Search

Interface-type resistive switching (RS) devices with lower operation current and more reliable switching repeatability exhibits great potential in the applications for data storage devices and ultra-low-energy computing. However, the working mechanism of such interface-type RS devices are much less studied compared to that of the filament-type devices, which hinders the design and application of the novel interface-type devices. In this work, we fabricate a metal/TiOx/TiN/Si (001) thin film memristor by using a one-step pulsed laser deposition. In situ transmission electron microscopy (TEM) imaging and current-voltage (I-V) characteristic demonstrate that the device is switched between high resistive state (HRS) and low resistive state (LRS) in a bipolar fashion with sweeping the applied positive and negative voltages. In situ scanning transmission electron microscopy (STEM) experiments with electron energy loss spectroscopy (EELS) reveal that the charged defects (such as oxygen vacancies) can migrate along the intrinsic grain boundaries of TiOx insulating phase under electric field without forming obvious conductive filaments, resulting in the modulation of Schottky barriers at the metal/semiconductor interfaces. The fundamental insights gained from this study presents a novel perspective on RS processes and opens up new technological opportunities for fabricating ultra-low-energy nitride-based memristive devices.

The XLZD collaboration is developing a two-phase xenon time projection chamber with an active mass of 60 to 80 t capable of probing the remaining WIMP-nucleon interaction parameter space down to the so-called neutrino fog. In this work we show that, based on the performance of currently operating detectors using the same technology and a realistic reduction of radioactivity in detector materials, such an experiment will also be able to competitively search for neutrinoless double beta decay in $^{136}$Xe using a natural-abundance xenon target. XLZD can reach a 3$\sigma$ discovery potential half-life of 5.7$\times$10$^{27}$ yr (and a 90% CL exclusion of 1.3$\times$10$^{28}$ yr) with 10 years of data taking, corresponding to a Majorana mass range of 7.3-31.3 meV (4.8-20.5 meV). XLZD will thus exclude the inverted neutrino mass ordering parameter space and will start to probe the normal ordering region for most of the nuclear matrix elements commonly considered by the community., Comment: 29 pages, 7 figures

This report describes the experimental strategy and technologies for a next-generation xenon observatory sensitive to dark matter and neutrino physics. The detector will have an active liquid xenon target mass of 60-80 tonnes and is proposed by the XENON-LUX-ZEPLIN-DARWIN (XLZD) collaboration. The design is based on the mature liquid xenon time projection chamber technology of the current-generation experiments, LZ and XENONnT. A baseline design and opportunities for further optimization of the individual detector components are discussed. The experiment envisaged here has the capability to explore parameter space for Weakly Interacting Massive Particle (WIMP) dark matter down to the neutrino fog, with a 3$\sigma$ evidence potential for the spin-independent WIMP-nucleon cross sections as low as $3\times10^{-49}\rm cm^2$ (at 40 GeV/c$^2$ WIMP mass). The observatory is also projected to have a 3$\sigma$ observation potential of neutrinoless double-beta decay of $^{136}$Xe at a half-life of up to $5.7\times 10^{27}$ years. Additionally, it is sensitive to astrophysical neutrinos from the atmosphere, sun, and galactic supernovae., Comment: 32 pages, 14 figures

The UltraViolet imaging of the Cosmic Assembly Near-infrared Deep Extragalactic Legacy Survey Fields (UVCANDELS) program provides deep HST F275W and F435W imaging over four CANDELS fields (GOODS-N, GOODS-S, COSMOS, and EGS). We combine this newly acquired UV imaging with existing HST imaging from CANDELS as well as existing ancillary data to obtain robust photometric redshifts and reliable estimates for galaxy physical properties for over 150,000 galaxies in the $\sim$430 arcmin$^2$ UVCANDELS area. Here, we leverage the power of the new UV photometry to not only improve the photometric redshift measurements in these fields, but also constrain the full redshift probability distribution combining multiple redshift fitting tools. Furthermore, using the full UV-to-IR photometric dataset, we measure the galaxy physical properties by fitting templates from population synthesis models with two different parameterizations (flexible and fixed-form) of the star-formation histories (SFHs). Compared to the flexible SFH parametrization, we find that the fixed-form SFHs systematically underestimate the galaxy stellar masses, both at the low- ($\lesssim10^9 M_\odot$) and high- ($\gtrsim10^{10} M_\odot$) mass end, by as much as $\sim0.5$ dex. This underestimation is primarily due the limited ability of fixed-form SFH parameterization to simultaneously capture the chaotic nature of star-formation in these galaxies., Comment: 22 pages, 6 figures; accepted to ApJS; catalogs available via MAST

