Search

Recent experiments have demonstrated that vibrational strong coupling (VSC) between molecular vibrations and the optical cavity field can modify vibrational energy transfer (VET) processes in molecular systems. However, the underlying mechanisms and the behavior of individual molecules under collective VSC remain largely incomplete. In this work, we combine state-of-the-art quantum vibrational spectral calculation, quantum wavepacket dynamics simulations, and ab initio machine-learning potential to elucidate how the vibrational dynamics of water OH stretches can be altered by VSC. Taking the (H$_2$O)$_{21}$-cavity system as an example, we show that the collective VSC breaks the localization picture, promotes the delocalization of OH stretches, and opens new intermolecular vibrational energy pathways involving both neighboring and remote water molecules. The manipulation of the VET process relies on the alignment of the transition dipole moment orientations of the corresponding vibrational states. The emergence of new energy transfer pathways is found to be attributed to cavity-induced vibrational resonance involving OH stretches across different water molecules, along with alterations in mode coupling patterns. Our fully quantum theoretical calculations not only confirm and extend previous findings on cavity-modified energy transfer processes but also provide new insights in energy transfer processes under collective VSC.

In this work, we present a case study in implementing a variational quantum algorithm for solving the Poisson equation, which is a commonly encountered partial differential equation in science and engineering. We highlight the practical challenges encountered in mapping the algorithm to physical hardware, and the software engineering considerations needed to achieve realistic results on today's non-fault-tolerant systems., Comment: 8 pages, 6 figures, submitted on 2024-10-10 to IEEE Quantum Software and its Engineering

Model-based reinforcement learning refers to a set of approaches capable of sample-efficient decision making, which create an explicit model of the environment. This model can subsequently be used for learning optimal policies. In this paper, we propose a temporal Gaussian Mixture Model composed of a perception model and a transition model. The perception model extracts discrete (latent) states from continuous observations using a variational Gaussian mixture likelihood. Importantly, our model constantly monitors the collected data searching for new Gaussian components, i.e., the perception model performs a form of structure learning (Smith et al., 2020; Friston et al., 2018; Neacsu et al., 2022) as it learns the number of Gaussian components in the mixture. Additionally, the transition model learns the temporal transition between consecutive time steps by taking advantage of the Dirichlet-categorical conjugacy. Both the perception and transition models are able to forget part of the data points, while integrating the information they provide within the prior, which ensure fast variational inference. Finally, decision making is performed with a variant of Q-learning which is able to learn Q-values from beliefs over states. Empirically, we have demonstrated the model's ability to learn the structure of several mazes: the model discovered the number of states and the transition probabilities between these states. Moreover, using its learned Q-values, the agent was able to successfully navigate from the starting position to the maze's exit.

The precise characterization and mitigation of systematic effects is one of the biggest roadblocks impeding the detection of the fluctuations of cosmological 21cm signals. Missing data in radio cosmological experiments, often due to radio frequency interference (RFI), poses a particular challenge to power spectrum analysis as it could lead to the ringing of bright foreground modes in Fourier space, heavily contaminating the cosmological signals. Here we show that the problem of missing data becomes even more arduous in the presence of systematic effects. Using a realistic numerical simulation, we demonstrate that partially flagged data combined with systematic effects can introduce significant foreground ringing. We show that such an effect can be mitigated through inpainting the missing data. We present a rigorous statistical framework that incorporates the process of inpainting missing data into a quadratic estimator of the 21cm power spectrum. Under this framework, the uncertainties associated with our inpainting method and its impact on power spectrum statistics can be understood. These results are applied to the latest Phase II observations taken by the Hydrogen Epoch of Reionization Array, forming a crucial component in power spectrum analyses as we move toward detecting 21cm signals in the ever more noisy RFI environment., Comment: 25 pages, 11 figures, submitted to ApJ

Accurate modeling of terrestrial carbon and water exchange requires robust ecological parameters that capture vegetation responses and adaptations to the local environment. The current generation of land models use Plant Functional Types (PFTs) to discretize vegetation functional diversity, but these coarse categorizations often overlook fine-scale variations shaped by local climate, soil, and forest age factors. The lack of governing equations for plant adaptation demands a paradigm shift in how we integrate diverse Earth observations to uncover ecological functional dependence on changing environments. To address this challenge, we developed DifferLand, a differentiable, hybrid physics and machine learning model that infers the spatial distributions of ecological parameters and their relationships with environmental factors constrained by satellite and in-situ observations. Our model unifies top-down and bottom-up observational constraints with process-based knowledge to generate a global analysis of ecological functions and their adaptation to environmental gradients. We found PFTs account for less than half of the explainable spatial parameter variations controlling carbon fluxes and vegetation states. The remaining parameter variability is largely driven by local climate and forest demography factors, and the learned environment-parameter relationships lead to enhanced spatial generalization at unseen locations. DifferLand identified growing season length, leaf economics, and agricultural intensity as the three orthogonal spatial gradients underlying parameter variations. Our novel framework can lead to new insights on global carbon cycling by learning directly from data and expanding our understanding of local responses of ecosystems to environmental drivers.

