Search

The increasing demand for multilingual capabilities in healthcare underscores the need for AI models adept at processing diverse languages, particularly in clinical documentation and decision-making. Arabic, with its complex morphology, syntax, and diglossia, poses unique challenges for natural language processing (NLP) in medical contexts. This case study evaluates Sporo AraSum, a language model tailored for Arabic clinical documentation, against JAIS, the leading Arabic NLP model. Using synthetic datasets and modified PDQI-9 metrics modified ourselves for the purposes of assessing model performances in a different language. The study assessed the models' performance in summarizing patient-physician interactions, focusing on accuracy, comprehensiveness, clinical utility, and linguistic-cultural competence. Results indicate that Sporo AraSum significantly outperforms JAIS in AI-centric quantitative metrics and all qualitative attributes measured in our modified version of the PDQI-9. AraSum's architecture enables precise and culturally sensitive documentation, addressing the linguistic nuances of Arabic while mitigating risks of AI hallucinations. These findings suggest that Sporo AraSum is better suited to meet the demands of Arabic-speaking healthcare environments, offering a transformative solution for multilingual clinical workflows. Future research should incorporate real-world data to further validate these findings and explore broader integration into healthcare systems., Comment: arXiv admin note: text overlap with arXiv:2411.06713

Convection in planets and stars is predicted to occur in the "ultimate regime'' of diffusivity-free, rapidly rotating turbulence, in which flows are characteristically unaffected by viscous and thermal diffusion. Boundary layer diffusion, however, has historically hindered experimental study of this regime. Here, we utilize the boundary-independent oscillatory thermal-inertial mode of rotating convection to realize the diffusivity-free scaling in liquid metal laboratory experiments. This oscillatory style of convection arises in rotating liquid metals (low Prandtl number fluids) and is driven by the temperature gradient in the fluid bulk, thus remaining independent of diffusive boundary dynamics. We triply verify the existence of the diffusivity-free regime via measurements of heat transfer efficiency $Nu$, dimensionless flow velocities $Re$, and internal temperature anomalies $\theta$, all of which are in quantitative agreement with planar asymptotically-reduced models. Achieving the theoretical diffusivity-free scalings in desktop-sized laboratory experiments provides the validation necessary to extrapolate and predict the convective flows in remote geophysical and astrophysical systems.

Multimodal data integration and label scarcity pose significant challenges for machine learning in medical settings. To address these issues, we conduct an in-depth evaluation of the newly proposed Multimodal Variational Mixture-of-Experts (MMVM) VAE on the challenging MIMIC-CXR dataset. Our analysis demonstrates that the MMVM VAE consistently outperforms other multimodal VAEs and fully supervised approaches, highlighting its strong potential for real-world medical applications., Comment: Findings paper presented at Machine Learning for Health (ML4H) symposium 2024, December 15-16, 2024, Vancouver, Canada, 13 pages. arXiv admin note: text overlap with arXiv:2403.05300

General movements (GMs) are spontaneous, coordinated body movements in infants that offer valuable insights into the developing nervous system. Assessed through the Prechtl GM Assessment (GMA), GMs are reliable predictors for neurodevelopmental disorders. However, GMA requires specifically trained clinicians, who are limited in number. To scale up newborn screening, there is a need for an algorithm that can automatically classify GMs from infant video recordings. This data poses challenges, including variability in recording length, device type, and setting, with each video coarsely annotated for overall movement quality. In this work, we introduce a tool for extracting features from these recordings and explore various machine learning techniques for automated GM classification., Comment: Findings paper presented at Machine Learning for Health (ML4H) symposium 2024, December 15-16, 2024, Vancouver, Canada, 6 pages

The orbital period of a cataclysmic variable stands as a crucial parameter for investigating the structure and physics of these binary systems, as well as understanding their evolution. We use photometric Gaia data for dwarf novae (DNe) in the quiescent state which are available for a number of years to determine new orbital periods and improve/modify previously suggested values. Two approaches are implemented for selecting high-inclination targets, either eclipsing or with ellipsoidal variations. We determine new orbital periods for 75 DNe and improve ephemerides for 27 more (three of which change significantly), contributing 9.4% of the known DNe periods of 0.05-2.0 days, and doubling the number of known periods exceeding 0.44 days. Their phase-folded light curves are presented and arranged by orbital period, illustrating the transition from short-period systems, dominated by radiation from the accretion disc and the hot spot, to longer-period DNe, where the Roche-lobe-filling secondary star is the primary visual flux source. This transition -- which occurs around the well-known period gap (around 2-3 hours) -- is expected, as DNe with larger orbital periods typically harbour more massive donors, which contribute to the visible flux. However, this transition is not abrupt. Within the same range of periods, we observe systems dominated by ellipsoidal variations, where the companion star is clearly visible, as well as others dominated by the disc and hot spot. The presence of some DNe with ellipsoidal variations near the lower edge of the period gap is striking, as the companions in these systems are expected to be cool low-mass M-dwarfs not visible in the light curve. This could indicate that we are observing systems where the donor star was originally much more massive and underwent significant nuclear evolution before mass-transfer began, as has been suggested previously for QZ Ser., Comment: Accepted for publication in A&A, 15 pages, 3 figures

