities for CME propagation models. In this work, we analyse and simulate the coronal and heliospheric evolution of a slow, streamer-blowout CME that erupted on 23 September 2021 and was encountered in situ by four spacecraft approximately equally distributed in heliocentric distance between 0.4 and 1 au. We employ the Open Solar Physics Rapid Ensemble Information (OSPREI) modelling suite in ensemble mode to predict the CME arrival and structure in a hindcast fashion and to compute the "best-fit" solutions at the different spacecraft individually and together. We find that the spread in the predicted quantities increases with heliocentric distance, suggesting that there may be a maximum (angular and radial) separation between an inner and an outer probe beyond which estimates of the in-situ magnetic field orientation (parameterised by flux rope model geometry) increasingly diverge. We discuss the importance of these exceptional observations and the results of our investigation in the context of advancing our understanding of CME structure and evolution as well as improving space weather forecasts., Comment: 20 pages, 12 figures, 3 tables; accepted for publication in MNRAS

The James Webb Space Telescope has uncovered a puzzling population of UV-faint broad-line active galactic nuclei (AGN), nicknamed ``Little Red Dots'' (LRD) owing to their compact morphology and red rest-frame optical colours. Interpreted as dust attenuated AGN, their inferred intrinsic luminosities and supermassive black hole (SMBH) masses rival those of UV-luminous quasars, although they are $>100$ times more abundant. If LRDs and quasars are members of the same underlying population, they should inhabit comparable mass dark matter halos, traced by similar overdensities of galaxies. Otherwise, they represent distinct populations with different physical properties and formation histories. Characterizing LRD environments thus provides a critical test of their nature. Here, we report the discovery of a LRD at $z=7.3$, attenuated by moderate amounts of dust, $A_V = {3.26}\,\rm{mag}$, with an intrinsic bolometric luminosity of $10^{46.7}\,\rm{erg}\,\rm{s}^{-1}$ and a SMBH mass of $7\times10^8\,\rm{M}_\odot$. Most notably, this object is embedded in an overdensity of eight nearby galaxies, allowing us to calculate the first spectroscopic estimate of the clustering of galaxies around LRDs. We find a LRD-galaxy cross-correlation length of $r_0\!=\!9\pm2\,\rm{h}^{-1}\,\rm{cMpc}$, comparable to that of $z\!\sim\!6$ UV-luminous quasars. The resulting estimate of their minimum dark matter halo mass of $\log_{10}(M_{\rm{halo, min}}/\rm{M}_{\odot})= 12.3_{-0.8}^{+0.7}$ indicates that nearly all halos above this mass must host actively accreting SMBHs at $z\approx7$, in strong contrast with the far smaller duty cycle of luminous quasars ($<1\%$). Our results, taken at face value, motivate a picture in which LRDs are the obscured counterparts of UV-luminous quasars, which provides a natural explanation for the short UV-luminous lifetimes inferred from both quasar clustering and quasar proximity zones., Comment: Submitted. Comments welcome!

Numerous challenges persist in High Energy Physics (HEP), the addressing of which requires advancements in detection technology, computational methods, data analysis frameworks, and phenomenological designs. We provide a concise yet comprehensive overview of recent progress across these areas, in line with advances in quantum technology. We will discuss the potential of quantum devices in detecting subtle effects indicative of new physics beyond the Standard Model, the transformative role of quantum algorithms and large-scale quantum computers in studying real-time non-perturbative dynamics in the early universe and at colliders, as well as in analyzing complex HEP data. Additionally, we emphasize the importance of integrating quantum properties into HEP experiments to test quantum mechanics at unprecedented high-energy scales and search for hints of new physics. Looking ahead, the continued integration of resources to fully harness these evolving technologies will enhance our efforts to deepen our understanding of the fundamental laws of nature., Comment: 27 pages, 7 figures