The rovibrational level populations, and subsequent emission in various astrophysical environments, is driven by inelastic collision processes. The available rovibrational rate coefficients for water have been calculated using a number of approximations. We present a numerically exact calculation for the rovibrational quenching for all water vibrational modes due to collisions with atomic hydrogen. The scattering theory implements a quantum close-coupling (CC) method on a high level ab initio six-dimensional (6D) potential energy surface (PES). Total rovibrational quenching cross sections for excited bending levels were compared with earlier results on a 4D PES with the rigid-bender close-coupling (RBCC) approximation. General agreement between 6D-CC and 4D-RBCC calculations are found, but differences are evident including the energy and amplitude of low-energy orbiting resonances. Quenching cross sections from the symmetric and asymmetric stretch modes are provided for the first time. The current 6D-CC calculation provides accurate inelastic data needed for astrophysical modeling., Comment: 20 pages, 5 figures

By converting the entire 3D space around the user into a screen, Extended Reality (XR) can ameliorate traditional displays' space limitations and facilitate the consumption of multiple pieces of information at a time. However, if designed inappropriately, these XR interfaces can overwhelm the user and complicate information access. In this work, we explored the design dimensions that can be adapted to enable suitable presentation and interaction within an XR interface. To investigate a specific use case of context-aware adaptations within our proposed design space, we concentrated on the spatial layout of the XR content and investigated non-adaptive and adaptive placement strategies. In this paper, we (1) present a comprehensive design space for XR interfaces, (2) propose Environment-referenced, an adaptive placement strategy that uses a relevant intermediary from the environment within a Hybrid Frame of Reference (FoR) for each XR object, and (3) evaluate the effectiveness of this adaptive placement strategy and a non-adaptive Body-Fixed placement strategy in four contextual scenarios varying in terms of social setting and user mobility in the environment. The performance of these placement strategies from our within-subjects user study emphasized the importance of intermediaries' relevance to the user's focus. These findings underscore the importance of context-aware interfaces, indicating that the appropriate use of an adaptive content placement strategy in a context can significantly improve task efficiency, accuracy, and usability.

Recent advancements in Augmented Reality (AR) research have highlighted the critical role of context awareness in enhancing interface effectiveness and user experience. This underscores the need for intelligent AR (iAR) interfaces that dynamically adapt across various contexts to provide optimal experiences. In this paper, we (a) propose a comprehensive framework for context-aware inference and adaptation in iAR, (b) introduce a taxonomy that describes context through quantifiable input data, and (c) present an architecture that outlines the implementation of our proposed framework and taxonomy within iAR. Additionally, we present an empirical AR experiment to observe user behavior and record user performance, context, and user-specified adaptations to the AR interfaces within a context-switching scenario. We (d) explore the nuanced relationships between context and user adaptations in this scenario and discuss the significance of our framework in identifying these patterns. This experiment emphasizes the significance of context-awareness in iAR and provides a preliminary training dataset for this specific Scenario.

Sufficiently capable models could subvert human oversight and decision-making in important contexts. For example, in the context of AI development, models could covertly sabotage efforts to evaluate their own dangerous capabilities, to monitor their behavior, or to make decisions about their deployment. We refer to this family of abilities as sabotage capabilities. We develop a set of related threat models and evaluations. These evaluations are designed to provide evidence that a given model, operating under a given set of mitigations, could not successfully sabotage a frontier model developer or other large organization's activities in any of these ways. We demonstrate these evaluations on Anthropic's Claude 3 Opus and Claude 3.5 Sonnet models. Our results suggest that for these models, minimal mitigations are currently sufficient to address sabotage risks, but that more realistic evaluations and stronger mitigations seem likely to be necessary soon as capabilities improve. We also survey related evaluations we tried and abandoned. Finally, we discuss the advantages of mitigation-aware capability evaluations, and of simulating large-scale deployments using small-scale statistics.

While the pretraining of Foundation Models (FMs) for remote sensing (RS) imagery is on the rise, models remain restricted to a few hundred million parameters. Scaling models to billions of parameters has been shown to yield unprecedented benefits including emergent abilities, but requires data scaling and computing resources typically not available outside industry R&D labs. In this work, we pair high-performance computing resources including Frontier supercomputer, America's first exascale system, and high-resolution optical RS data to pretrain billion-scale FMs. Our study assesses performance of different pretrained variants of vision Transformers across image classification, semantic segmentation and object detection benchmarks, which highlight the importance of data scaling for effective model scaling. Moreover, we discuss construction of a novel TIU pretraining dataset, model initialization, with data and pretrained models intended for public release. By discussing technical challenges and details often lacking in the related literature, this work is intended to offer best practices to the geospatial community toward efficient training and benchmarking of larger FMs.