Electron beam accelerators are essential in many scientific and technological fields. Their operation relies heavily on the stability and precision of the electron beam. Traditional diagnostic techniques encounter difficulties in addressing the complex and dynamic nature of electron beams. Particularly in the context of free-electron lasers (FELs), it is fundamentally impossible to measure the lasing-on and lasingoff electron power profiles for a single electron bunch. This is a crucial hurdle in the exact reconstruction of the photon pulse profile. To overcome this hurdle, we developed a machine learning model that predicts the temporal power profile of the electron bunch in the lasing-off regime using machine parameters that can be obtained when lasing is on. The model was statistically validated and showed superior predictions compared to the state-of-the-art batch calibrations. The work we present here is a critical element for a virtual pulse reconstruction diagnostic (VPRD) tool designed to reconstruct the power profile of individual photon pulses without requiring repeated measurements in the lasing-off regime. This promises to significantly enhance the diagnostic capabilities in FELs at large., Comment: 10 pages, 4 figures, Machine Learning and the Physical Sciences Workshop, NeurIPS 2024 https://neurips.cc/virtual/2024/100009

This study compares Sporo Health's AI Scribe, a proprietary model fine-tuned for medical scribing, with various LLMs (GPT-4o, GPT-3.5, Gemma-9B, and Llama-3.2-3B) in clinical documentation. We analyzed de-identified patient transcripts from partner clinics, using clinician-provided SOAP notes as the ground truth. Each model generated SOAP summaries using zero-shot prompting, with performance assessed via recall, precision, and F1 scores. Sporo outperformed all models, achieving the highest recall (73.3%), precision (78.6%), and F1 score (75.3%) with the lowest performance variance. Statistically significant differences (p < 0.05) were found between Sporo and the other models, with post-hoc tests showing significant improvements over GPT-3.5, Gemma-9B, and Llama 3.2-3B. While Sporo outperformed GPT-4o by up to 10%, the difference was not statistically significant (p = 0.25). Clinical user satisfaction, measured with a modified PDQI-9 inventory, favored Sporo. Evaluations indicated Sporo's outputs were more accurate and relevant. This highlights the potential of Sporo's multi-agentic architecture to improve clinical workflows., Comment: arXiv admin note: substantial text overlap with arXiv:2410.15528

Supervised learning has become a cornerstone of modern machine learning, yet a comprehensive theory explaining its effectiveness remains elusive. Empirical phenomena, such as neural analogy-making and the linear representation hypothesis, suggest that supervised models can learn interpretable factors of variation in a linear fashion. Recent advances in self-supervised learning, particularly nonlinear Independent Component Analysis, have shown that these methods can recover latent structures by inverting the data generating process. We extend these identifiability results to parametric instance discrimination, then show how insights transfer to the ubiquitous setting of supervised learning with cross-entropy minimization. We prove that even in standard classification tasks, models learn representations of ground-truth factors of variation up to a linear transformation. We corroborate our theoretical contribution with a series of empirical studies. First, using simulated data matching our theoretical assumptions, we demonstrate successful disentanglement of latent factors. Second, we show that on DisLib, a widely-used disentanglement benchmark, simple classification tasks recover latent structures up to linear transformations. Finally, we reveal that models trained on ImageNet encode representations that permit linear decoding of proxy factors of variation. Together, our theoretical findings and experiments offer a compelling explanation for recent observations of linear representations, such as superposition in neural networks. This work takes a significant step toward a cohesive theory that accounts for the unreasonable effectiveness of supervised deep learning.

Concept-based machine learning methods have increasingly gained importance due to the growing interest in making neural networks interpretable. However, concept annotations are generally challenging to obtain, making it crucial to leverage all their prior knowledge. By creating concept-enriched models that incorporate concept information into existing architectures, we exploit their interpretable capabilities to the fullest extent. In particular, we propose Concept-Guided Conditional Diffusion, which can generate visual representations of concepts, and Concept-Guided Prototype Networks, which can create a concept prototype dataset and leverage it to perform interpretable concept prediction. These results open up new lines of research by exploiting pre-existing information in the quest for rendering machine learning more human-understandable., Comment: Interpretable AI: Past, Present and Future Workshop at NeurIPS 2024

Finding clusters of data points with similar characteristics and generating new cluster-specific samples can significantly enhance our understanding of complex data distributions. While clustering has been widely explored using Variational Autoencoders, these models often lack generation quality in real-world datasets. This paper addresses this gap by introducing TreeDiffusion, a deep generative model that conditions Diffusion Models on hierarchical clusters to obtain high-quality, cluster-specific generations. The proposed pipeline consists of two steps: a VAE-based clustering model that learns the hierarchical structure of the data, and a conditional diffusion model that generates realistic images for each cluster. We propose this two-stage process to ensure that the generated samples remain representative of their respective clusters and enhance image fidelity to the level of diffusion models. A key strength of our method is its ability to create images for each cluster, providing better visualization of the learned representations by the clustering model, as demonstrated through qualitative results. This method effectively addresses the generative limitations of VAE-based approaches while preserving their clustering performance. Empirically, we demonstrate that conditioning diffusion models on hierarchical clusters significantly enhances generative performance, thereby advancing the state of generative clustering models., Comment: 25 pages, submitted to ICLR 2025