In order to solve problems of practical importance, quantum computers will likely need to incorporate quantum error correction, where a logical qubit is redundantly encoded in many noisy physical qubits. The large physical-qubit overhead typically associated with error correction motivates the search for more hardware-efficient approaches. Here, using a microfabricated superconducting quantum circuit, we realize a logical qubit memory formed from the concatenation of encoded bosonic cat qubits with an outer repetition code of distance $d=5$. The bosonic cat qubits are passively protected against bit flips using a stabilizing circuit. Cat-qubit phase-flip errors are corrected by the repetition code which uses ancilla transmons for syndrome measurement. We realize a noise-biased CX gate which ensures bit-flip error suppression is maintained during error correction. We study the performance and scaling of the logical qubit memory, finding that the phase-flip correcting repetition code operates below threshold, with logical phase-flip error decreasing with code distance from $d=3$ to $d=5$. Concurrently, the logical bit-flip error is suppressed with increasing cat-qubit mean photon number. The minimum measured logical error per cycle is on average $1.75(2)\%$ for the distance-3 code sections, and $1.65(3)\%$ for the longer distance-5 code, demonstrating the effectiveness of bit-flip error suppression throughout the error correction cycle. These results, where the intrinsic error suppression of the bosonic encodings allows us to use a hardware-efficient outer error correcting code, indicate that concatenated bosonic codes are a compelling paradigm for reaching fault-tolerant quantum computation., Comment: Comments on the manuscript welcome!

We study lattice polytopes which arise as the convex hull of chip vectors for \textit{self-reachable} chip configurations on a tree $T$. We show that these polytopes always have the integer decomposition property and characterize the vertex sets of these polytopes. Additionally, in the case of self-reachable configurations with the smallest possible number of chips, we show that these polytopes are unimodularly equivalent to a unit cube.

Actuator saturation is a common nonlinearity. In wave energy conversion, force saturation conveniently limits drivetrain size and cost with minimal impact on energy generation. However, such nonlinear dynamics typically demand numerical simulation, which increases computational cost and diminishes intuition. This paper instead uses describing functions to approximate a force saturation nonlinearity as a linear impedance mismatch. In the frequency domain, the impact of controller impedance mismatch (such as force limit, finite bandwidth, or parameter error) on electrical power production is shown analytically and graphically for a generic nondimensionalized single degree of freedom wave energy converter in regular waves. Results are visualized with Smith charts. Notably, systems with a specific ratio of reactive to real mechanical impedance are least sensitive to force limits, a criteria which conflicts with resonance and bandwidth considerations. The describing function method shows promise to enable future studies such as large-scale design optimization and co-design., Comment: 6 pages, 7 figures. For code, see https://github.com/symbiotic-engineering/IFAC_CAMS_2024/ . To be presented at IFAC CAMS 2024 conference and to appear in IFAC-PapersOnLine

In this paper, we explore the notion of a \emph{self-reachable} chip configuration on a simple graph, that is a chip configuration which can be re-obtained from itself after a (nonempty) sequence of vertex firings. In particular, we focus on the case of trees and provide a characterization for such configurations, as well as show that all self-reachable configurations with the same number of chips on a tree are reachable from one another. We conclude with a recursive enumeration formula for the number of self-reachable configurations.

