Search

Purpose: The purpose of the study was to evaluate the implementation and impact of a computer science-integrated math curriculum, the Beauty and Joy of Computing (BJC), combined with curriculum and professional development support for Computer Science Principles (CSP) teachers, funded by an Innovation and Research (EIR) grant. Methods: The study used a randomized controlled trial design where schools from the IDEA public schools were randomly assigned to use the BJC curriculum (treatment) or, the business as usual, Code.org (comparison). The study compared AP CSP test scores and student attitudes towards CSP between students in the treatment and comparison group. Results: The study found that the average AP CSP scores were low in both the treatment and comparison groups. About one-quarter of test takers (25%) earned a passing score of 3 or higher (compared to national average of 64%). The results of the analyses show that there were no statistically significant differences between the BJC and comparison groups. The study also showed that students in the treatment group responded more favorably on all student engagement in CSP constructs than comparison students, the size of the differences between the treatment and comparison condition were small and none were statistically significant. Implications: Since the evaluation examined student outcomes in the same year that the new curriculum was implemented for this first time, finding significant impacts on student outcomes may require more time. Nevertheless, the results of this evaluation inform IDEA staff, practitioners, and other researchers on the intervention's effectiveness, and they provide insights on how to improve curriculum use to further improve student outcomes.

Purpose: The purpose of the study was to evaluate the implementation and impact of the IDEA Public Schools' Mathematics Curriculum Redesign (MCR) effort, funded by an Innovation and Research (EIR) grant. Methods: The study used a quasi-experimental matching design to examine the effect of MCR on student achievement outcomes. The study compared outcomes between the two cohorts of students at IDEA Public Schools that piloted the MCR and matched cohorts of students in IDEA schools that had delayed implementation of MCR. Results:?The study found statistically significant differences in IDEA Geometry test scores (ES = 0.39) between treatment and comparison group students in Grade 9 and on AP Calculus exam scores between treatment and comparison group students in Grade 12 (ES = 0.57). The study also found that the MCR program met the evaluation threshold for fidelity of implementation for the two out of three key program components (development of redesigned math curriculum and teacher use of math curriculum) in some years. The program did not meet the evaluation threshold for fidelity of implementation for the third key program components (teacher training and professional development to use redesigned math curriculum) in any of the years. Implications:?Because curriculum development and implementation are ongoing and iterative, realizing all of the benefits is not immediate and may take more time to achieve desired results. Continuous improvement of the curriculum and consistent supports to teachers will contribute to continued benefits for teachers and students while moving toward reaching anticipated student outcomes.

Unconventional superconductivity, where electron pairing does not involve electron-phonon interactions, is often attributed to magnetic correlations in a material. Well known examples include high-T_c cuprates and uranium-based heavy fermion superconductors. Less explored are unconventional superconductors with strong spin-orbit coupling, where interactions between spin-polarised electrons and external magnetic field can result in multiple superconducting phases and field-induced transitions between them, a rare phenomenon in the superconducting state. Here we report a magnetic-field driven phase transition in \beta-PdBi2, a layered non-magnetic superconductor. Our tunnelling spectroscopy on thin PdBi2 monocrystals incorporated in planar superconductor-insulator-normal metal junctions reveals a marked discontinuity in the superconducting properties with increasing in-plane field, which is consistent with a transition from conventional (s-wave) to nodal pairing. Our theoretical analysis suggests that this phase transition may arise from spin polarisation and spin-momentum locking caused by locally broken inversion symmetry, with p-wave pairing becoming energetically favourable in high fields. Our findings also reconcile earlier predictions of unconventional multigap superconductivity in \beta-PdBi2 with previous experiments where only a single s-wave gap could be detected., Comment: 29 pages, including 4 main Figures, Methods, 8 Supplementary Figures and 5 Supplementary Notes

Distributing entanglement over long distances remains a challenge due to its fragility when exposed to environmental effects. In this work, we compare various entanglement distribution protocols in a realistic noisy fiber network. We focus specifically on two schemes that only require the sending of a non-entangled carrier photon to remote nodes of the network. These protocols rely on optical CNOT gates and we vary the probability with which they can be successfully performed. Encoding our entangled states in photon polarization, we analyse the effect of depolarizing noise on the photonic states as the carrier passes through the fibers. Building a robust model of photon loss and calculating the distillable entanglement of the noisy states, we find the entanglement distribution rate. We discover that methods involving a separable carrier can reach a higher rate than the standard entanglement distribution protocol, provided that the success probability of the optical CNOT gates is sufficiently high., Comment: 12 pages, 10 figures. Comments welcome