This work explores the feasibility of employing ultrasound (US) US technology in a wrist-worn IoT device for low-power, high-fidelity heart-rate (HR) extraction. US offers deep tissue penetration and can monitor pulsatile arterial blood flow in large vessels and the surrounding tissue, potentially improving robustness and accuracy compared to PPG. We present an IoT wearable system prototype utilizing a commercial microcontroller MCU employing the onboard ADC to capture high frequency US signals and an innovative low-power US pulser. An envelope filter lowers the bandwidth of the US signal by a factor of >5x, reducing the system's acquisition requirements without compromising accuracy (correlation coefficient between HR extracted from enveloped and raw signals, r(92)=0.99, p<0.001). The full signal processing pipeline is ported to fixed point arithmetic for increased energy efficiency and runs entirely onboard. The system has an average power consumption of 5.8mW, competitive with PPG based systems, and the HR extraction algorithm requires only 68kB of RAM and 71ms of processing time on an ARM Cortex-M4 MCU. The system is estimated to run continuously for more than 7 days on a smartwatch battery. To accurately evaluate the proposed circuit and algorithm and identify the anatomical location on the wrist with the highest accuracy for HR extraction, we collected a dataset from 10 healthy adults at three different wrist positions. The dataset comprises roughly 5 hours of HR data with an average of 80.6+-16.3 bpm. During recording, we synchronized the established ECG gold standard with our US-based method. The comparisons yields a Pearson correlation coefficient of r(92)=0.99, p<0.001 and a mean error of 0.69+-1.99 bpm in the lateral wrist position near the radial artery. The dataset and code have been open-sourced at https://github.com/mgiordy/Ultrasound-Heart-Rate

AI-powered medical scribes have emerged as a promising solution to alleviate the documentation burden in healthcare. Ambient AI scribes provide real-time transcription and automated data entry into Electronic Health Records (EHRs), with the potential to improve efficiency, reduce costs, and enhance scalability. Despite early success, the accuracy of AI scribes remains critical, as errors can lead to significant clinical consequences. Additionally, AI scribes face challenges in handling the complexity and variability of medical language and ensuring the privacy of sensitive patient data. This case study aims to evaluate Sporo Health's AI scribe, a multi-agent system leveraging fine-tuned medical LLMs, by comparing its performance with OpenAI's GPT-4o Mini on multiple performance metrics. Using a dataset of de-identified patient conversation transcripts, AI-generated summaries were compared to clinician-generated notes (the ground truth) based on clinical content recall, precision, and F1 scores. Evaluations were further supplemented by clinician satisfaction assessments using a modified Physician Documentation Quality Instrument revision 9 (PDQI-9), rated by both a medical student and a physician. The results show that Sporo AI consistently outperformed GPT-4o Mini, achieving higher recall, precision, and overall F1 scores. Moreover, the AI generated summaries provided by Sporo were rated more favorably in terms of accuracy, comprehensiveness, and relevance, with fewer hallucinations. These findings demonstrate that Sporo AI Scribe is an effective and reliable tool for clinical documentation, enhancing clinician workflows while maintaining high standards of privacy and security.

We have computed the even-N moments N =< 20 of the gluon-gluon splitting function P_{gg} at the fourth order of perturbative QCD via the renormalization of off-shell operator matrix elements. Our results, derived analytically for a general compact simple gauge group, agree with all results obtained for this function so far, in particular with the lowest five moments obtained via structure functions in deep-inelastic scattering. Using our new moments and all available endpoint constraints, we construct improved approximations for the four-loop P_{gg}(x) that should be sufficient for a wide range of collider-physics applications. The N^3LO contributions to the scale derivative of the gluon distribution, resulting from these and the corresponding quark-to-gluon splitting functions, amount to 1% or less at x >~ 10^{-4} at a standard reference scale with alpha_s = 0.2., Comment: 15 pages, LaTeX, 3 figures, 2 ancillary files (FORM file with results and FORTRAN subroutine). arXiv admin note: text overlap with arXiv:2404.09701

The structure of many real-world datasets is intrinsically hierarchical, making the modeling of such hierarchies a critical objective in both unsupervised and supervised machine learning. Recently, novel approaches for hierarchical clustering with deep architectures have been proposed. In this work, we take a critical perspective on this line of research and demonstrate that many approaches exhibit major limitations when applied to realistic datasets, partly due to their high computational complexity. In particular, we show that a lightweight procedure implemented on top of pre-trained non-hierarchical clustering models outperforms models designed specifically for hierarchical clustering. Our proposed approach is computationally efficient and applicable to any pre-trained clustering model that outputs logits, without requiring any fine-tuning. To highlight the generality of our findings, we illustrate how our method can also be applied in a supervised setup, recovering meaningful hierarchies from a pre-trained ImageNet classifier.