Identifying amyloid-beta positive patients is crucial for determining eligibility for Alzheimer's disease (AD) clinical trials and new disease-modifying treatments, but currently requires PET or CSF sampling. Previous MRI-based deep learning models for predicting amyloid positivity, using only T1w sequences, have shown moderate performance. We trained deep learning models to predict amyloid PET positivity and evaluated whether multi-contrast inputs improve performance. A total of 4,058 exams with multi-contrast MRI and PET-based quantitative amyloid deposition were obtained from three public datasets: the Alzheimer's Disease Neuroimaging Initiative (ADNI), the Open Access Series of Imaging Studies 3 (OASIS3), and the Anti-Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4). Two separate EfficientNet models were trained for amyloid positivity prediction: one with only T1w images and the other with both T1w and T2-FLAIR images as network inputs. The area under the curve (AUC), accuracy, sensitivity, and specificity were determined using an internal held-out test set. The trained models were further evaluated using an external test set. In the held-out test sets, the T1w and T1w+T2FLAIR models demonstrated AUCs of 0.62 (95% CI: 0.60, 0.64) and 0.67 (95% CI: 0.64, 0.70) (p = 0.006); accuracies were 61% (95% CI: 60%, 63%) and 64% (95% CI: 62%, 66%) (p = 0.008); sensitivities were 0.88 and 0.71; and specificities were 0.23 and 0.53, respectively. The trained models showed similar performance in the external test set. Performance of the current model on both test sets exceeded that of the publicly available model. In conclusion, the use of multi-contrast MRI, specifically incorporating T2-FLAIR in addition to T1w images, significantly improved the predictive accuracy of PET-determined amyloid status from MRI scans using a deep learning approach., Comment: 18 pages, 7 figures, 4 tables

The development of high-performance materials for microelectronics, energy storage, and extreme environments depends on our ability to describe and direct property-defining microstructural order. Our present understanding is typically derived from laborious manual analysis of imaging and spectroscopy data, which is difficult to scale, challenging to reproduce, and lacks the ability to reveal latent associations needed for mechanistic models. Here, we demonstrate a multi-modal machine learning (ML) approach to describe order from electron microscopy analysis of the complex oxide La$_{1-x}$Sr$_x$FeO$_3$. We construct a hybrid pipeline based on fully and semi-supervised classification, allowing us to evaluate both the characteristics of each data modality and the value each modality adds to the ensemble. We observe distinct differences in the performance of uni- and multi-modal models, from which we draw general lessons in describing crystal order using computer vision., Comment: 30 pages, 5 figures, 2 tables

Precision medicine leverages patient heterogeneity to estimate individualized treatment regimens, formalized, data-driven approaches designed to match patients with optimal treatments. In the presence of competing events, where multiple causes of failure can occur and one cause precludes others, it is crucial to assess the risk of the specific outcome of interest, such as one type of failure over another. This helps clinicians tailor interventions based on the factors driving that particular cause, leading to more precise treatment strategies. Currently, no precision medicine methods simultaneously account for both survival and competing risk endpoints. To address this gap, we develop a nonparametric individualized treatment regime estimator. Our two-phase method accounts for both overall survival from all events as well as the cumulative incidence of a main event of interest. Additionally, we introduce a multi-utility value function that incorporates both outcomes. We develop random survival and random cumulative incidence forests to construct individual survival and cumulative incidence curves. Simulation studies demonstrated that our proposed method performs well, which we applied to a cohort of peripheral artery disease patients at high risk for limb loss and mortality., Comment: 22 pages, 4 figures

The He II reionization epoch is expected to take place at $z\sim3-5$. In this stage, the helium and metals in the inter-galactic medium (IGM) are further ionized with additional contributions from harder non-stellar sources, and some large-scale gravitationally bound systems approach virialization. The "Probing the He II re-Ionization ERa via Absorbing C IV Historical Yield (HIERACHY)" program utilizes high- and medium-resolution spectra of bright background quasars at $z\approx3.9-5.2$ to investigate Ly$\alpha$, C IV, and other metal absorption lines during this epoch. Additionally, we employ narrow-band imaging to search for Ly$\alpha$ emitters associated with C IV absorbers, alongside multi-wavelength observations to identify and study particularly intriguing cases. In this paper, we present the design of the HIERACHY program, its current status, major scientific goals, and examples of initial data products from completed Magellan/MIKE, MagE spectroscopy, and MDM imaging observations. We also provide a brief outlook on future multi-wavelength observations that may significantly impact the related science., Comment: 25 pages, 10 figures, 2 tables, accepted for publication by ApJ