The Vietoris-Rips filtration, the standard filtration on metric data in topological data analysis, is notoriously sensitive to outliers. Sheehy's subdivision-Rips bifiltration $\mathcal{SR}(-)$ is a density-sensitive refinement that is robust to outliers in a strong sense, but whose 0-skeleton has exponential size. For $X$ a finite metric space of constant doubling dimension and fixed $\epsilon>0$, we construct a $(1+\epsilon)$-homotopy interleaving approximation of $\mathcal{SR}(X)$ whose $k$-skeleton has size $O(|X|^{k+2})$. For $k\geq 1$ constant, the $k$-skeleton can be computed in time $O(|X|^{k+3})$., Comment: 20 pages

Three-dimensional (3D) deep molecular generative models offer the advantage of goal-directed generation based on 3D-dependent properties, such as binding affinity for structure-based design within binding pockets. Traditional benchmarks created to evaluate SMILES or molecular graphs generators, such as GuacaMol or MOSES, are limited to evaluate 3D generators as they do not assess the quality of the generated molecular conformation. In this work, we hence developed GenBench3D, which implements a new benchmark for models producing molecules within a binding pocket. Our main contribution is the Validity3D metric, evaluating the conformation quality using the likelihood of bond lengths and valence angles based on reference values observed in the Cambridge Structural Database. The LiGAN, 3D-SBDD, Pocket2Mol, TargetDiff, DiffSBDD and ResGen models were benchmarked. We show that only between 0% and 11% of generated molecules have valid conformations. Performing local relaxation of generated molecules in the pocket considerably improved the Validity3D for all models by a minimum increase of 40%. For LiGAN, 3D-SBDD, or TargetDiff, the set of valid relaxed molecules shows on average higher Vina score (i.e. worse) than the set of raw generated molecules, indicating that the binding affinity of raw generated molecules might be overestimated. Using the other scoring functions, that give higher importance to ligand strain, only yield improved scores when using valid relaxed molecules. Using valid relaxed molecules, TargetDiff and Pocket2Mol show better median Vina, Glide and Gold PLP scores than other models. We have publicly released GenBench3D on GitHub for broader use: https://github.com/bbaillif/genbench3d

Chirality refers to the asymmetry of objects that cannot be superimposed on their mirror image. It is a concept that exists in various scientific fields and has profound consequences. Although these are perhaps most widely recognized within biology, chemistry, and pharmacology, recent advances in chiral phonons, topological systems, crystal enantiomorphic materials, and magneto-chiral materials have brought this topic to the forefront of condensed matter physics research. Our review discusses the symmetry requirements and the features associated with structural chirality in inorganic materials. This allows us to explore the nature of phase transitions in these systems, the coupling between order parameters, and their impact on the material's physical properties. We highlight essential contributions to the field, particularly recent progress in the study of chiral phonons, altermagnetism, magnetochirality between others. Despite the rarity of naturally occurring inorganic chiral crystals, this review also highlights a significant knowledge gap, presenting challenges and opportunities for structural chirality mostly at the fundamental level, e.g., chiral displacive phase transitions and ferrochirality, possibilities of tuning and switching structural chirality by external means (electric, magnetic, or strain fields), whether chirality could be an independent order parameter, and whether structural chirality could be quantified, etc. Beyond simply summarising this field of research, this review aims to inspire further research in materials science by addressing future challenges, encouraging the exploration of chirality beyond traditional boundaries, and seeking the development of innovative materials with superior or new properties.

We study the size of Sheehy's subdivision bifiltrations, up to homotopy. We focus in particular on the subdivision-Rips bifiltration $\mathcal{SR}(X)$ of a metric space $X$, the only density-sensitive bifiltration on metric spaces known to satisfy a strong robustness property. Given a simplicial filtration $\mathcal{F}$ with a total of $m$ maximal simplices across all indices, we introduce a nerve-based simplicial model for its subdivision bifiltration $\mathcal{SF}$ whose $k$-skeleton has size $O(m^{k+1})$. We also show that the $0$-skeleton of any simplicial model of $\mathcal{SF}$ has size at least $m$. We give several applications: For an arbitrary metric space $X$, we introduce a $\sqrt{2}$-approximation to $\mathcal{SR}(X)$, denoted $\mathcal{J}(X)$, whose $k$-skeleton has size $O(|X|^{k+2})$. This improves on the previous best approximation bound of $\sqrt{3}$, achieved by the degree-Rips bifiltration, which implies that $\mathcal{J}(X)$ is more robust than degree-Rips. Moreover, we show that the approximation factor of $\sqrt{2}$ is tight; in particular, there exists no exact model of $\mathcal{SR}(X)$ with poly-size skeleta. On the other hand, we show that for $X$ in a fixed-dimensional Euclidean space with the $\ell_p$-metric, there exists an exact model of $\mathcal{SR}(X)$ with poly-size skeleta for $p\in \{1, \infty\}$, as well as a $(1+\epsilon)$-approximation to $\mathcal{SR}(X)$ with poly-size skeleta for any $p \in (1, \infty)$ and fixed ${\epsilon > 0}$., Comment: 37 pages