Rotating turbulent convection occurs ubiquitously in natural convective systems encompassing planetary cores, oceans, and atmospheres, as well as in many industrial applications. While the global heat and mass transfer of water-like rotating Rayleigh-B\'enard convection is well-documented, the characteristics of rotating convection in liquid metals remain less well understood. In this study, we characterize rotating Rayleigh-B\'enard convection in liquid gallium (Prandtl number $Pr \approx 0.027$) within a slender cylinder (diameter-to-height aspect ratio $\Gamma = D/H = 1/2$) using novel thermovelocimetric diagnostic techniques that integrate simultaneous multi-point thermometry and ultrasonic Doppler velocity measurements. This approach experimentally reveals the formation of a stable azimuthal wavenumber $m = 2$ global-scale vortical structure at low supercriticality. We propose that enhanced wall modes facilitated by the slender cylinder geometry interact with the bulk flow to create these large-scale axialized vortices. Our findings extend results from the previous $Pr \sim 1$ studies across various cylindrical aspect ratios. In particular, we find evidence of a different scaling for wall mode precession frequency that possibly exists in liquid metal, offering new insights into the coupling effects in low-$Pr$ rotating convective turbulence., Comment: 10 main text + 4 appendix figures

IJCLab has been leading the development and deployment of low-$\beta$ Superconducting Radio Frequency (SRF) cavities for proton and heavy ion accelerators. We are launching an electron accelerator project for sustainable Energy Recovery Linac (iSAS/PERLE) with state-of-the-art SRF cavities at 800~MHz. Our proposal includes advanced heat treatment of such cavities to reach an excellent quality factor of $3\times 10^{10}$ at $22$~MV/m. In this paper, we overview the status of this activity., Comment: Contribution to LCWS2024 proceedings

We examine the fluid flow forced by precession of a rotating cylindrical container using numerical simulations and experimental flow measurements with ultrasonic Doppler velocimetry (UDV). The analysis is based on the decomposition of the flow field into contributions with distinct azimuthal symmetry or analytically known inertial modes and the corresponding calculation of their amplitudes. We show that the predominant fraction of the kinetic energy of the precession-driven fluid flow is contained only within a few large-scale modes. The most striking observation shown by simulations and experiments is the transition from a flow dominated by large-scale structures to a more turbulent behaviour with the small-scale fluctuations becoming increasingly important. At a fixed rotation frequency (parametrized by the Reynolds number, ${\rm{Re}}$) this transition occurs when a critical precession ratio is exceeded and consists of a two-stage collapse of the directly driven flow going along with a massive modification of the azimuthal circulation (the zonal flow) and the appearance of an axisymmetric double-roll mode limited to a narrow range of precession ratios. A similar behaviour is found in experiments which make it possible to follow the transition up to Reynolds numbers of ${\rm{Re}}\approx 2\times 10^6$. We find that the critical precession ratio decreases with rotation, initially showing a particular scaling $\propto {\rm{Re}}^{-\frac{1}{5}}$ but developing an asymptotic behaviour for ${\rm{Re}}\gtrsim 10^5$ which might be explained by the onset of turbulence in boundary layers., Comment: 2024, 29 pages, 16 figures, accepted for publication in Journal of Fluid Mechanics

We present constraints on the $f(R)$ gravity model using a sample of 1,005 galaxy clusters in the redshift range $0.25 - 1.78$ that have been selected through the thermal Sunyaev-Zel'dovich effect (tSZE) from South Pole Telescope (SPT) data and subjected to optical and near-infrared confirmation with the Multi-component Matched Filter (MCMF) algorithm. We employ weak gravitational lensing mass calibration from the Dark Energy Survey (DES) Year 3 data for 688 clusters at $z < 0.95$ and from the Hubble Space Telescope (HST) for 39 clusters with $0.6 < z < 1.7$. Our cluster sample is a powerful probe of $f(R)$ gravity, because this model predicts a scale-dependent enhancement in the growth of structure, which impacts the halo mass function (HMF) at cluster mass scales. To account for these modified gravity effects on the HMF, our analysis employs a semi-analytical approach calibrated with numerical simulations. Combining calibrated cluster counts with primary cosmic microwave background (CMB) temperature and polarization anisotropy measurements from the Planck2018 release, we derive robust constraints on the $f(R)$ parameter $f_{R0}$. Our results, $\log_{10} |f_{R0}| < -5.32$ at the 95 % credible level, are the tightest current constraints on $f(R)$ gravity from cosmological scales. This upper limit rules out $f(R)$-like deviations from general relativity that result in more than a $\sim$20 % enhancement of the cluster population on mass scales $M_\mathrm{200c}>3\times10^{14}M_\odot$., Comment: 21 pages, 6 figures, submitted to Phys. Rev. D

Antibodies offer great potential for the treatment of various diseases. However, the discovery of therapeutic antibodies through traditional wet lab methods is expensive and time-consuming. The use of generative models in designing antibodies therefore holds great promise, as it can reduce the time and resources required. Recently, the class of diffusion models has gained considerable traction for their ability to synthesize diverse and high-quality samples. In their basic form, however, they lack mechanisms to optimize for specific properties, such as binding affinity to an antigen. In contrast, the class of offline Reinforcement Learning (RL) methods has demonstrated strong performance in navigating large search spaces, including scenarios where frequent real-world interaction, such as interaction with a wet lab, is impractical. Our novel method, BetterBodies, which combines Variational Autoencoders (VAEs) with RL guided latent diffusion, is able to generate novel sets of antibody CDRH3 sequences from different data distributions. Using the Absolut! simulator, we demonstrate the improved affinity of our novel sequences to the SARS-CoV spike receptor-binding domain. Furthermore, we reflect biophysical properties in the VAE latent space using a contrastive loss and add a novel Q-function based filtering to enhance the affinity of generated sequences. In conclusion, methods such as ours have the potential to have great implications for real-world biological sequence design, where the generation of novel high-affinity binders is a cost-intensive endeavor.