Preclinical diffusion MRI (dMRI) has proven value in methods development and validation, characterizing the biological basis of diffusion phenomena, and comparative anatomy. While dMRI enables in vivo non-invasive characterization of tissue, ex vivo dMRI is increasingly being used to probe tissue microstructure and brain connectivity. Ex vivo dMRI has several experimental advantages that facilitate high spatial resolution and high signal-to-noise ratio (SNR) images, cutting-edge diffusion contrasts, and direct comparison with histological data as a methodological validation. However, there are a number of considerations that must be made when performing ex vivo experiments. The steps from tissue preparation, image acquisition and processing, and interpretation of results are complex, with many decisions that not only differ dramatically from in vivo imaging of small animals, but ultimately affect what questions can be answered using the data. This work concludes a 3-part series of recommendations and considerations for preclinical dMRI. Herein, we describe best practices for dMRI of ex vivo tissue, with a focus on image pre-processing, data processing and model fitting, and tractography. In each section, we attempt to provide guidelines and recommendations, but also highlight areas for which no guidelines exist (and why), and where future work should lie. We end by providing guidelines on code sharing and data sharing, and point towards open-source software and databases specific to small animal and ex vivo imaging., Comment: Part 3 of 3 in "Considerations and recommendations for preclinical diffusion MRI"

Recent advances in multi-modal, highly capable LLMs have enabled voice-enabled AI agents. These agents are enabling new applications, such as voice-enabled autonomous customer service. However, with all AI capabilities, these new capabilities have the potential for dual use. In this work, we show that voice-enabled AI agents can perform the actions necessary to perform common scams. To do so, we select a list of common scams collected by the government and construct voice-enabled agents with directions to perform these scams. We conduct experiments on our voice-enabled agents and show that they can indeed perform the actions necessary to autonomously perform such scams. Our results raise questions around the widespread deployment of voice-enabled AI agents.

Massive main-sequence stars have convective cores and radiative envelopes, but also sub-surface convection zones caused by partial ionisation. However, the convective properties depend on opacity and a star's metallicity. Non-rotating 1D evolution models of main-sequence stars with the metallicity of the SMC suggest tenuous sub-surface convection zones using the Rayleigh number as a criterion for convection owing to their lower metallicity. We test whether massive stars of different metallicities both inside and outside of asteroseismically calibrated stability windows for sub-surface convection exhibit different properties in stochastic low-frequency (SLF) variability. We extracted customised light curves from the TESS mission for a sample of massive stars using an effective point spread function (ePSF) method, and compared their morphologies in terms of characteristic frequency and amplitude using a Gaussian process (GP) regression methodology. We demonstrate that the properties of SLF variability are generally consistent across the metallicity range from the Milky Way down to the SMC, for stars both inside and outside of the sub-surface stability windows. We conclude that non-rotating 1D stellar structure models cannot alone be used to explain SLF variability in light curves of massive stars. The similar properties of SLF variability across a range of metallicity values, which follow the same trends in mass and age in the HR diagram at both high and low metallicity, support a transition in the dominant mechanism causing SLF variability from younger to more evolved stars. Specifically, core-excited internal gravity waves (IGWs) are favoured for younger stars lacking sub-surface convection zones, especially at low metallicity, and sub-surface convection zones are favoured for more evolved massive stars. (abstract abridged for arXiv), Comment: Accepted and in press, A&A (https://doi.org/10.1051/0004-6361/202451419)

The propagation of wave fields and their interactions with matter are important for established and emerging fields in optical sciences. Efficient methods for predicting such behaviour have been employed routinely for coherent sources. However, most real world optical systems exhibit partial coherence, for which the present mathematical description involves high dimensional complex functions and hence poses challenges for numerical implementations. This demands significant computational resources to determine the properties of partially coherent wavefields. Here, we describe the novel Phase-Space (PS) propagator, an efficient and self-consistent technique for free space propagation of wave fields which are partially coherent in the spatial domain. The PS propagator makes use of the fact that the propagation of a wave field in free space is equivalent to a shearing of the corresponding PSD function. Computationally, this approach is simpler and the need for using different propagation methods for near and far-field regions is removed., Comment: 17 pages, 4 figures, to be submitted to Optics Communications For the associated Python package, see https://github.com/jakerogers-1/Phase-Space-Propagator