The $t_{2g}$ band-structure of SrTiO$_3$-based two-dimensional electron gasses (2DEGs), have been found to play a role in features such as the superconducting dome, high-mobility transport, and the magnitude of spin-orbit coupling. This adds to the already very diverse range of phenomena, including magnetism and extreme magnetoresistance, exhibited by this particular material platform. Tuning and/or combining these intriguing attributes could yield significant progress within quantum and spintronics technologies. Doing so demands precise control of the parameters, which requires a better understanding of the factors that affect them. Here we present effects of the $t_{2g}$ band-order inversion, stemming from the growth of spinel-structured $\gamma$-Al$_2$O$_3$ onto perovskite SrTiO$_3$. Electronic transport measurements show that with LaAlO$_3$/SrTiO$_3$ as the reference, the carrier density and electron mobility are enhanced, and the sample displays a reshaping of the superconducting dome. Additionally, unpaired spins are evidenced by increasing Anomalous Hall Effect with decreasing temperature, entering the same temperature range as the superconducting transition, and a Kondo-like upturn in the sheet resistance. Finally, it is argued that the high-mobility $d_{xz/yz}$-band is more likely than the $d_{xy}$-band to host the supercurrent., Comment: 10 pages, 5 figures

Human excrement composting (HEC) is a sustainable strategy for human excrement (HE) management that recycles nutrients and mitigates health risks while reducing reliance on freshwater, fossil fuels, and fertilizers. We present a comprehensive microbial time series analysis of HEC and show that the initial gut-like microbiome of HEC systems transitions to a microbiome similar to soil and traditional compost in fifteen biological replicates tracked weekly for one year., Comment: Main text: 9 pages, 2 figures; Extended data: 10 figures; Supplemental Text: 32 pages, 8 figures, 2 tables

Among the most puzzling early discoveries of JWST are "Little Red Dots" -- compact red sources that host broad Balmer emission lines and, in many cases, exhibit a "V shaped" change in slope in the rest-optical. The physical properties of Little Red Dots currently have order-of-magnitude uncertainties, because models to explain the continuum of these sources differ immensely. Here, we leverage the complete selection of red sources in the RUBIES program, supplemented with public PRISM spectra, to study the origin of this "V shape". By fitting a broken power law with a flexible inflection point, we find that a large fraction (20/44, nearly all spatially unresolved) of extremely red H$\alpha$ emitters at $2

Surfactants are key ingredients in foaming and cleansing products across various industries such as personal and home care, industrial cleaning, and more, with the critical micelle concentration (CMC) being of major interest. Predictive models for CMC of pure surfactants have been developed based on recent ML methods, however, in practice surfactant mixtures are typically used due to to performance, environmental, and cost reasons. This requires accounting for synergistic/antagonistic interactions between surfactants; however, predictive ML models for a wide spectrum of mixtures are missing so far. Herein, we develop a graph neural network (GNN) framework for surfactant mixtures to predict the temperature-dependent CMC. We collect data for 108 surfactant binary mixtures, to which we add data for pure species from our previous work [Brozos et al. (2024), J. Chem. Theory Comput.]. We then develop and train GNNs and evaluate their accuracy across different prediction test scenarios for binary mixtures relevant to practical applications. The final GNN models demonstrate very high predictive performance when interpolating between different mixture compositions and for new binary mixtures with known species. Extrapolation to binary surfactant mixtures where either one or both surfactant species are not seen before, yields accurate results for the majority of surfactant systems. We further find superior accuracy of the GNN over a semi-empirical model based on activity coefficients, which has been widely used to date. We then explore if GNN models trained solely on binary mixture and pure species data can also accurately predict the CMCs of ternary mixtures. Finally, we experimentally measure the CMC of 4 commercial surfactants that contain up to four species and industrial relevant mixtures and find a very good agreement between measured and predicted CMC values.

Deploying deep reinforcement learning (DRL) in real-world inventory management presents challenges, including dynamic environments and uncertain problem parameters, e.g. demand and lead time distributions. These challenges highlight a research gap, suggesting a need for a unifying framework to model and solve sequential decision-making under parameter uncertainty. We address this by exploring an underexplored area of DRL for inventory management: training generally capable agents (GCAs) under zero-shot generalization (ZSG). Here, GCAs are advanced DRL policies designed to handle a broad range of sampled problem instances with diverse inventory challenges. ZSG refers to the ability to successfully apply learned policies to unseen instances with unknown parameters without retraining. We propose a unifying Super-Markov Decision Process formulation and the Train, then Estimate and Decide (TED) framework to train and deploy a GCA tailored to inventory management applications. The TED framework consists of three phases: training a GCA on varied problem instances, continuously estimating problem parameters during deployment, and making decisions based on these estimates. Applied to periodic review inventory problems with lost sales, cyclic demand patterns, and stochastic lead times, our trained agent, the Generally Capable Lost Sales Network (GC-LSN) consistently outperforms well-known traditional policies when problem parameters are known. Moreover, under conditions where demand and/or lead time distributions are initially unknown and must be estimated, we benchmark against online learning methods that provide worst-case performance guarantees. Our GC-LSN policy, paired with the Kaplan-Meier estimator, is demonstrated to complement these methods by providing superior empirical performance.