We compute the differential cross section for direct quarkonium production in high-energy electron-nucleus collisions at small $x$. Our computation is performed within the nonrelativistic QCD factorization formalism that separates the calculation into short distance coefficients and long distance matrix elements that depend on the color and spin of the state. We obtain the short distance coefficients of the production of the heavy quark pair within the framework of the Color Glass Condensate effective field theory, which resums coherent multiple interactions of the heavy quark pair with the nucleus to all orders. Our results are expressed as the convolution of perturbatively calculable perturbative functions with multi-point light-like Wilson line correlators. In the correlation limit, we establish the correspondence between our CGC formulation with calculations employing the transverse momentum dependent (TMD) framework. We extend this correspondence by resumming kinematic power corrections within the improved TMD framework, which interpolates between the TMD formalism and $k_\perp$ factorization formalism. We present a detailed numerical analysis, focusing on $J/\psi$ production in the kinematics accessible at the future Electron-Ion Collider, highlighting the importance of genuine higher-order saturation contributions when the electron collides with a large nucleus., Comment: 31 pages, 8 figures

The physics case for quarkonium-production studies accessible at the US Electron Ion Collider is described., Comment: Latex, 84 pages. Review prepared for Progress in Particle and Nuclear Physics

The industrial landscape is undergoing a significant transformation, moving away from traditional wired fieldbus networks to cutting-edge 5G mobile networks. This transition, extending from local applications to company-wide use and spanning multiple factories, is driven by the promise of low-latency communication and seamless connectivity for various devices in industrial settings. However, besides these tremendous benefits, the integration of 5G as the communication infrastructure in industrial networks introduces a new set of risks and threats to the security of industrial systems. The inherent complexity of 5G systems poses unique challenges for ensuring a secure integration, surpassing those encountered with any technology previously utilized in industrial networks. Most importantly, the distinct characteristics of industrial networks, such as real-time operation, required safety guarantees, and high availability requirements, further complicate this task. As the industrial transition from wired to wireless networks is a relatively new concept, a lack of guidance and recommendations on securely integrating 5G renders many industrial systems vulnerable and exposed to threats associated with 5G. To address this situation, in this paper, we summarize the state-of-the-art and derive a set of recommendations for the secure integration of 5G into industrial networks based on a thorough analysis of the research landscape. Furthermore, we identify opportunities to utilize 5G to further enhance security and indicate remaining challenges, potentially identifying future academic potential, Comment: 15 pages, 5 figures

Reinforcement learning (RL) has seen significant success across various domains, but its adoption is often limited by the black-box nature of neural network policies, making them difficult to interpret. In contrast, symbolic policies allow representing decision-making strategies in a compact and interpretable way. However, learning symbolic policies directly within on-policy methods remains challenging. In this paper, we introduce SYMPOL, a novel method for SYMbolic tree-based on-POLicy RL. SYMPOL employs a tree-based model integrated with a policy gradient method, enabling the agent to learn and adapt its actions while maintaining a high level of interpretability. We evaluate SYMPOL on a set of benchmark RL tasks, demonstrating its superiority over alternative tree-based RL approaches in terms of performance and interpretability. To the best of our knowledge, this is the first method, that allows a gradient-based end-to-end learning of interpretable, axis-aligned decision trees within existing on-policy RL algorithms. Therefore, SYMPOL can become the foundation for a new class of interpretable RL based on decision trees. Our implementation is available under: https://github.com/s-marton/SYMPOL

Metal-assisted chemical etching of silicon is a promising method for fabricating nanostructures with a high aspect ratio. To define a pattern for the catalyst, lift-off processes are commonly used. The lift-off step however is often a process bottle neck due to low yield, especially for smaller structures. To bypass the lift-off process, other methods such as electroplating can be utilized. In this paper, we suggest an electroplated bi-layer catalyst for vapour phase metal-assisted chemical etching as an alternative to the commonly utilised lift-off process. Samples were successfully etched in vapour, and resulting structures had feature sizes down to 10 nm.

In applied time-to-event analysis, a flexible parametric approach is to model the hazard rate as a piecewise constant function of time. However, the change points and values of the piecewise constant hazard are usually unknown and need to be estimated. In this paper, we develop a fully data-driven procedure for piecewise constant hazard estimation. We work in a general counting process framework which nests a wide range of popular models in time-to-event analysis including Cox's proportional hazards model with potentially high-dimensional covariates, competing risks models as well as more general multi-state models. To construct our estimator, we set up a regression model for the increments of the Breslow estimator and then use fused lasso techniques to approximate the piecewise constant signal in this regression model. In the theoretical part of the paper, we derive the convergence rate of our estimator as well as some results on how well the change points of the piecewise constant hazard are approximated by our method. We complement the theory by both simulations and a real data example, illustrating that our results apply in rather general event histories such as multi-state models.