Asteroseismology is the study of the interior physics and structure of stars using their pulsations. It is applicable to stars across the Hertzsprung-Russell (HR) diagram and a powerful technique to measure masses, radii and ages, but also directly constrain interior rotation, chemical mixing, and magnetism. This is because a star's self-excited pulsation modes are sensitive to its structure. Asteroseismology generally requires long-duration and high-precision time series data. The method of forward asteroseismic modelling, which is the statistical comparison of observed pulsation mode frequencies to theoretically predicted pulsation frequencies calculated from a grid of models, provides precise constraints for calibrating various transport phenomena. In this introduction to asteroseismology, we provide an overview of its principles, and the typical data sets and methodologies used to constrain stellar interiors. Finally, we present key highlights of asteroseismic results from across the HR diagram, and conclude with ongoing challenges and future prospects for this ever-expanding field within stellar astrophysics., Comment: 21 pages; This is a pre-print of an invited chapter for the Encyclopedia of Astrophysics (edited by I. Mandel, section editor F.R.N. Schneider) to be published by Elsevier as a Reference Module

2D electron density profiles obtained from coherence imaging spectroscopy in different MAST-U divertor conditions are compared. The data includes variations of strike point position, core electron density, and heating power. The improved performance of the long-legged divertors results in a lower electron density and particle flux at the target compared to configurations with smaller strike point major radius, while also being characterized by lower temperatures and deeper detachment. Comparisons against SOLPS simulations generally show good agreement in profile shape along and across the separatrix. The peaking of the electron density downstream of the detachment front is associated with significant neutral drag acting on the plasma flow.

Language models (LMs) can produce errors that are hard to detect for humans, especially when the task is complex. RLHF, the most popular post-training method, may exacerbate this problem: to achieve higher rewards, LMs might get better at convincing humans that they are right even when they are wrong. We study this phenomenon under a standard RLHF pipeline, calling it "U-SOPHISTRY" since it is Unintended by model developers. Specifically, we ask time-constrained (e.g., 3-10 minutes) human subjects to evaluate the correctness of model outputs and calculate humans' accuracy against gold labels. On a question-answering task (QuALITY) and programming task (APPS), RLHF makes LMs better at convincing our subjects but not at completing the task correctly. RLHF also makes the model harder to evaluate: our subjects' false positive rate increases by 24.1% on QuALITY and 18.3% on APPS. Finally, we show that probing, a state-of-the-art approach for detecting Intended Sophistry (e.g. backdoored LMs), does not generalize to U-SOPHISTRY. Our results highlight an important failure mode of RLHF and call for more research in assisting humans to align them.

The provisions in the Every Student Succeeds Act (ESSA) reauthorization of the Elementary and Secondary Education Act (ESEA) include (1) the requirement for states to periodically review resource allocation in local education agencies (LEAs) serving a significant number of schools identified for comprehensive support and improvement (CSI), targeted support and improvement (TSI), and additional targeted support and improvement (ATSI) and (2) the requirement that CSI and ATSI plans both identify and address local resource inequities. ESSA requires states to periodically review resource allocation to support school improvement in schools that qualify for comprehensive and targeted support. In 2019, the Resource Allocation Review Community of Practice (RAR CoP) of the United States Department of Education (the Department) provided a useful foundation. The RAR CoP calls upon state agencies to connect the additional resource inequity provisions outlined in ESSA and align resource allocation reviews with the comprehensive school improvement. Building on the guidance offered by the RAR CoP, Utah, with the support of WestEd's Region 15 Comprehensive Center (R15CC), was among the first states to design and implement an RAR process. This report shares some discoveries and lessons learned during the design and implementation process.

This document provides state education agency (SEA) leaders with a systematized approach to identifying and defining elements, major tasks, and deliverables for conducting a meaningful resource allocation review (RAR). Based on the findings of an intensive four-year design process, field research, and recommendations from the pilot audit of the RAR process conducted by the U.S. Department of Education, this document: (1) offers education leaders at the state and local level with concrete examples of resource inequities; (2) presents considerations for decision-making around resources, and (3) proposes ways to reduce the burden on LEAs and schools by streamlining the RAR completion process.

This report summarizes an investigation of literature published in 1985-2023 on universal design (UD) of large-scale assessments. State education agencies, K-12 assessment vendors, teacher trainers, and classroom teachers are increasingly applying the principles of UD to the instruction and assessment of students from special populations. In theory, UD ensures that instruction and assessment are created from the beginning to be accessible to the widest population of students possible. Yet, it is unclear how these UD frameworks and their associated principles and elements are being applied, or even which UD framework is used. This obscurity and lack of defined UD principles in the research literature creates obstacles in research replication and in identification of evidence-based practices. This scoping review identifies how the concept of UD has been applied to the broad range of U.S. district and state large-scale academic assessments. We included literature on the application of UD to the broad range of K-12 district and state large-scale academic assessments in the U.S.