Despite the promising performance of convolutional neural networks (CNNs) in brain tumor diagnosis from magnetic resonance imaging (MRI), their integration into the clinical workflow has been limited. That is mainly due to the fact that the features contributing to a model's prediction are unclear to radiologists and hence, clinically irrelevant, i.e., lack of explainability. As the invaluable sources of radiologists' knowledge and expertise, radiology reports can be integrated with MRI in a contrastive learning (CL) framework, enabling learning from image-report associations, to improve CNN explainability. In this work, we train a multimodal CL architecture on 3D brain MRI scans and radiology reports to learn informative MRI representations. Furthermore, we integrate tumor location, salient to several brain tumor analysis tasks, into this framework to improve its generalizability. We then apply the learnt image representations to improve explainability and performance of genetic marker classification of pediatric Low-grade Glioma, the most prevalent brain tumor in children, as a downstream task. Our results indicate a Dice score of 31.1% between the model's attention maps and manual tumor segmentation (as an explainability measure) with test classification performance of 87.7%, significantly outperforming the baselines. These enhancements can build trust in our model among radiologists, facilitating its integration into clinical practices for more efficient tumor diagnosis.

Fine-tuning language models has become increasingly popular following the proliferation of open models and improvements in cost-effective parameter efficient fine-tuning. However, fine-tuning can influence model properties such as safety. We assess how fine-tuning can impact different open models' propensity to output toxic content. We assess the impacts of fine-tuning Gemma, Llama, and Phi models on toxicity through three experiments. We compare how toxicity is reduced by model developers during instruction-tuning. We show that small amounts of parameter-efficient fine-tuning on developer-tuned models via low-rank adaptation on a non-adversarial dataset can significantly alter these results across models. Finally, we highlight the impact of this in the wild, demonstrating how toxicity rates of models fine-tuned by community contributors can deviate in hard-to-predict ways., Comment: To be presented at NeurIPS 2024 Safe Generative AI Workshop

In this article, we present results of our recent work of epitaxially-grown aluminum (epi-Al) on antimonide heterostructures, where the epi-Al thin film is grown at either room temperature or below zero $^o$C. A sharp superconducting transition at $T \sim 1.3$ K is observed in these epi-Al films, and the critical magnetic field follows the BCS (Bardeen-Cooper-Schrieffer) model. We further show that supercurrent states are achieved in Josephson junctions fabricated in the epi-Al/antimonide heterostructures with mobility $\mu \sim 1.0 \times 10^6$ cm$^2$/Vs, making these heterostructures a promising platform for the exploration of Josephson junction effects for quantum microelectronics applications, and the realization of robust topological superconducting states that potentially allow the realization of intrinsically fault-tolerant qubits and quantum gates.

Josephson junction field-effect transistors (JJFETs) have recently re-emerged as promising candidates for superconducting computing. For JJFETs to perform Boolean logic operations, the so-called gain factor $\alpha_{R}$ must be larger than 1. In a conventional JJFET made with a classical channel material, due to a gradual dependence of superconducting critical current on the gate bias, $\alpha_{R}$ is much smaller than 1. In this Letter, we propose a new device structure of quantum enhanced JJFETs in a zero-energy-gap InAs/GaSb heterostructure. We demonstrate that, due to an excitonic insulator quantum phase transition in this zero-gap heterostructure, the superconducting critical current displays a sharp transition as a function of gate bias, and the deduced gain factor $\alpha_{R}$ ~ 0.06 is more than 50 times that (~ 0.001) reported in a classical JJFET. Further optimization may allow achieving a gain factor larger than 1 for logic applications.

For heterostructures formed by a quantum spin Hall insulator (QSHI) placed in proximity to a superconductor (SC), no external magnetic field is necessary to drive the system into a phase supporting topological superconductivity with Majorana zero energy states, making them very attractive for the realization of non-Abelian states and fault-tolerant qubits. Despite considerable work investigating QSHI edge states, there is still an open question about their resilience to large magnetic fields and the implication of such resilience for the formation of a quasi-1D topological superconducting state. In this work, we investigate the transport properties of helical edge states in a QSHI-SC junction formed by a InAs/GaSb (15nm/5nm) double quantum well and a superconducting tantalum (Ta) constriction. We observe a robust conductance plateau up to 2 T, signaling resilient edge state transport. Such resilience is consistent with the Dirac point for the edge states being buried in the bulk valence band. Using a modified Landauer-Buttiker analysis, we find that the conductance is consistent with 98% Andreev reflection probability owing to the high transparency of the InAs/GaSb-Ta interface. We further theoretically show that a buried Dirac point does not affect the robustness of the quasi-1D topological superconducting phase, and favors the hybridization of Majorana Kramer pairs and fermionic modes in the QSHI resulting in extended MKP states, highlighting the subtle role of buried Dirac points in probing MKPs., Comment: 14 pages, 10 figures, and 2 tables

We report the rotationally resolved spectrum of a single polyatomic molecular ion in the gas phase. Building upon the recently developed inelastic recoil spectroscopy (IRS) technique, we have achieved a spectral resolution sufficient to observe resolved rotational-vibrational transitions of a trapped cyclopropenyl cation, c-C3H3+. The high precision of IRS shown in this work presents an attractive platform for astrochemistry, single molecule chiral detection, and tests of fundamental physics.