Deep learning-based object detection algorithms enable the simultaneous classification and localization of any number of objects in image data. Many of these algorithms are capable of operating in real-time on high resolution images, attributing to their widespread usage across many fields. We present an end-to-end object detection pipeline designed for real-time rare event searches for the Migdal effect, using high-resolution image data from a state-of-the-art scientific CMOS camera in the MIGDAL experiment. The Migdal effect in nuclear scattering, crucial for sub-GeV dark matter searches, has yet to be experimentally confirmed, making its detection a primary goal of the MIGDAL experiment. Our pipeline employs the YOLOv8 object detection algorithm and is trained on real data to enhance the detection efficiency of nuclear and electronic recoils, particularly those exhibiting overlapping tracks that are indicative of the Migdal effect. When deployed online on the MIGDAL readout PC, we demonstrate our pipeline to process and perform the rare event search on 2D image data faster than the peak 120 frame per second acquisition rate of the CMOS camera. Applying these same steps offline, we demonstrate that we can reduce a sample of 20 million camera frames to around 1000 frames while maintaining nearly all signal that YOLOv8 is able to detect, thereby transforming a rare search into a much more manageable search. Our studies highlight the potential of pipelines similar to ours significantly improving the detection capabilities of experiments requiring rapid and precise object identification in high-throughput data environments.

Transformers have revolutionized machine learning across diverse domains, yet understanding their behavior remains crucial, particularly in high-stakes applications. This paper introduces the contextual counting task, a novel toy problem aimed at enhancing our understanding of Transformers in quantitative and scientific contexts. This task requires precise localization and computation within datasets, akin to object detection or region-based scientific analysis. We present theoretical and empirical analysis using both causal and non-causal Transformer architectures, investigating the influence of various positional encodings on performance and interpretability. In particular, we find that causal attention is much better suited for the task, and that no positional embeddings lead to the best accuracy, though rotary embeddings are competitive and easier to train. We also show that out of distribution performance is tightly linked to which tokens it uses as a bias term.

We compare the various chirality measures most widely used in the literature to quantify chiral symmetry in extended solids, i.e., the continuous chirality measure, the Hausdorff distance, and the angular momentum. By studying these functions in an algebraically tractable case, we can evaluate their strengths and weaknesses when applied to more complex crystals. Going beyond those classical calculations, we propose a new method to quantify the handedness of a crystal based on a pseudoscalar function, i.e., the helicity. This quantity, borrowed from hydrodynamics, can be computed from the eigenvector carrying the system from the high-symmetry non-chiral phase to the low-symmetry chiral phase. Different model systems like K$_3$NiO$_2$, CsCuCl$_3$ and MgTi$_2$O$_4$ are used as test cases where we show the superior interest of using helicity to quantify chirality together with the handedness distinction.

We demonstrate through first-principles calculations applied to the K$_{3}$NiO$_{2}$ crystal that a structural phase transition from an achiral to a chiral phase can be mediated by a degenerate soft phonon mode and controlled by pressure and epitaxial strain. Breaking such degeneracy with an electric field generates a competition between enantiomorphic, polar, and orthorhombic displacements. Originated by the interaction between spontaneous chiral and induced polar and axial modes, an optimal parameter window for converting the handedness of the system into its opposite kind is observed.