Although access to sufficient screen space is crucial to knowledge work, workers often find themselves with limited access to display infrastructure in remote or public settings. While virtual displays can be used to extend the available screen space through extended reality (XR) head-worn displays (HWD), we must better understand the implications of working with them in public settings from both users' and bystanders' viewpoints. To this end, we conducted two user studies. We first explored the usage of a hybrid AR display across real-world settings and tasks. We focused on how users take advantage of virtual displays and what social and environmental factors impact their usage of the system. A second study investigated the differences between working with a laptop, an AR system, or a VR system in public. We focused on a single location and participants performed a predefined task to enable direct comparisons between the conditions while also gathering data from bystanders. The combined results suggest a positive acceptance of XR technology in public settings and show that virtual displays can be used to accompany existing devices. We highlighted some environmental and social factors. We saw that previous XR experience and personality can influence how people perceive the use of XR in public. In addition, we confirmed that using XR in public still makes users stand out and that bystanders are curious about the devices, yet have no clear understanding of how they can be used., Comment: arXiv admin note: substantial text overlap with arXiv:2310.09786

This short paper reports a study of the electronic dissociation energies, De, of water clusters from direct ab initio (mostly CCSD(T)) calculations and the q-AQUA and MB-pol potentials. These clusters range in size from 6-25 monomers. These are all in very good agreement with each other, as shown in a recent Perspective by Herman and Xantheas. To the best of our knowledge, we present for the first time results for the De per monomer. To our surprise this quantity appears to be converging to a value close to 12 kcal/mol. An estimate of 1.5 - 2 kcal/mol for the {\Delta}ZPE for these clusters puts the value of D0 at 10 to 10.5 kcal/mol. This value is remarkably (and probably fortuitously) close to the reported sublimation enthalpy of 10.2 kcal/mol at 10 K. However, given that these De energies correspond to dissociation of the cluster to N isolated monomers the interpretation of ``vaporization" of these ``solid" clusters is qualitatively reasonable., Comment: 7 pages 2 figures. Working draft

In this study, we investigate in situ etching of \b{eta}-Ga2O3 in a metal-organic chemical vapor deposition (MOCVD) system using tert-Butyl chloride (TBCl). We report the successful etching of both heteroepitaxial (-201)-oriented and homoepitaxial (010)-oriented \b{eta}-Ga2O3 films over a wide range of substrate temperature, TBCl molar flows, and reactor pressures. We identify that the likely etchant is HCl (g) formed by the pyrolysis of TBCl in the hydrodynamic boundary layer above the substrate. The temperature dependence of the etch rate reveals two distinct regimes characterized by markedly different apparent activation energies. The extracted apparent activation energies suggest that at temperatures below ~800 {\deg}C the etch rate is likely limited by desorption of etch products, while at higher substrate temperatures chemisorption of HCl limits the etch rate. The relative etch rates of heteroepitaxial (-201) and homoepitaxial (010) \b{eta}-Ga2O3 were observed to scale by the ratio of the surface energies indicating an anisotropic etch. For (010) homoepitaxial films, relatively smooth post-etch surface morphology was achieved by tuning the etching parameters., Comment: 12 pages, 8 figures

Dexterous telemanipulation is crucial in advancing human-robot systems, especially in tasks requiring precise and safe manipulation. However, it faces significant challenges due to the physical differences between human and robotic hands, the dynamic interaction with objects, and the indirect control and perception of the remote environment. Current approaches predominantly focus on mapping the human hand onto robotic counterparts to replicate motions, which exhibits a critical oversight: it often neglects the physical interaction with objects and relegates the interaction burden to the human to adapt and make laborious adjustments in response to the indirect and counter-intuitive observation of the remote environment. This work develops an End-Effects-Oriented Learning-based Dexterous Telemanipulation (EFOLD) framework to address telemanipulation tasks. EFOLD models telemanipulation as a Markov Game, introducing multiple end-effect features to interpret the human operator's commands during interaction with objects. These features are used by a Deep Reinforcement Learning policy to control the robot and reproduce such end effects. EFOLD was evaluated with real human subjects and two end-effect extraction methods for controlling a virtual Shadow Robot Hand in telemanipulation tasks. EFOLD achieved real-time control capability with low command following latency (delay<0.11s) and highly accurate tracking (MSE<0.084 rad)., Comment: Accepted by IROS 2024

The 21 cm transition from neutral Hydrogen promises to be the best observational probe of the Epoch of Reionisation (EoR). This has led to the construction of low-frequency radio interferometric arrays, such as the Hydrogen Epoch of Reionization Array (HERA), aimed at systematically mapping this emission for the first time. Precision calibration, however, is a requirement in 21 cm radio observations. Due to the spatial compactness of HERA, the array is prone to the effects of mutual coupling, which inevitably lead to non-smooth calibration errors that contaminate the data. When unsmooth gains are used in calibration, intrinsically spectrally-smooth foreground emission begins to contaminate the data in a way that can prohibit a clean detection of the cosmological EoR signal. In this paper, we show that the effects of mutual coupling on calibration quality can be reduced by applying custom time-domain filters to the data prior to calibration. We find that more robust calibration solutions are derived when filtering in this way, which reduces the observed foreground power leakage. Specifically, we find a reduction of foreground power leakage by 2 orders of magnitude at k=0.5.