Machines need data and metadata to be machine-actionable and FAIR (findable, accessible, interoperable, reusable) to manage increasing data volumes. Knowledge graphs and ontologies are key to this, but their use is hampered by high access barriers due to required prior knowledge in semantics and data modelling. The Rosetta Statement approach proposes modeling English natural language statements instead of a mind-independent reality. We propose a metamodel for creating semantic schema patterns for simple statement types. The approach supports versioning of statements and provides a detailed editing history. Each Rosetta Statement pattern has a dynamic label for displaying statements as natural language sentences. Implemented in the Open Research Knowledge Graph (ORKG) as a use case, this approach allows domain experts to define data schema patterns without needing semantic knowledge. Future plans include combining Rosetta Statements with semantic units to organize ORKG into meaningful subgraphs, improving usability. A search interface for querying statements without needing SPARQL or Cypher knowledge is also planned, along with tools for data entry and display using Large Language Models and NLP. The Rosetta Statement metamodel supports a two-step knowledge graph construction procedure. Domain experts can model semantic content without support from ontology engineers, lowering entry barriers and increasing cognitive interoperability. The second level involves developing semantic graph patterns for reasoning, requiring collaboration with ontology engineers.

Sapphire is a technologically highly relevant material, but it poses many challenges to performing epitaxial thin-film deposition. We have identified and applied the conditions for adsorption-controlled homoepitaxial growth of c-plane sapphire. The films thus grown are atomically smooth, have a controlled termination, and are of outstanding crystallinity. Their chemical purity exceeds that of the substrates. The films exhibit exceptional optical properties such as a single-crystal-like bandgap and a low density of F+ centers.

Despite the proliferation of large vision-language foundation models, estimation of the learning and forgetting outcomes following fine-tuning of these models remains largely unexplored. Inspired by work highlighting the significance of the modality gap in contrastive dual-encoders, we propose the Inter-Intra Modal Measure (IIMM). Combining terms quantifying the similarity between image embeddings and the similarity between incorrect image and label embedding pairs, the IIMM functions as a strong predictor of performance changes with fine-tuning. Our extensive empirical analysis across four state-of-the-art vision-language models (CLIP, SigLIP, CoCa, EVA-02-CLIP) and five fine-tuning techniques (full fine-tuning, BitFit, attention-weight tuning, LoRA, CLIP-Adapter) demonstrates a strong, statistically significant linear relationship: fine-tuning on tasks with higher IIMM scores produces greater in-domain performance gains but also induces more severe out-of-domain performance degradation, with some parameter-efficient fine-tuning (PEFT) methods showing extreme forgetting. We compare our measure against transfer scores from state-of-the-art model selection methods and show that the IIMM is significantly more predictive of accuracy gains. With only a single forward pass of the target data, practitioners can leverage this key insight to heuristically evaluate the degree to which a model can be expected to improve following fine-tuning. Given additional knowledge about the model's performance on a few diverse tasks, this heuristic further evolves into a strong predictor of expected performance changes when training for new tasks.

Knowledge graphs and ontologies are becoming increasingly vital as they align with the FAIR Guiding Principles (Findable, Accessible, Interoperable, Reusable). We address eleven challenges that may impede the full realization of the potential of FAIR knowledge graphs, as conventional solutions are perceived to be overly complex and lacking in cognitive interoperability. We extend the concept of "semantic units" as a conceptual solution by adding further subcategories. Semantic units structure a knowledge graph into identifiable and semantically meaningful subgraphs, with each subgraph being represented by a resource that instantiates a semantic unit class. We introduce some-instance, most-instances, every-instance, and all-instances resources as new types of representational entities in addition to named-individual, class, and property resources. We combine these new resource types with the concept of semantic units and introduce new subcategories of statement units and semantically meaningful collections of statement units (i.e., compound units) that provide solutions to the eleven challenges. These include, for instance, schemes for modelling assertional, contingent, prototypical, and universal statements, including class axioms, as well as absence statements, negations, and cardinality restrictions. The schemes are alternatives to existing OWL-based modelling schemes, and we provide corresponding representations for them that do not involve blank nodes. With question units we also introduce a way of representing questions in a knowledge graph that can be made readily executable as graph queries. We also provide schemes for directive statements, directive conditional statements, and logical arguments. We argue that semantic units provide a framework that increases the overall expressivity and cognitive interoperability of knowledge graphs compared to conventional OWL-based solutions., Comment: arXiv admin note: text overlap with arXiv:2301.01227

Explaining the formation pathways of amides on ice-grain mantels is crucial to understanding the prebiotic chemistry in an interstellar medium. In this computational study, we explore different radical-neutral formation pathways for some of the observed amides (formamide, acetamide, urea, and N-methylformamide) via intermediate carbamoyl (NH2CO) radical precursors and their isomers. We assess the relative energy of four NH2CO isomers in the gas phase and evaluate their binding energy on small water clusters to discern the affinity that the isomers present to an ice model. We consider three possible reaction pathways for the formation of the carbamoyl radicals, namely, the OH + HCN, CN + H2O, and NH2 + CO reaction channels. We computed the binding energy distribution for the HCN and CH3CN precursors on an ice model consisting of a set of clusters of 22 water molecules each to serve as a starting point for the reactivity study on the ice surface. The computations revealed that the lowest barrier to the formation of an NH2CO isomer corresponds to the NH2 + CO reaction (12.6 kJ/mol). The OH + HCN reaction pathway results in the exothermic formation of the N-radical form of carbamoyl HN(C=O)H with a reaction barrier of 26.7 kJ/mol. We found that the CN + H2O reaction displays a high energy barrier of 70.6 kJ/mol. Finally, we also probed the direct formation of the acetamide radical precursor via the OH + CH3CN reaction and found that the most probable outcome on interstellar ices is the H-abstraction reaction to yield CH2CN and H2O. Based on these results, we believe that including alternative reaction pathways, leading to the formation of amides via the N-radical form of carbamoyl, would provide an improvement in the prediction of the amide abundances in astrochemical models, especially regarding the chemistry of star-forming regions., Comment: Journal is ACS Earth Space Chem, 2024. ASAP (As Soon As Publishable)