The promise of a strong magnetoelectric (ME) coupling in a multiferroic (MF) material is not only of fundamental interest, but also forms the basis of next generation memory devices where the direction of magnetization can be reversed by an external electric field. Using group-theory led first-principles calculations, we determine the ME properties of a relatively understudied family of layered oxides with the general formula $A_4B_3$O$_9$. We show how the tetrahedral rotations in these oxides can lead to a variety of hitherto unknown structural phases with different symmetries. In particular, a polar phase in the $Cmc2_1$ space group has been identified where a weak ferromagnetic mode arises spontaneously via a canting of the antiferromagnetically ordered $B$-site spins. In this polar phase, the polar mode couples to the magnetic modes through a rare $\Gamma$-point ME-MF coupling scheme such that the net magnetization can be directly reversed by an electric field switching of the polar mode. Moreover, in agreement with previous experimental observations, we predict room-temperature magnetism in $A_4B_3$O$_9$ layered oxides which is supported by our calculations of the magnetic exchange interaction parameters, further indicating the potential of these compounds in practical technological applications.

Objectives: To demonstrate that a virtual imaging trial platform can accurately emulate a major clinical trial, specifically the National Lung Screening Trial (NLST) that compared computed tomography (CT) and chest radiography (CXR) imaging for lung cancer screening. Design, Setting, and Participants: A virtual patient population of 294 subjects was created from human models (XCAT) emulating the NLST, with two types of simulated cancerous lung nodules. Each virtual patient in the cohort was assessed using simulated CT and CXR systems to generate images reflecting the NLST imaging technologies. Deep learning models trained for lesion detection, AI CT-Reader, and AI CXR-Reader served as virtual readers. Main Outcomes and Measures: The primary outcome was the difference in the Receiver Operating Characteristic Area Under the Curve (AUC) for CT and CXR modalities. Results: The study analyzed paired CT and CXR simulated images from 294 virtual patients. The AI CT-Reader outperformed the AI CXR-Reader across all levels of analysis. At the patient level, CT demonstrated superior diagnostic performance with an AUC of 0.92 (95% CI: 0.90 to 0.95), compared to CXR AUC of 0.72 (0.67 to 0.77). Subgroup analyses of lesion types revealed CT had significantly better detection of homogeneous lesions (AUC 0.97, 95% CI: 0.95 to 0.98) compared to heterogeneous lesions(0.89; 0.86 to 0.93). Furthermore, when the specificity of the AI CT-Reader was adjusted to match the NLST sensitivity of 94% for CT, the VLST results closely mirrored the NLST findings, further highlighting the alignment between the two studies. Conclusion and Relevance: The VIT results closely replicated those of the earlier NLST, underscoring its potential to replicate real clinical imaging trials. Integration of virtual trials may aid in the evaluation and improvement of imaging-based diagnosis., Comment: 19 pages, 4 figures, 1 table

The first public 0.9-4.4{\mu}m NIRCam images of the North Ecliptic Pole (NEP) Time Domain Field (TDF) uncovered galaxies displaying point-source features in their cores as seen in the longer wavelength filters. We visually identified a sample of 66 galaxies (~1 galaxy per arcmin2) with point-like cores and have modeled their two-dimensional light profiles with GalFit, identifying 16 galactic nuclei with measurable point-source components. GalFit suggests the visual sample is a mix of both compact stellar bulge and point-source galaxy cores. This core classification is complemented by spectral energy distribution (SED) modeling to infer the sample's active galactic nucleus (AGN) and host-galaxy parameters. For galaxies with measurable point-source components, the median fractional AGN contribution to their 0.1-30.0{\mu}m flux is 0.44, and 14/16 are color-classified AGN. We conclude that near-infrared point-source galaxy cores are signatures of AGN. In addition, we define an automated sample-selection criterion to identify these point-source features. These criteria can be used in other extant and future NIRCam images to streamline the search for galaxies with unresolved IR-luminous AGN. The James Webb Space Telescope's superb angular resolution and sensitivity at infrared wavelengths is resurrecting the morphological identification of AGN., Comment: 17 pages, 9 figures, 2 tables