In this paper, we introduce the first release of a large-scale dataset capturing discourse on $\mathbb{X}$ (a.k.a., Twitter) related to the upcoming 2024 U.S. Presidential Election. Our dataset comprises 22 million publicly available posts on X.com, collected from May 1, 2024, to July 31, 2024, using a custom-built scraper, which we describe in detail. By employing targeted keywords linked to key political figures, events, and emerging issues, we aligned data collection with the election cycle to capture evolving public sentiment and the dynamics of political engagement on social media. This dataset offers researchers a robust foundation to investigate critical questions about the influence of social media in shaping political discourse, the propagation of election-related narratives, and the spread of misinformation. We also present a preliminary analysis that highlights prominent hashtags and keywords within the dataset, offering initial insights into the dominant themes and conversations occurring in the lead-up to the election. Our dataset is available at: url{https://github.com/sinking8/usc-x-24-us-election

Biofouling represents a critical challenge in marine transportation, healthcare, and food manufacturing, among other industries, as it promotes contamination and increases maintenance costs. Zwitterionic polymers, known for their exceptional antifouling properties, offer a promising solution for biofouling deterrence. Despite the rapid development of zwitterionic polymers in recent years, the design rules, especially concerning the choice of cationic moieties to optimize biofouling deterrence, remain elusive. In this study, we leveraged a versatile all-dry synthesis scheme to achieve a selection of 9 zwitterionic polymers, 5 of which are unprecedented for this synthesis paradigm, thus systematically unraveling that molecular design rule. Notably, we developed a synthesis strategy to enable nanoscale compositional gradient along the coating cross-section, which ensures the robustness of the zwitterionic polymer coatings irrespective of the choice of cation-anion combinations. That robustness is enabled by an organosilicon-based layer at the coating-substrate interface, which simultaneously enhances coating adhesion and chemical stability while ensuring high concentration of zwitterionic moieties at the polymer-liquid interface to maximize biofouling deterrence. The antifouling efficacy was assessed using biofilms of Pseudomonas aeruginosa or Bacillus subtilis. All coatings demonstrated antifouling efficacy, with a novel zwitterionic polymer comprising a combination of imidazolium and carboxyl groups achieving the greatest antibiofilm effects, which we attributed to the strong hydration. This study highlights the coating architecture, i.e., one with nanoscale gradient and varying crosslinking densities, as a valid strategy to render zwitterionic polymers robust coatings and the imidazolium-based carboxybetaine as a promising next-generation antibiofouling chemistry.

Understanding the genetic basis of complex traits is a longstanding challenge in the field of genomics. Genome-wide association studies (GWAS) have identified thousands of variant-trait associations, but most of these variants are located in non-coding regions, making the link to biological function elusive. While traditional approaches, such as transcriptome-wide association studies (TWAS), have advanced our understanding by linking genetic variants to gene expression, they often overlook gene-gene interactions. Here, we review current approaches to integrate different molecular data, leveraging machine learning methods to identify gene modules based on co-expression and functional relationships. These integrative approaches, like PhenoPLIER, combine TWAS and drug-induced transcriptional profiles to effectively capture biologically meaningful gene networks. This integration provides a context-specific understanding of disease processes while highlighting both core and peripheral genes. These insights pave the way for novel therapeutic targets and enhance the interpretability of genetic studies in personalized medicine., Comment: 26 pages, 4 figures GitHub: https://github.com/pivlab/annual_review_of_biomedical_data_science HTML: https://pivlab.github.io/annual_review_of_biomedical_data_science/

Extracting information from quantum many-body systems remains a key challenge in quantum technologies due to experimental limitations. In this work, we employ a single spin qubit to probe a strongly interacting system, creating an environment conducive to qubit decoherence. By focusing on the XXZ spin chain, we observe diverse dynamics in the qubit evolution, reflecting different parameters of the chain. This demonstrates that a spin qubit can probe both quantitative properties of the spin chain and qualitative characteristics, such as the bipartite entanglement entropy, phase transitions, and perturbation propagation velocity within the system. This approach reveals the power of small quantum systems to probe the properties of large, strongly correlated quantum systems., Comment: 5 pages, 2 figures, 1 supplemental material