Progress in machine learning has facilitated the development of potentials that offer both the accuracy of first-principles techniques and vast increases in the speed of evaluation. Recently,"$\Delta$-machine learning" has been used to elevate the quality of a potential energy surface (PES) based on low-level, e.g., density functional theory (DFT) energies and gradients to close to the gold-standard coupled cluster level of accuracy. We have demonstrated the success of this approach for molecules, ranging in size from H$_3$O$^+$ to 15-atom acetyl-acetone and tropolone. These were all done using the B3LYP functional. Here we investigate the generality of this approach for the PBE, M06, M06-2X, and PBE0+MBD functionals, using ethanol as the example molecule. Linear regression with permutationally invariant polynomials is used to fit both low-level and correction PESs. These PESs are employed for standard RMSE analysis for training and test datasets, and then general fidelity tests such as energetics of stationary points, normal mode frequencies, and torsional potentials are examined. We achieve similar improvements in all cases. Interestingly, we obtained significant improvement over DFT gradients where coupled cluster gradients were not used to correct the low-level PES. Finally, we present some results for correcting a recent molecular mechanics force field for ethanol and comment on the possible generality of this approach.

This letter illustrates the opinion of the molecular dynamics (MD) community on the need to adopt a new FAIR paradigm for the use of molecular simulations. It highlights the necessity of a collaborative effort to create, establish, and sustain a database that allows findability, accessibility, interoperability, and reusability of molecular dynamics simulation data. Such a development would democratize the field and significantly improve the impact of MD simulations on life science research. This will transform our working paradigm, pushing the field to a new frontier. We invite you to support our initiative at the MDDB community (https://mddbr.eu/community/)

Surveys in the Milky Way and Large Magellanic Cloud revealed that the majority of massive stars will interact with companions during their lives. However, knowledge of the binary properties of massive stars at low metallicity, which approaches the conditions of the Early Universe, remains sparse. We present the Binarity at LOw Metallicity (BLOeM) campaign - an ESO large programme designed to obtain 25 epochs of spectroscopy for 929 massive stars in the SMC - the lowest metallicity conditions in which multiplicity is probed to date (Z = 0.2 Zsun). BLOeM will provide (i) the binary fraction, (ii) the orbital configurations of systems with periods P < 3 yr, (iii) dormant OB+BH binaries, and (iv) a legacy database of physical parameters of massive stars at low metallicity. The stars are observed with the LR02 setup of the giraffe instrument of the Very Large Telescope (3960-4570A, resolving power R=6200; typical signal-to-noise ratio S/N=70-100). This paper utilises the first 9 epochs obtained over a three-month time. We describe the survey and data reduction, perform a spectral classification of the stacked spectra, and construct a Hertzsprung-Russell diagram of the sample via spectral-type and photometric calibrations. The sample covers spectral types from O4 to F5, spanning the effective temperature and luminosity ranges 6.5

Exploring nonequilibrium hot carriers from plasmonic metal nanostructures is a dynamic field in optoelectronics, driving photochemical reactions such as solar fuel generation. The hot carrier injection mechanism and the reaction rate are highly impacted by the metal/molecule interaction. However, determining the primary type of the reaction and thus the injection mechanism of the hot carriers has remained elusive. In this work, we reveal an electron injection mechanism deviating from a purely outersphere process for the reduction of ferricyanide redox molecule in a gold/p-type gallium nitride (Au/p- GaN) photocathode system. Combining our experimental approach with ab initio simulations, we discover that the efficient inner-sphere transfer of low-energy electrons leads to a continuous enhancement in the photocathode device performance in the interband regime. These findings provide important mechanistic insights, showing our methodology as a powerful tool for analyzing and engineering hot-carrier-driven processes in plasmonic photocatalytic systems and optoelectronic devices.

In this paper, we explore a connection between type universes and memory allocation. Type universe hierarchies are used in dependent type theories to ensure consistency, by forbidding a type from quantifying over all types. Instead, the types of types (universes) form a hierarchy, and a type can only quantify over types in other universes (with some exceptions), restricting cyclic reasoning in proofs. We present a perspective where universes also describe \emph{where} values are allocated in the heap, and the choice of universe algebra imposes a structure on the heap overall. The resulting type system provides a simple declarative system for reasoning about and restricting memory allocation, without reasoning about reads or writes. We present a theoretical framework for equipping a type system with higher-order references restricted by a universe hierarchy, and conjecture that many existing universe algebras give rise to interesting systems for reasoning about allocation. We present 3 instantiations of this approach to enable reasoning about allocation in the simply typed $\lambda$-calculus: (1) the standard ramified universe hierarchy, which we prove guarantees termination of the language extended with higher-order references by restricting cycles in the heap; (2) an extension with an \emph{impredicative} base universe, which we conjecture enables full-ground references (with terminating computation but cyclic ground data structures); (3) an extension with \emph{universe polymorphism}, which divides the heap into fine-grained regions., Comment: Submitted to ESOP 2025