The intrigue of waves on periodic lattices and gratings has resonated with physicists and mathematicians alike for decades. In-depth analysis has been devoted to the seemingly simplest array system: a one-dimensionally periodic lattice of two-dimensional scatterers embedded in a dispersionless medium governed by the Helmholtz equation. We investigate such a system and experimentally confirm the existence of a new class of generalised Rayleigh--Bloch waves that have been recently theorised to exist in classical wave regimes, without the need for resonant scatterers. Airborne acoustics serves as such a regime and here we experimentally observe the first generalised Rayleigh--Bloch waves above the first cut-off, i.e., in the radiative regime. We consider radiative acoustic lattice resonances along a diffraction grating and connect them to generalised Rayleigh--Bloch waves by considering both short and long arrays of non-resonant 2D cylindrical Neumann scatterers embedded in air. On short arrays, we observe finite lattice resonances under continuous wave excitation, and on long arrays, we observe propagating Rayleigh--Bloch waves under pulsed excitation. We interpret their existence by considering multiple wave scattering theory and, in doing so, unify differing nomenclatures used to describe waves on infinite periodic and finite arrays and the interpretation of their dispersive properties., Comment: 10 pages, 5 figures

We study the Picard-Newton iteration for the incompressible Boussinesq equations, which is a two-step iteration resulting from the composition of the Picard and Newton iterations. We prove that this iterative method retains Newton's quadratic convergence but has less restrictive sufficient conditions for convergence than Newton and also is unconditionally stable under a small data condition. In this sense, Picard-Newton can be considered as a Newton iteration that is nonlinearly preconditioned with Picard. Our numerical tests illustrate this quadratic convergence and stability on benchmark problems. Furthermore, the tests show convergence for significantly higher Rayleigh number than both Picard and Newton, which illustrates the larger convergence basin of Picard-Newton that the theory predicts. We also introduce Anderson acceleration into the Picard step in our Picard-Newton numerical tests, and this enables convergence for even higher Rayleigh number., Comment: 20 pages, 14 figures, 5 tables

Populism is a concept that is often used but notoriously difficult to measure. Common qualitative measurements like holistic grading or content analysis require great amounts of time and labour, making it difficult to quickly scope out which politicians should be classified as populist and which should not, while quantitative methods show mixed results when it comes to classifying populist rhetoric. In this paper, we develop a pipeline to train and validate an automated classification model to estimate the use of populist language. We train models based on sentences that were identified as populist and pluralist in 300 US governors' speeches from 2010 to 2018 and in 45 speeches of presidential candidates in 2016. We find that these models classify most speeches correctly, including 84% of governor speeches and 89% of presidential speeches. These results extend to different time periods (with 92% accuracy on more recent American governors), different amounts of data (with as few as 70 training sentences per category achieving similar results), and when classifying politicians instead of individual speeches. This pipeline is thus an effective tool that can optimise the systematic and swift classification of the use of populist language in politicians' speeches.

Optimization-based coupling (OBC) is an attractive alternative to traditional Lagrange multiplier approaches in multiple modeling and simulation contexts. However, application of OBC to time-dependent problem has been hindered by the computational costs of finding the stationary points of the associated Lagrangian, which requires primal and adjoint solves. This issue can be mitigated by using OBC in conjunction with computationally efficient reduced order models (ROM). To demonstrate the potential of this combination, in this paper we develop an optimization-based ROM-ROM coupling for a transient advection-diffusion transmission problem. The main challenge in this formulation is the generation of adjoint snapshots and reduced bases for the adjoint systems required by the optimizer. One of the main contributions of the paper is a new technique for efficient adjoint snapshot collection for gradient-based optimizers in the context of optimization-based ROM-ROM couplings. We present numerical studies demonstrating the accuracy of the approach along with comparison between various approaches for selecting a reduced order basis for the adjoint systems, including decay of snapshot energy, iteration counts, and timings., Comment: 27 pages, 5 figures, 12 tables

Combined, joint, intra-governmental, inter-agency and multinational (CJIIM) operations require rapid data sharing without the bottlenecks of metadata curation and alignment. Curation and alignment is particularly infeasible for external open source information (OSINF), e.g., social media, which has become increasingly valuable in understanding unfolding situations. Large language models (transformers) facilitate semantic data and metadata alignment but are inefficient in CJIIM settings characterised as denied, degraded, intermittent and low bandwidth (DDIL). Vector symbolic architectures (VSA) support semantic information processing using highly compact binary vectors, typically 1-10k bits, suitable in a DDIL setting. We demonstrate a novel integration of transformer models with VSA, combining the power of the former for semantic matching with the compactness and representational structure of the latter. The approach is illustrated via a proof-of-concept OSINF data discovery portal that allows partners in a CJIIM operation to share data sources with minimal metadata curation and low communications bandwidth. This work was carried out as a bridge between previous low technology readiness level (TRL) research and future higher-TRL technology demonstration and deployment.