This paper introduces Diffuse-TreeVAE, a deep generative model that integrates hierarchical clustering into the framework of Denoising Diffusion Probabilistic Models (DDPMs). The proposed approach generates new images by sampling from a root embedding of a learned latent tree VAE-based structure, it then propagates through hierarchical paths, and utilizes a second-stage DDPM to refine and generate distinct, high-quality images for each data cluster. The result is a model that not only improves image clarity but also ensures that the generated samples are representative of their respective clusters, addressing the limitations of previous VAE-based methods and advancing the state of clustering-based generative modeling., Comment: 8 pages, 7 figures, Structured Probabilistic Inference & Generative Modeling workshop of ICML 2024

Chemical processes taking place on ice-grain mantles are pivotal to the complex chemistry of interstellar environments. In this study, we conducted a comprehensive analysis of the catalytic effects of an amorphous solid water (ASW) surface on the reaction between ammonia (NH$_3$) and formaldehyde (H$_2$CO) to form aminomethanol (NH$_2$CH$_2$OH) using density functional theory. We identified potential catalytic sites based on the binding energy distribution of NH$_3$ and H$_2$CO reactants, on a set-of-clusters surface model composed of 22 water molecules and found a total of 14 reaction paths. Our results indicate that the catalytic sites can be categorized into four groups, depending on the interactions of the carbonyl oxygen and the amino group with the ice surface in the reactant complex. A detailed analysis of the reaction mechanism using Intrinsic Reaction Coordinate and reaction force analysis revealed three distinct chemical events for this reaction: formation of the C--N bond, breaking of the N--H bond, and formation of the O--H hydroxyl bond. Depending on the type of catalytic site, these events can occur within a single, concerted, albeit asynchronous, step, or can be isolated in a step-wise mechanism, with the lowest overall transition state energy observed at 1.3 kcal mol$^{-1}$. A key requirement for the low-energy mechanism is the presence of a pair of dangling OH bonds on the surface, found at 5\% of the potential catalytic sites on an ASW porous surface.

We propose a novel method, scTree, for single-cell Tree Variational Autoencoders, extending a hierarchical clustering approach to single-cell RNA sequencing data. scTree corrects for batch effects while simultaneously learning a tree-structured data representation. This VAE-based method allows for a more in-depth understanding of complex cellular landscapes independently of the biasing effects of batches. We show empirically on seven datasets that scTree discovers the underlying clusters of the data and the hierarchical relations between them, as well as outperforms established baseline methods across these datasets. Additionally, we analyze the learned hierarchy to understand its biological relevance, thus underpinning the importance of integrating batch correction directly into the clustering procedure.

Concept Bottleneck Models (CBMs) have emerged as a promising interpretable method whose final prediction is based on intermediate, human-understandable concepts rather than the raw input. Through time-consuming manual interventions, a user can correct wrongly predicted concept values to enhance the model's downstream performance. We propose Stochastic Concept Bottleneck Models (SCBMs), a novel approach that models concept dependencies. In SCBMs, a single-concept intervention affects all correlated concepts, thereby improving intervention effectiveness. Unlike previous approaches that model the concept relations via an autoregressive structure, we introduce an explicit, distributional parameterization that allows SCBMs to retain the CBMs' efficient training and inference procedure. Additionally, we leverage the parameterization to derive an effective intervention strategy based on the confidence region. We show empirically on synthetic tabular and natural image datasets that our approach improves intervention effectiveness significantly. Notably, we showcase the versatility and usability of SCBMs by examining a setting with CLIP-inferred concepts, alleviating the need for manual concept annotations., Comment: Published at 38th Conference on Neural Information Processing Systems (NeurIPS 2024)

Effective summarization of unstructured patient data in electronic health records (EHRs) is crucial for accurate diagnosis and efficient patient care, yet clinicians often struggle with information overload and time constraints. This review dives into recent literature and case studies on both the significant impacts and outstanding issues of patient chart review on communications, diagnostics, and management. It also discusses recent efforts to integrate artificial intelligence (AI) into clinical summarization tasks, and its transformative impact on the clinician's potential, including but not limited to reductions of administrative burden and improved patient-centered care.

This white paper offers a comprehensive overview of the protocols utilized in the field of intelligent transportation systems (ITS). The paper presents a comprehensive overview of protocols from all layers of the Open System Interconnection (OSI) model. The protocols encompass a range of communication technologies, including ad-hoc, mobile broadband, and fiber-optic cable. All of these protocols are utilized in the field of intelligent transportation systems (ITS) and must exchange information at some point. The current development leads to a much tighter connection between ecosystems that were previously isolated or not connected. This situation necessitates a comprehensive understanding of the protocols at all layers to ensure the establishment of safe and secure applications within the field of intelligent transportation systems (ITS)., Comment: 39 pages, 1 figure

We give a self-contained introduction to isolated points on curves and their counterpoint, parameterized points, that situates these concepts within the study of the arithmetic of curves. In particular, we show how natural geometric constructions of infinitely many degree $d$ points on curves motivate the definitions of $\mathbb{P}^1$- and AV-parameterized points and explain how a result of Faltings implies that there are only finitely many isolated points on any curve. We use parameterized points to deduce properties of the density degree set and review how the minimum density degree relates to the gonality. The paper includes several examples that illustrate the possible behaviors of degree $d$ points., Comment: 26 pages