We describe a family of architectures to support transductive inference by allowing memory to grow to a finite but a-priori unknown bound while making efficient use of finite resources for inference. Current architectures use such resources to represent data either eidetically over a finite span ("context" in Transformers), or fading over an infinite span (in State Space Models, or SSMs). Recent hybrid architectures have combined eidetic and fading memory, but with limitations that do not allow the designer or the learning process to seamlessly modulate the two, nor to extend the eidetic memory span. We leverage ideas from Stochastic Realization Theory to develop a class of models called B'MOJO to seamlessly combine eidetic and fading memory within an elementary composable module. The overall architecture can be used to implement models that can access short-term eidetic memory "in-context," permanent structural memory "in-weights," fading memory "in-state," and long-term eidetic memory "in-storage" by natively incorporating retrieval from an asynchronously updated memory. We show that Transformers, existing SSMs such as Mamba, and hybrid architectures such as Jamba are special cases of B'MOJO and describe a basic implementation, to be open sourced, that can be stacked and scaled efficiently in hardware. We test B'MOJO on transductive inference tasks, such as associative recall, where it outperforms existing SSMs and Hybrid models; as a baseline, we test ordinary language modeling where B'MOJO achieves perplexity comparable to similarly-sized Transformers and SSMs up to 1.4B parameters, while being up to 10% faster to train. Finally, we show that B'MOJO's ability to modulate eidetic and fading memory results in better inference on longer sequences tested up to 32K tokens, four-fold the length of the longest sequences seen during training.

The Dragonfly Telephoto Array employs a unique design to detect very large and diffuse galaxies, which might be missed with conventional telescopes. The Dragonfly Ultrawide Survey (DFUWS) is a new wide-field survey which will cover 10,000 deg$^2$ of the northern sky, and it provides an ideal dataset to find these large diffuse galaxies. From 3100 deg$^2$ of DFUWS data, we identified eleven large, low surface brightness galaxies as a pilot sample for spectroscopic follow-up. These are the largest galaxies in the examined area that appear smooth and isolated, with effective radii of 12"-27". Eight are below 24 $\mathrm{mag\,arcsec^{-2}}$ in central $g$-band surface brightness. Keck Cosmic Web Imager (KCWI) spectra of the diffuse light show that all eleven galaxies in this sample are quiescent, and seven qualify as ultra-diffuse galaxies (UDGs). Eight galaxies have distances between 15 and 30 Mpc, while the other three are in the Pegasus cluster at 50 Mpc. Their spectra show evidence of a $\sim 1$Gyr old stellar population in addition to an even older stellar population. The intermediate-age component is present in group and satellite galaxies but not in the Pegasus cluster UDGs. All galaxies in this sample are detected in both Dragonfly and Legacy imaging, and the sample partially overlaps with existing UDG catalogs. This pilot sample provides an excellent training set for our analysis of the upcoming full 10,000 deg$^2$ DFUWS data, from which we may expect to discover even larger, previously-unknown galaxies., Comment: Submitted to ApJ

Language models are capable of iteratively improving their outputs based on natural language feedback, thus enabling in-context optimization of user preference. In place of human users, a second language model can be used as an evaluator, providing feedback along with numerical ratings which the generator attempts to optimize. However, because the evaluator is an imperfect proxy of user preference, this optimization can lead to reward hacking, where the evaluator's ratings improve while the generation quality remains stagnant or even decreases as judged by actual user preference. The concern of reward hacking is heightened in iterative self-refinement where the generator and the evaluator use the same underlying language model, in which case the optimization pressure can drive them to exploit shared vulnerabilities. Using an essay editing task, we show that iterative self-refinement leads to deviation between the language model evaluator and human judgment, demonstrating that reward hacking can occur spontaneously in-context with the use of iterative self-refinement. In addition, we study conditions under which reward hacking occurs and observe two factors that affect reward hacking severity: model size and context sharing between the generator and the evaluator.

Backdoors are hidden behaviors that are only triggered once an AI system has been deployed. Bad actors looking to create successful backdoors must design them to avoid activation during training and evaluation. Since data used in these stages often only contains information about events that have already occurred, a component of a simple backdoor trigger could be a model recognizing data that is in the future relative to when it was trained. Through prompting experiments and by probing internal activations, we show that current large language models (LLMs) can distinguish past from future events, with probes on model activations achieving 90% accuracy. We train models with backdoors triggered by a temporal distributional shift; they activate when the model is exposed to news headlines beyond their training cut-off dates. Fine-tuning on helpful, harmless and honest (HHH) data does not work well for removing simpler backdoor triggers but is effective on our backdoored models, although this distinction is smaller for the larger-scale model we tested. We also find that an activation-steering vector representing a model's internal representation of the date influences the rate of backdoor activation. We take these results as initial evidence that, at least for models at the modest scale we test, standard safety measures are enough to remove these backdoors.