We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of evaluation. In addition, we discuss issues around safety and representation, as well as methods we used to minimize the potential harm of our models.

Distributing quantum correlations to each node of a network is a key aspect of quantum networking. Here, we present a robust, physically-motivated protocol by which global quantum correlations, as characterised by the discord, can be distributed to quantum memories using a mixed state of information carriers which possess only classical correlations. In addition to this, said distribution is measurement-outcome independent, and the distribution is done using only bilocal unitary operations and projective measurements. We also explore the scaling of this protocol for larger networks and illustrate the structure of the quantum correlations, showing its dependence on the local operations performed., Comment: 10 pages, 5 figures. Comments are welcome

Transforming a random variable to improve its normality leads to a followup test for whether the transformed variable follows a normal distribution. Previous work has shown that the Anderson Darling test for normality suffers from resubstitution bias following Box-Cox transformation, and indicates normality much too often. The work reported here extends this by adding the Shapiro-Wilk statistic and the two-parameter Box Cox transformation, all of which show severe bias. We also develop a recalibration to correct the bias in all four settings. The methodology was motivated by finding reference ranges in biomarker studies where parametric analysis, possibly on a power-transformed measurand, can be much more informative than nonparametric. Setting environmental standards illustrates another potential application., Comment: 27 pages, 16 figures

The Milky Way is a mosaic of stars from different origins. In particular, metal-poor accreted star candidates offer a unique opportunity to better understand the accretion history of the Milky Way. In this work, we aim to explore the assembly history of the Milky Way by investigating accreted stars in terms of their ages, dynamical properties, and chemical abundances. We also aim to better characterize the impact of incorporating asteroseismic information on age and chemical abundance calculations of metal-poor accreted stars for which TESS data is available. In this study, we conducted an in-depth examination of 30 metal-poor accreted star candidates, using TESS and Gaia data, as well as MIKE spectra. We find satisfactory agreement between seismic and predicted/spectroscopic surface gravity (log g) values, demonstrating the reliability of spectroscopic data from our methodology. We found that while age determination is highly dependent on the log g and asteroseismic information used, the overall chemical abundance distributions are similar for different log g. However, we found that calcium (Ca) abundances are more sensitive to the adopted log g. Our study reveals that the majority of our stars have properties compatible to those reported for the Gaia-Sausage-Enceladus, with a minority of stars that might be associated to Splash. We found an age distribution with a median of 11.3 Gyr with lower and upper uncertainties of 4.1 and 1.3 Gyr respectively when including asteroseismic information. As regarding some key chemical signatures we note that these stars are metal-poor ([Fe/H]) < -0.8), alpha-rich ([alpha]/Fe] > 0.2), copper-poor ([Cu/Fe] < 0 ) and with chemical abundances typical of accreted stars. These findings illustrate the importance of multi-dimensional analyses in unraveling the complex accretion history of the Milky Way., Comment: Accepted for publication in A&A

We present a detailed study of the chemical diversity of the metal-poor Milky Way (MW) using data from the GALAH DR3 survey. Considering 17 chemical abundances relative to iron ([X/Fe]) for 9,923 stars, we employ Principal Component Analysis (PCA) and Extreme Deconvolution (XD) to identify 10 distinct stellar groups. This approach, free from chemical or dynamical cuts, reveals known populations, including the accreted halo, thick disc, thin disc, and in-situ halo. The thick disc is characterised by multiple substructures, suggesting it comprises stars formed in diverse environments. Our findings highlight the limited discriminatory power of magnesium in separating accreted and disc stars. Elements such as Ba, Al, Cu, and Sc are critical in distinguishing disc from accreted stars, while Ba, Y, Eu and Zn differentiate disc and accreted stars from the in-situ halo. This study demonstrates the potential power of combining a latent space representation of the data (PCA) with a clustering algorithm (XD) in Galactic archaeology, in providing new insights into the galaxy's assembly and evolutionary history.

Mitotic activity is an important feature for grading several cancer types. Counting mitotic figures (MFs) is a time-consuming, laborious task prone to inter-observer variation. Inaccurate recognition of MFs can lead to incorrect grading and hence potential suboptimal treatment. In this study, we propose an artificial intelligence (AI)-aided approach to detect MFs in digitised haematoxylin and eosin-stained whole slide images (WSIs). Advances in this area are hampered by the limited number and types of cancer datasets of MFs. Here we establish the largest pan-cancer dataset of mitotic figures by combining an in-house dataset of soft tissue tumours (STMF) with five open-source mitotic datasets comprising multiple human cancers and canine specimens (ICPR, TUPAC, CCMCT, CMC and MIDOG++). This new dataset identifies 74,620 MFs and 105,538 mitotic-like figures. We then employed a two-stage framework (the Optimised Mitoses Generator Network (OMG-Net) to classify MFs. The framework first deploys the Segment Anything Model (SAM) to automate the contouring of MFs and surrounding objects. An adapted ResNet18 is subsequently trained to classify MFs. OMG-Net reaches an F1-score of 0.84 on pan-cancer MF detection (breast carcinoma, neuroendocrine tumour and melanoma), largely outperforming the previous state-of-the-art MIDOG++ benchmark model on its hold-out testing set (e.g. +16% F1-score on breast cancer detection, p<0.001) thereby providing superior accuracy in detecting MFs on various types of tumours obtained with different scanners.