We investigate the design of magnetic ordering in one-dimensional mesoscopic magnetic Ising chains by modulating long-range interactions. These interactions are affected by geometrical modifications to the chain, which adjust the energy hierarchy and the resulting magnetic ground states. Consequently, the magnetic ordering can be tuned between antiferromagnetic and dimer antiferromagnetic phases. These phases are experimentally observed in chains fabricated using both conventional electron-beam lithography and ion implantation techniques, demonstrating the feasibility of controlling magnetic properties at the mesoscale. The ability of attaining these magnetic structures by thermal annealing, underlines the potential of using such systems instead of simulated annealers in tackling combinatorial optimization tasks., Comment: 12 pages, 9 figures

The matching of competences, such as skills, occupations or knowledges, is a key desiderata for candidates to be fit for jobs. Automatic extraction of competences from CVs and Jobs can greatly promote recruiters' productivity in locating relevant candidates for job vacancies. This work presents the first model that jointly extracts and classifies competence from Danish job postings. Different from existing works on skill extraction and skill classification, our model is trained on a large volume of annotated Danish corpora and is capable of extracting a wide range of Danish competences, including skills, occupations and knowledges of different categories. More importantly, as a single BERT-like architecture for joint extraction and classification, our model is lightweight and efficient at inference. On a real-scenario job matching dataset, our model beats the state-of-the-art models in the overall performance of Danish competence extraction and classification, and saves over 50% time at inference.

The integration of Large Language Models (LLMs) into medical applications has sparked widespread interest across the healthcare industry, from drug discovery and development to clinical decision support, assisting telemedicine, medical devices, and healthcare insurance applications. This perspective paper aims to discuss the inner workings of building LLM-powered medical AI applications and introduces a comprehensive framework for their development. We review existing literature and outline the unique challenges of applying LLMs in specialized medical contexts. Additionally, we introduce a three-step framework to organize medical LLM research activities: 1) Modeling: breaking down complex medical workflows into manageable steps for developing medical-specific models; 2) Optimization: optimizing the model performance with crafted prompts and integrating external knowledge and tools, and 3) System engineering: decomposing complex tasks into subtasks and leveraging human expertise for building medical AI applications. Furthermore, we offer a detailed use case playbook that describes various LLM-powered medical AI applications, such as optimizing clinical trial design, enhancing clinical decision support, and advancing medical imaging analysis. Finally, we discuss various challenges and considerations for building medical AI applications with LLMs, such as handling hallucination issues, data ownership and compliance, privacy, intellectual property considerations, compute cost, sustainability issues, and responsible AI requirements.

Many reinforcement learning (RL) algorithms are too costly to use in practice due to the large sizes $S, A$ of the problem's state and action space. To resolve this issue, we study transfer RL with latent low rank structure. We consider the problem of transferring a latent low rank representation when the source and target MDPs have transition kernels with Tucker rank $(S , d, A )$, $(S , S , d), (d, S, A )$, or $(d , d , d )$. In each setting, we introduce the transfer-ability coefficient $\alpha$ that measures the difficulty of representational transfer. Our algorithm learns latent representations in each source MDP and then exploits the linear structure to remove the dependence on $S, A $, or $S A$ in the target MDP regret bound. We complement our positive results with information theoretic lower bounds that show our algorithms (excluding the ($d, d, d$) setting) are minimax-optimal with respect to $\alpha$.

How many permutations are needed so that every infinite-coinfinite set of natural numbers with asymptotic density can be rearranged to no longer have the same density? We prove that the density number $\mathfrak{dd}$, which answers this question, is equal to the least size of a non-meager set of reals, $\mathsf{non} (\mathcal{M})$. The same argument shows that a slight modification of the rearrangement number $\mathfrak{rr}$ of~\cite{BBBHHL20} is equal to $\mathsf{non} (\mathcal{M})$, and similarly for a cardinal invariant related to large-scale topology introduced by Banakh~\cite{Ba23}, thus answering a question of the latter. We then consider variants of $\mathfrak{dd}$ given by restricting the possible densities of the original set and / or of the permuted set, providing lower and upper bounds for these cardinals and proving consistency of strict inequalities. We finally look at cardinals defined in terms of relative density and of asymptotic mean, and relate them to the rearrangement numbers of~\cite{BBBHHL20}.