Transit agencies that operate battery-electric buses must carefully manage fast-charging infrastructure to extend daily bus range without degrading on-time performance. To support this need, we propose a mixed-integer linear programming model to schedule opportunity charging that minimizes the amount of departure delay in all trips served by electric buses. Our novel approach directly tracks queuing at chargers in order to set and propagate departure delays. Allowing but minimizing delays makes it possible to optimize performance when delays due to traffic conditions and charging needs are inevitable, in contrast with existing methods that require charging to occur during scheduled layover time. To solve the model, we develop two algorithms based on decomposition. The first is an exact solution method based on Combinatorial Benders (CB) decomposition, which avoids directly enumerating the model's logic-based "big M" constraints and their inevitable computational challenges. The second, inspired by the CB approach but more efficient, is a polynomial-time heuristic based on linear programming that we call 3S. Computational experiments on both a simple notional transit network and the real bus system of King County, Washington, USA demonstrate the performance of both methods. The 3S method appears particularly promising for creating good charging schedules quickly at real-world scale.

We present a nation-wide network analysis of non-fatal opioid-involved overdose journeys in the United States. Leveraging a unique proprietary dataset of Emergency Medical Services incidents, we construct a journey-to-overdose geospatial network capturing nearly half a million opioid-involved overdose events spanning 2018-2023. We analyze the structure and sociological profiles of the nodes, which are counties or their equivalents, characterize the distribution of overdose journey lengths, and investigate changes in the journey network between 2018 and 2023. Our findings include that authority and hub nodes identified by the HITS algorithm tend to be located in urban areas and involved in overdose journeys with particularly long geographical distances.

Developing novel lead-free ferroelectric materials is crucial for next-generation microelectronic technologies that are energy efficient and environment friendly. However, materials discovery and property optimization are typically time-consuming due to the limited throughput of traditional synthesis methods. In this work, we use a high-throughput combinatorial synthesis approach to fabricate lead-free ferroelectric superlattices and solid solutions of (Ba0.7Ca0.3)TiO3 (BCT) and Ba(Zr0.2Ti0.8)O3 (BZT) phases with continuous variation of composition and layer thickness. High-resolution X-ray diffraction (XRD) and analytical scanning transmission electron microscopy (STEM) demonstrate high film quality and well-controlled compositional gradients. Ferroelectric and dielectric property measurements identify the optimal property point achieved at the morphotropic phase boundary (MPB) with a composition of 48BZT-52BCT. Displacement vector maps reveal that ferroelectric domain sizes are tunable by varying {BCT-BZT}N superlattice geometry. This high-throughput synthesis approach can be applied to many other material systems to expedite new materials discovery and properties optimization, allowing for the exploration of a large area of phase space within a single growth., Comment: submitted

UVCANDELS is a HST Cycle-26 Treasury Program awarded 164 orbits of primary ultraviolet (UV) F275W imaging and coordinated parallel optical F435W imaging in four CANDELS fields: GOODS-N, GOODS-S, EGS, and COSMOS, covering a total area of $\sim426$ arcmin$^2$. This is $\sim2.7$ times larger than the area covered by previous deep-field space UV data combined, reaching a depth of about 27 and 28 ABmag ($5\sigma$ in $0.2"$ apertures) for F275W and F435W, respectively. Along with the new photometric catalogs, we present an analysis of the rest-frame UV luminosity function (LF), relying on our UV-optimized aperture photometry method yielding a factor of $1.5\times$ increase than the H-isophot aperture photometry in the signal-to-noise ratios of galaxies in our F275W imaging. Using well tested photometric redshift measurements we identify 5810 galaxies at redshifts $0.6

We discover that many natural-language prompts can be replaced by corresponding prompts that are unintelligible to humans but that provably elicit similar behavior in language models. We call these prompts "evil twins" because they are obfuscated and uninterpretable (evil), but at the same time mimic the functionality of the original natural-language prompts (twins). Remarkably, evil twins transfer between models. We find these prompts by solving a maximum-likelihood problem which has applications of independent interest., Comment: EMNLP 2024 Main, camera-ready

Transportation distance information is a powerful resource, but location records are often censored due to privacy concerns or regulatory mandates. We outline methods to approximate, sample from, and compare distributions of distances between censored location pairs, a task with applications to public health informatics, logistics, and more. We validate empirically via simulation and demonstrate applicability to practical geospatial data analysis tasks.