Two decades on, the study of hypervelocity stars is still in its infancy. These stars can provide novel constraints on the total mass of the Galaxy and its Dark Matter distribution. However how these stars are accelerated to such high velocities is unclear. Various proposed production mechanisms for these stars can be distinguished using chemo-dynamic tagging. The advent of Gaia and other large surveys have provided hundreds of candidate hyper velocity objects to target for ground based high resolution follow-up observations. We conduct high resolution spectroscopic follow-up observations of 16 candidate late-type hyper velocity stars using the Apache Point Observatory and the McDonald Observatory. We derive atmospheric parameters and chemical abundances for these stars. We measure up to 22 elements, including the following nucleosynthetic families: {\alpha} (Mg, Si, Ca, Ti), light/Odd-Z (Na, Al, V, Cu, Sc), Fe-peak (Fe, Cr, Mn, Co, Ni, Zn), and Neutron Capture (Sr, Y, Zr, Ba, La, Nd, Eu). Our kinematic analysis shows one candidate is unbound, two are marginally bound, and the remainder are bound to the Galaxy. Finally, for the three unbound or marginally bound stars, we perform orbit integration to locate possible globular cluster or dwarf galaxy progenitors. We do not find any likely candidate systems for these stars and conclude that the unbound stars are likely from the the stellar halo, in agreement with the chemical results. The remaining bound stars are all chemically consistent with the stellar halo as well., Comment: 16 pages, 7 figures

In this report, we introduce the Gemini 1.5 family of models, representing the next generation of highly compute-efficient multimodal models capable of recalling and reasoning over fine-grained information from millions of tokens of context, including multiple long documents and hours of video and audio. The family includes two new models: (1) an updated Gemini 1.5 Pro, which exceeds the February version on the great majority of capabilities and benchmarks; (2) Gemini 1.5 Flash, a more lightweight variant designed for efficiency with minimal regression in quality. Gemini 1.5 models achieve near-perfect recall on long-context retrieval tasks across modalities, improve the state-of-the-art in long-document QA, long-video QA and long-context ASR, and match or surpass Gemini 1.0 Ultra's state-of-the-art performance across a broad set of benchmarks. Studying the limits of Gemini 1.5's long-context ability, we find continued improvement in next-token prediction and near-perfect retrieval (>99%) up to at least 10M tokens, a generational leap over existing models such as Claude 3.0 (200k) and GPT-4 Turbo (128k). Finally, we highlight real-world use cases, such as Gemini 1.5 collaborating with professionals on completing their tasks achieving 26 to 75% time savings across 10 different job categories, as well as surprising new capabilities of large language models at the frontier; when given a grammar manual for Kalamang, a language with fewer than 200 speakers worldwide, the model learns to translate English to Kalamang at a similar level to a person who learned from the same content.

For machine learning components used as part of autonomous systems (AS) in carrying out critical tasks it is crucial that assurance of the models can be maintained in the face of post-deployment changes (such as changes in the operating environment of the system). A critical part of this is to be able to monitor when the performance of the model at runtime (as a result of changes) poses a safety risk to the system. This is a particularly difficult challenge when ground truth is unavailable at runtime. In this paper we introduce a process for creating safety monitors for ML components through the use of degraded datasets and machine learning. The safety monitor that is created is deployed to the AS in parallel to the ML component to provide a prediction of the safety risk associated with the model output. We demonstrate the viability of our approach through some initial experiments using publicly available speed sign datasets.

We propose a model order reduction approach to speed up the computation of seismograms, i.e. the solution of the seismic wave equation evaluated at a receiver location, for different model parameters. Our approach achieves a reduction of the unknowns by a factor of approximately 1000 for various numerical experiments for a 2D subsurface model of Groningen, the Netherlands, even if the wave speeds of the subsurface are relatively varied. Moreover, using parallel computing, the reduced model can approximate the (time domain) seismogram in a lower wall clock time than an implicit Newmark-beta method. To realize this reduction, we exploit the fact that seismograms are low-pass filtered for the observed seismic events by considering the Laplace-transformed problem in frequency domain. Therefore, we can avoid the high frequencies that would require many reduced basis functions to reach the desired accuracy and generally make the reduced order approximation of wave problems challenging. Instead, we can prove for our ansatz that for a fixed subsurface model the reduced order approximation converges exponentially fast in the frequency range of interest in the Laplace domain. We build the reduced model from solutions of the Laplace-transformed problem via a (Proper Orthogonal Decomposition-)Greedy algorithm targeting the construction of the reduced model to the time domain seismograms; the latter is achieved by using an a posteriori error estimator that does not require computing any time domain counterparts. Finally, we show that we obtain a stable reduced model thus overcoming the challenge that standard model reduction approaches do not necessarily yield a stable reduced model for wave problems., Comment: 26 pages, plus supplementary material