A double cover $Y$ of $\mathbb{P}^1 \times \mathbb{P}^2$ ramified over a general $(2,2)$-divisor will have the structure of a geometrically standard conic bundle ramified over a smooth plane quartic $\Delta \subset \mathbb{P}^2$ via the second projection. These threefolds are rational over algebraically closed fields, but over nonclosed fields, including over $\mathbb{R}$, their rationality is an open problem. In this paper, we characterize rationality over $\mathbb{R}$ when $\Delta(\mathbb{R})$ has at least two connected components (extending work of M. Ji and the second author) and over local fields when all odd degree fibers of the first projection have nonsquare discriminant. We obtain these applications by proving general results comparing the conic bundle structure on $Y$ with the conic bundle structure on a well-chosen intersection of two quadrics. The difference between these two conic bundles is encoded by a constant Brauer class, and we prove that this class measures a certain failure of Galois descent for the codimension 2 Chow group of $Y$., Comment: 18 pages

Hybrid atom-photon gates play an important role for the realization of a quantum interface capable of mapping atomic states to photons for communication across quantum networks. Here, we propose a feasible theoretical scheme for implementing a hybrid atom-photon controlled-Z gate between an atom and a microwave photon in a superconducting coplanar waveguide resonator based on the Gaussian soft control technique. The gate protocol employs a classical auxiliary field that induces an atomic transition between one state of the atomic qubit and Rydberg states for obtaining strong coupling of the atom and microwave resonator. By tailoring the amplitude of this field with Gaussian temporal modulation, the gate performances are improved in various aspects. Numerical simulations demonstrate that the controlled-Z gate based on Gaussian soft control is resilient to the variation of the atom-photon coupling strength, deviation in the gate time, and less sensitive to the Rydberg level shifts caused by stray electric fields.

Nowadays, cities provide much more than shopping opportunities or working spaces. Individual locations such as parks and squares are used as meeting points and local recreation areas by many people. To ensure that they remain attractive in the future, the design of such squares must be regularly adapted to the needs of the public. These utilization trends can be derived using public data collection. The more diverse and rich the data sets are, the easier it is to optimize public space design through data analysis. Traditional data collection methods such as questionnaires, observations, or videos are either labor intensive or cannot guarantee to preserve the individual's privacy. This work presents a privacy-preserving, low-power, and low-cost smart sensing system that is capable of anonymously collecting data about public space utilization by analyzing the occupancy distribution of public seating. To support future urban planning the sensor nodes are capable of monitoring environmental noise, chair utilization, and their position, temperature, and humidity and provide them over a city-wide Long Range Wide Area Network (LoRaWAN). The final sensing system's robust operation is proven in a trial run at two public squares in a city with 16 sensor nodes over a duration of two months. By consuming 33.65 mWh per day with all subsystems enabled, including sitting detection based on a continuous acceleration measurement operating on a robust and simple threshold algorithm, the custom-designed sensor node achieves continuous monitoring during the 2-month trial run. The evaluation of the experimental results clearly shows how the two locations are used, which confirms the practicability of the proposed solution. All data collected during the field trial is publicly available as open data.

The advent of the Internet of Things (IoT) is boosting the proliferation of sensors and smart devices in industry and daily life. Continuous monitoring IoT systems are also finding applications in agriculture, particularly in the realm of smart farming. The adoption of wearable sensors to record the activity of livestock has garnered increasing interest. Such a device enables farmers to locate, monitor, and constantly assess the health status of their cattle more efficiently and effectively, even in challenging terrain and remote locations. This work presents a maintenance-free and robust smart sensing system that is capable of tracking cattle in remote locations and collecting activity parameters, such as the individual's grazing- and resting time. To support the paradigm of smart farming, the cattle tracker is capable of monitoring the cow's activity by analyzing data from an accelerometer, magnetometer, temperature sensor, and Global Navigation Satellite System (GNSS) module, providing them over Long Range Wide Area Network (LoRaWAN) to a backend server. By consuming 511.9 J per day with all subsystems enabled and a data transmission every 15 minutes, the custom-designed sensor node achieves a battery lifetime of 4 months. When exploiting the integrated solar energy harvesting subsystem, this can be even increased by 40% to up to 6 months. The final sensing system's robust operation is proven in a trial run with two cows on a pasture for over three days. Evaluations of the experimental results clearly show behavior patterns, which confirms the practicability of the proposed solution.

We calculate the production and polarization of direct $J/\psi$ in the improved color evaporation model in $e+p$ photoproduction. We present the production as functions of transverse momentun, mass of the hadronic final state, and inelasticity. We also present the polarization parameters $\lambda_\vartheta$, $\lambda_{\varphi}$, and $\lambda_{\vartheta \varphi}$ in the helicity and the Collins-Soper frames, as well as the frame-invariant polarization parameter $\tilde{\lambda}$ as a function of transverse momentum and inelasticity. We find agreement with both $J/\psi$ unpolarized cross sections and the invariant polarization parameters as a function of $p_T$., Comment: arXiv admin note: substantial text overlap with arXiv:2102.09118