This document presents a summary of the 2023 Terrestrial Very-Long-Baseline Atom Interferometry Workshop hosted by CERN. The workshop brought together experts from around the world to discuss the exciting developments in large-scale atom interferometer (AI) prototypes and their potential for detecting ultralight dark matter and gravitational waves. The primary objective of the workshop was to lay the groundwork for an international TVLBAI proto-collaboration. This collaboration aims to unite researchers from different institutions to strategize and secure funding for terrestrial large-scale AI projects. The ultimate goal is to create a roadmap detailing the design and technology choices for one or more km-scale detectors, which will be operational in the mid-2030s. The key sections of this report present the physics case and technical challenges, together with a comprehensive overview of the discussions at the workshop together with the main conclusions., Comment: Summary of the Terrestrial Very-Long-Baseline Atom Interferometry Workshop held at CERN: https://indico.cern.ch/event/1208783/

We present AstroCLIP, a single, versatile model that can embed both galaxy images and spectra into a shared, physically meaningful latent space. These embeddings can then be used - without any model fine-tuning - for a variety of downstream tasks including (1) accurate in-modality and cross-modality semantic similarity search, (2) photometric redshift estimation, (3) galaxy property estimation from both images and spectra, and (4) morphology classification. Our approach to implementing AstroCLIP consists of two parts. First, we embed galaxy images and spectra separately by pretraining separate transformer-based image and spectrum encoders in self-supervised settings. We then align the encoders using a contrastive loss. We apply our method to spectra from the Dark Energy Spectroscopic Instrument and images from its corresponding Legacy Imaging Survey. Overall, we find remarkable performance on all downstream tasks, even relative to supervised baselines. For example, for a task like photometric redshift prediction, we find similar performance to a specifically-trained ResNet18, and for additional tasks like physical property estimation (stellar mass, age, metallicity, and sSFR), we beat this supervised baseline by 19\% in terms of $R^2$. We also compare our results to a state-of-the-art self-supervised single-modal model for galaxy images, and find that our approach outperforms this benchmark by roughly a factor of two on photometric redshift estimation and physical property prediction in terms of $R^2$, while remaining roughly in-line in terms of morphology classification. Ultimately, our approach represents the first cross-modal self-supervised model for galaxies, and the first self-supervised transformer-based architectures for galaxy images and spectra., Comment: 18 pages, accepted in Monthly Notices of the Royal Astronomical Society, Presented at the NeurIPS 2023 AI4Science Workshop

We introduce multiple physics pretraining (MPP), an autoregressive task-agnostic pretraining approach for physical surrogate modeling. MPP involves training large surrogate models to predict the dynamics of multiple heterogeneous physical systems simultaneously by learning features that are broadly useful across diverse physical tasks. In order to learn effectively in this setting, we introduce a shared embedding and normalization strategy that projects the fields of multiple systems into a single shared embedding space. We validate the efficacy of our approach on both pretraining and downstream tasks over a broad fluid mechanics-oriented benchmark. We show that a single MPP-pretrained transformer is able to match or outperform task-specific baselines on all pretraining sub-tasks without the need for finetuning. For downstream tasks, we demonstrate that finetuning MPP-trained models results in more accurate predictions across multiple time-steps on new physics compared to training from scratch or finetuning pretrained video foundation models. We open-source our code and model weights trained at multiple scales for reproducibility and community experimentation.

Large Language Models have not yet been broadly adapted for the analysis of scientific datasets due in part to the unique difficulties of tokenizing numbers. We propose xVal, a numerical encoding scheme that represents any real number using just a single token. xVal represents a given real number by scaling a dedicated embedding vector by the number value. Combined with a modified number-inference approach, this strategy renders the model end-to-end continuous when considered as a map from the numbers of the input string to those of the output string. This leads to an inductive bias that is generally more suitable for applications in scientific domains. We empirically evaluate our proposal on a number of synthetic and real-world datasets. Compared with existing number encoding schemes, we find that xVal is more token-efficient and demonstrates improved generalization., Comment: 10 pages 7 figures. Supplementary: 5 pages 2 figures