With the advent of novel cancer treatment options such as immunotherapy, studying the tumour immune micro-environment is crucial to inform on prognosis and understand response to therapeutic agents. A key approach to characterising the tumour immune micro-environment may be through combining (1) digitised microscopic high-resolution optical images of hematoxylin and eosin (H&E) stained tissue sections obtained in routine histopathology examinations with (2) automated immune cell detection and classification methods. However, current individual immune cell classification models for digital pathology present relatively poor performance. This is mainly due to the limited size of currently available datasets of individual immune cells, a consequence of the time-consuming and difficult problem of manually annotating immune cells on digitised H&E whole slide images. In that context, we introduce Immunocto, a massive, multi-million automatically generated database of 6,848,454 human cells, including 2,282,818 immune cells distributed across 4 subtypes: CD4$^+$ T cell lymphocytes, CD8$^+$ T cell lymphocytes, B cell lymphocytes, and macrophages. For each cell, we provide a 64$\times$64 pixels H&E image at $\mathbf{40}\times$ magnification, along with a binary mask of the nucleus and a label. To create Immunocto, we combined open-source models and data to automatically generate the majority of contours and labels. The cells are obtained from a matched H&E and immunofluorescence colorectal dataset from the Orion platform, while contours are obtained using the Segment Anything Model. A classifier trained on H&E images from Immunocto produces an average F1 score of 0.74 to differentiate the 4 immune cell subtypes and other cells. Immunocto can be downloaded at: https://zenodo.org/uploads/11073373.

Identifying critical research within the growing body of academic work is an intrinsic aspect of conducting quality research. Systematic review processes used in evidence-based medicine formalise this as a procedure that must be followed in a research program. However, it comes with an increasing burden in terms of the time required to identify the important articles of research for a given topic. In this work, we develop a method for building a general-purpose filtering system that matches a research question, posed as a natural language description of the required content, against a candidate set of articles obtained via the application of broad search terms. Our results demonstrate that transformer models, pre-trained on biomedical literature, and then fine tuned for the specific task, offer a promising solution to this problem. The model can remove large volumes of irrelevant articles for most research questions. Furthermore, analysis of the specific research questions in our training data suggest natural avenues for further improvement., Comment: Paper Submitted to `Multimedia Tools and Applications`

Chemical cartography of the Galactic disk provides insights to its structure and assembly history over cosmic time. In this work, we use chemical cartography to explore chemical gradients and azimuthal substructure in the Milky Way disk with giant stars from APOGEE DR17. We confirm the existence of a radial metallicity gradient in the disk of $\Delta$[Fe/H]/$\Delta$R $\sim -0.066 \pm 0.0004$ dex/kpc and a vertical metallicity gradient of $\Delta$[Fe/H]/$\Delta$Z $\sim -0.164 \pm 0.001$ dex/kpc. We find azimuthal variations ($\pm0.1$ dex) on top of the radial metallicity gradient that have been previously established with other surveys. The APOGEE giants show strong correlations with stellar age and the intensity of azimuthal variations in iron; older stellar populations show the largest deviations from the radial metallicity gradient. Beyond iron, we show that other elements (e.g., Mg, O) display azimuthal variations at the $\pm0.05$ dex-level across the Galactic disk. We illustrate that moving into the orbit-space could help constrain the mechanisms producing these azimuthal metallicity variations. These results suggest that the spiral arms of the Galaxy are not solely responsible for azimuthal metallicity variations and other Galactic processes are at play., Comment: 19 pages, 11 figures

White dwarfs (WDs) polluted by exoplanetary material provide the unprecedented opportunity to directly observe the interiors of exoplanets. However, spectroscopic surveys are often limited by brightness constraints, and WDs tend to be very faint, making detections of large populations of polluted WDs difficult. In this paper, we aim to increase considerably the number of WDs with multiple metals in their atmospheres. Using 96,134 WDs with Gaia DR3 BP/RP (XP) spectra, we constructed a 2D map using an unsupervised machine learning technique called Uniform Manifold Approximation and Projection (UMAP) to organize the WDs into identifiable spectral regions. The polluted WDs are among the distinct spectral groups identified in our map. We have shown that this selection method could potentially increase the number of known WDs with 5 or more metal species in their atmospheres by an order of magnitude. Such systems are essential for characterizing exoplanet diversity and geology., Comment: 16 pages, 10 figures, accepted to ApJ on June 13, 2024