The Local Ultraviolet to Infrared Treasury (LUVIT) is a Hubble Space Telescope program that combines newly acquired data in the near ultraviolet (NUV), optical, and near infrared (NIR) with archival optical and NIR imaging to produce multiband panchromatic resolved stellar catalogs for 23 pointings in 22 low-mass, star-forming galaxies ranging in distance from the outskirts of the Local Group to ~3.8 Mpc. We describe the survey design, detail the LUVIT broadband filter observations and the archival datasets included in the LUVIT reductions, and summarize the simultaneous multiband data reduction steps. The spatial distributions and color-magnitude diagrams (CMDs) from the resulting stellar catalogs are presented for each target, from the NUV to the NIR. We demonstrate in which regions of the CMDs stars with NUV and optical, optical and NIR, and NUV through NIR detections reside. For each target, we use the results from artificial star tests to measure representative completeness, bias, and total photometric uncertainty as a function of magnitude in each broadband filter. We also assess which LUVIT targets have significant spatial variation in the fraction of stars recovered at a given magnitude. The panchromatic LUVIT stellar catalogs will provide a rich legacy dataset for a host of resolved stellar population studies., Comment: 48 pages, 14 figures, 8 tables, accepted for publication in ApJS

By analyzing the spectral energy distributions (SEDs) of resolved stars in nearby galaxies, we can constrain their stellar properties and line-of-sight dust extinction. From the Scylla survey, we obtain ultraviolet to near-infrared photometry from Wide Field Camera 3 onboard the {\it Hubble Space Telescope} for more than 1.5 million stars in the SMC and LMC. We use the Bayesian Extinction and Stellar Tool (BEAST) to analyze the multi-band SEDs of these sources and characterize their initial masses, ages, metallicities, distances, and line-of-sight extinction properties (e.g.~$A_V$, $R_V$). We apply quality cuts and perform validation simulations to construct a catalog of over 550,000 stars with high-reliability SED fits, which we use to analyze the stellar content and extinction properties of the SMC and LMC. We detect stars with masses as low as 0.6 $M_{\odot}$. Observed stellar age distributions show a jump in stars around 6 Gyrs ago, which is in agreement with other star-formation histories. Extinctions ($A_V$) in both galaxies follow a log-normal distribution. We compare $A_V$ with ancillary gas and dust tracers like $HI$, $H_\alpha$, and far infrared (FIR) dust emission and find positive correlations on a field-by-field basis. We convert observed $A_V$ to predicted dust surface densities using the Draine et. al. (2014) model and find $A_V$-based dust surface densities are a factor of $\sim$2.5 lower than observed FIR-based dust surface densities, a correction factor similar to other studies., Comment: Submitted to ApJ, 31 pages

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

We consider robustness to distribution shifts in the context of diagnostic models in healthcare, where the prediction target $Y$, e.g., the presence of a disease, is causally upstream of the observations $X$, e.g., a biomarker. Distribution shifts may occur, for instance, when the training data is collected in a domain with patients having particular demographic characteristics while the model is deployed on patients from a different demographic group. In the domain of applied ML for health, it is common to predict $Y$ from $X$ without considering further information about the patient. However, beyond the direct influence of the disease $Y$ on biomarker $X$, a predictive model may learn to exploit confounding dependencies (or shortcuts) between $X$ and $Y$ that are unstable under certain distribution shifts. In this work, we highlight a data generating mechanism common to healthcare settings and discuss how recent theoretical results from the causality literature can be applied to build robust predictive models. We theoretically show why ignoring covariates as well as common invariant learning approaches will in general not yield robust predictors in the studied setting, while including certain covariates into the prediction model will. In an extensive simulation study, we showcase the robustness (or lack thereof) of different predictors under various data generating processes. Lastly, we analyze the performance of the different approaches using the PTB-XL dataset, a public dataset of annotated ECG recordings.

We introduce the notion of causally-null-compactifiable space-times which can be canonically converted into a compact timed-metric-spaces using the cosmological time of Andersson-Howard-Galloway and the null distance of Sormani-Vega. We produce a large class of such space-times including future developments of compact initial data sets and regions which exhaust asymptotically flat space-times. We then present various notions of intrinsic distances between these space-times (introducing the timed-Hausdorff distance) and prove some of these notions of distance are definite in the sense that they equal zero iff there is a time-oriented Lorentzian isometry between the space-times. These definite distances enable us to define various notions of convergence of space-times to limit space-times which are not necessarily smooth. Many open questions and conjectures are included throughout., Comment: 135 pages with wide margins and space, color images