Online commerce relies heavily on user generated reviews to provide unbiased information about products that they have not physically seen. The importance of reviews has attracted multiple exploitative online behaviours and requires methods for monitoring and detecting reviews. We present a machine learning methodology for review detection and extraction, and demonstrate that it generalises for use across websites that were not contained in the training data. This method promises to drive applications for automatic detection and evaluation of reviews, regardless of their source. Furthermore, we showcase the versatility of our method by implementing and discussing three key applications for analysing reviews: Sentiment Inconsistency Analysis, which detects and filters out unreliable reviews based on inconsistencies between ratings and comments; Multi-language support, enabling the extraction and translation of reviews from various languages without relying on HTML scraping; and Fake review detection, achieved by integrating a trained NLP model to identify and distinguish between genuine and fake reviews.

This paper focuses on the opportunities and the ethical and societal risks posed by advanced AI assistants. We define advanced AI assistants as artificial agents with natural language interfaces, whose function is to plan and execute sequences of actions on behalf of a user, across one or more domains, in line with the user's expectations. The paper starts by considering the technology itself, providing an overview of AI assistants, their technical foundations and potential range of applications. It then explores questions around AI value alignment, well-being, safety and malicious uses. Extending the circle of inquiry further, we next consider the relationship between advanced AI assistants and individual users in more detail, exploring topics such as manipulation and persuasion, anthropomorphism, appropriate relationships, trust and privacy. With this analysis in place, we consider the deployment of advanced assistants at a societal scale, focusing on cooperation, equity and access, misinformation, economic impact, the environment and how best to evaluate advanced AI assistants. Finally, we conclude by providing a range of recommendations for researchers, developers, policymakers and public stakeholders.

Safety and responsibility evaluations of advanced AI models are a critical but developing field of research and practice. In the development of Google DeepMind's advanced AI models, we innovated on and applied a broad set of approaches to safety evaluation. In this report, we summarise and share elements of our evolving approach as well as lessons learned for a broad audience. Key lessons learned include: First, theoretical underpinnings and frameworks are invaluable to organise the breadth of risk domains, modalities, forms, metrics, and goals. Second, theory and practice of safety evaluation development each benefit from collaboration to clarify goals, methods and challenges, and facilitate the transfer of insights between different stakeholders and disciplines. Third, similar key methods, lessons, and institutions apply across the range of concerns in responsibility and safety - including established and emerging harms. For this reason it is important that a wide range of actors working on safety evaluation and safety research communities work together to develop, refine and implement novel evaluation approaches and best practices, rather than operating in silos. The report concludes with outlining the clear need to rapidly advance the science of evaluations, to integrate new evaluations into the development and governance of AI, to establish scientifically-grounded norms and standards, and to promote a robust evaluation ecosystem., Comment: 10 pages excluding bibliography

The recently proposed zero-added-loss multiplexing (ZALM) source of entangled photons enables higher efficiency in entanglement distribution than spontaneous parametric down-conversion sources and can be carried out using both space-to-ground and ground-to-ground links. We demonstrate the flexibility of ZALM architectures to be adapted to alternative entanglement distribution protocols. Focusing on the counter-intuitive result that entanglement can be generated between distant parties without using any entanglement as a resource, we analyze two protocols for entanglement distribution to memories via separable states. Modelling them in a ZALM setup, we consider the effects of noise both in the communication channels and in the memories. We thereby identify the optimal protocol to use, with respect to the highest entanglement generated, given the noise conditions of the network., Comment: 11 pages, 6 figures

Telecom C-band single photons exhibit the lowest attenuation in optical fibers, enabling long-haul quantum-secured communication. However, efficient coupling with optical fibers is crucial for these single photons to be effective carriers in long-distance transmission. In this work, we demonstrate an efficient fiber-coupled single photon source at the telecom C-band using InAs/InP quantum dots coupled to a tapered nanobeam. The tapered nanobeam structure facilitates directional emission that is mode-matched to a lensed fiber, resulting in a collection efficiency of up to 65% from the nanobeam to a single-mode fiber. Using this approach, we demonstrate single photon count rates of 575 $\pm$ 5 Kcps and a single photon purity of $g^2$ (0) = 0.015 $\pm$ 0.003. Additionally, we demonstrate Hong-Ou Mandel interference from the emitted photons with a visibility of 0.84 $\pm$ 0.06. From these measurements, we determine a photon coherence time of 450 $\pm$ 20 ps, a factor of just 8.3 away from the lifetime limit. This work represents an important step towards the development of telecom C-band single-photon sources emitting bright, pure, and indistinguishable photons, which are necessary to realize fiber-based long-distance quantum networks

This report introduces a new family of multimodal models, Gemini, that exhibit remarkable capabilities across image, audio, video, and text understanding. The Gemini family consists of Ultra, Pro, and Nano sizes, suitable for applications ranging from complex reasoning tasks to on-device memory-constrained use-cases. Evaluation on a broad range of benchmarks shows that our most-capable Gemini Ultra model advances the state of the art in 30 of 32 of these benchmarks - notably being the first model to achieve human-expert performance on the well-studied exam benchmark MMLU, and improving the state of the art in every one of the 20 multimodal benchmarks we examined. We believe that the new capabilities of the Gemini family in cross-modal reasoning and language understanding will enable a wide variety of use cases. We discuss our approach toward post-training and deploying Gemini models responsibly to users through services including Gemini, Gemini Advanced, Google AI Studio, and Cloud Vertex AI.

We introduce a new system of surface integral equations for Maxwell's transmission problem in three dimensions. This system has two remarkable features, both of which we prove. First, it is well-posed at all frequencies. Second, the underlying linear operator has a uniformly bounded inverse as the frequency approaches zero, ensuring that there is no low-frequency breakdown. The system is derived from a formulation we introduced in our previous work, which required additional integral constraints to ensure well -posedness across all frequencies. In this study, we eliminate those constraints and demonstrate that our new self adjoint, constraints-free linear system expressed in the desirable form of an identity plus a compact weakly-singular operator is stable for all frequencies. Furthermore, we propose and analyze a fully discrete numerical method for these systems and provide a proof of spectrally accurate convergence for the computational method. We also computationally demonstrate the high-order accuracy of the algorithm using benchmark scatterers with curved surfaces., Comment: 11 figures

Earth system models (ESMs) are vital for understanding past, present, and future climate, but they suffer from legacy technical infrastructure. ESMs are primarily implemented in Fortran, a language that poses a high barrier of entry for early career scientists and lacks a GPU runtime, which has become essential for continued advancement as GPU power increases and CPU scaling slows. Fortran also lacks differentiability - the capacity to differentiate through numerical code - which enables hybrid models that integrate machine learning methods. Converting an ESM from Fortran to Python/JAX could resolve these issues. This work presents a semi-automated method for translating individual model components from Fortran to Python/JAX using a large language model (GPT-4). By translating the photosynthesis model from the Community Earth System Model (CESM), we demonstrate that the Python/JAX version results in up to 100x faster runtimes using GPU parallelization, and enables parameter estimation via automatic differentiation. The Python code is also easy to read and run and could be used by instructors in the classroom. This work illustrates a path towards the ultimate goal of making climate models fast, inclusive, and differentiable.

Pediatric Low-Grade Neuroepithelial Tumors (PLGNT) are the most common pediatric cancer type, accounting for 40% of brain tumors in children, and identifying PLGNT molecular subtype is crucial for treatment planning. However, the gold standard to determine the PLGNT subtype is biopsy, which can be impractical or dangerous for patients. This research improves the performance of Convolutional Neural Networks (CNNs) in classifying PLGNT subtypes through MRI scans by introducing a loss function that specifically improves the model's Area Under the Receiver Operating Characteristic (ROC) Curve (AUROC), offering a non-invasive diagnostic alternative. In this study, a retrospective dataset of 339 children with PLGNT (143 BRAF fusion, 71 with BRAF V600E mutation, and 125 non-BRAF) was curated. We employed a CNN model with Monte Carlo random data splitting. The baseline model was trained using binary cross entropy (BCE), and achieved an AUROC of 86.11% for differentiating BRAF fusion and BRAF V600E mutations, which was improved to 87.71% using our proposed AUROC loss function (p-value 0.045). With multiclass classification, the AUROC improved from 74.42% to 76. 59% (p-value 0.0016).

The recent astrometric data of hundreds of millions of stars from Gaia DR3 has allowed a precise determination of the Milky Way rotation curve up to $28$~kpc. The data suggests a rapid decline in the density of dark matter beyond $19$~kpc. We fit the whole rotation curve with four components (gas, disk, bulge and halo) and compute the microlensing optical depth to the Large Magellanic Cloud. With this model of the galaxy we reanalyse the microlensing events of the MACHO and EROS-2 Collaborations. Using the published MACHO efficiency function for the duration of their survey, together with the rate of expected events according to the new density profile, we find that the Dark Matter halo could be composed up to $100\%$ of massive compact halo objects for any mass between $10^{-4}$ to $100~M_\odot$, except a broad range around $0.01~M_\odot$, where it cannot be larger than $\sim20~\%$. For the EROS-2 survey, using a modified efficiency curve for consistency with the MACHO analysis, we also find compatibility with a 100\% MACHO halo, but with a tighter constraint around $0.001~M_\odot$ where the halo fraction cannot be larger than $\sim12~\%$. This result assumes that MACHOs all have the same mass. If these were distributed in an extended mass function like that of the Thermal History Model, the constraints are weakened, allowing $100\%$ of all DM in the form of Primordial Black Holes., Comment: 15 pages, 5 figures