Sample Return Capsules (SRCs) entering Earth's atmosphere at hypervelocity from interplanetary space are a valuable resource for studying meteor phenomena. The 24 September 2023 arrival of the OSIRIS-REx (Origins, Spectral Interpretation, Resource Identification, and Security-Regolith Explorer) SRC provided an unprecedented chance for geophysical observations of a well-characterized source with known parameters, including timing and trajectory. A collaborative effort involving researchers from 16 institutions executed a carefully planned geophysical observational campaign at strategically chosen locations, deploying over 400 ground-based sensors encompassing infrasound, seismic, distributed acoustic sensing (DAS), and GPS technologies. Additionally, balloons equipped with infrasound sensors were launched to capture signals at higher altitudes. This campaign (the largest of its kind so far) yielded a wealth of invaluable data anticipated to fuel scientific inquiry for years to come. The success of the observational campaign is evidenced by the near-universal detection of signals across instruments, both proximal and distal. This paper presents a comprehensive overview of the collective scientific effort, field deployment, and preliminary findings. The early findings have the potential to inform future space missions and terrestrial campaigns, contributing to our understanding of meteoroid interactions with planetary atmospheres. Furthermore, the dataset collected during this campaign will improve entry and propagation models as well as augment the study of atmospheric dynamics and shock phenomena generated by meteoroids and similar sources., Comment: 87 pages, 14 figures

Recent advances in multimodal imaging acquisition techniques have allowed us to measure different aspects of brain structure and function. Multimodal fusion, such as linked independent component analysis (LICA), is popularly used to integrate complementary information. However, it has suffered from missing data, commonly occurring in neuroimaging data. Therefore, in this paper, we propose a Full Information LICA algorithm (FI-LICA) to handle the missing data problem during multimodal fusion under the LICA framework. Built upon complete cases, our method employs the principle of full information and utilizes all available information to recover the missing latent information. Our simulation experiments showed the ideal performance of FI-LICA compared to current practices. Further, we applied FI-LICA to multimodal data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study, showcasing better performance in classifying current diagnosis and in predicting the AD transition of participants with mild cognitive impairment (MCI), thereby highlighting the practical utility of our proposed method., Comment: 17 pages, 6 figures

The Dragonfly Spectral Line Mapper is an innovative all-refracting telescope designed to carry out ultra-low surface brightness wide-field mapping of visible wavelength line emission. Equipped with ultranarrowband (0.8 nm bandwidth) filters mounted in Dragonfly Filter-Tilter instrumentation, the Dragonfly Spectral Line Mapper maps H$\alpha$, [NII]$\lambda$6583, and [OIII]$\lambda$5007 line emission produced by structures with sizes ranging from $\sim$1 to 1000 kpc in the local Universe. These spatial scales encompass that of the exceedingly diffuse and faintly radiating circumgalactic medium, which is singularly difficult to detect with conventional mirror-based telescope instrumentation. Extremely careful control of systematics is required to directly image these large scale structures, necessitating high fidelity sky background subtraction, wavelength calibration, and specialized flat-fielding methods. In this paper, we discuss the on-sky performance of the Dragonfly Spectral Line Mapper with these methods in place., Comment: 10 pages, 5 figures, SPIE Astronomical Telescopes and Instrumentation 2024 Proceedings

The Dragonfly Spectral Line Mapper (DSLM) is a semi-autonomous, distributed-aperture based telescope design, featuring a modular setup of 120 Canon telephoto lenses, and equal numbers of ultra-narrowband filters, detectors, and other peripherals. Here we introduce the observatory software stack for this highly-distributed system. Its core is the Dragonfly Communication Protocol (DCP), a pure-Python hardware communication framework for standardized hardware interaction. On top of this are 120 REST-ful FastAPI web servers, hosted on Raspberry Pis attached to each unit, orchestrating command translation to the hardware and providing diagnostic feedback to a central control system running the global instrument control software. We discuss key features of this software suite, including docker containerization for environment management, class composition as a flexible framework for array commands, and a state machine algorithm which controls the telescope during autonomous observations., Comment: 18 pages, presented at the SPIE Astronomical Telescopes and Instrumentation conference in Yokohama, Japan

The Dragonfly Spectral Line Mapper is a mosaic telescope comprising 120 Canon telephoto lenses, based on the design of the Dragonfly Telephoto Array. With a wide field of view, and the addition of the "Dragonfly Filter-Tilter" instrumentation holding ultra narrow bandpass filters in front of each lens, the Dragonfly Spectral Line mapper is optimized for ultra low surface brightness imaging of visible wavelength line emission. The Dragonfly Spectral Line Mapper was constructed and commissioned in four phases from March 2022 to November 2023. During this time, four individual mounts of 30 lenses each were constructed and commissioned. The commissioning of the telescope included the deployment of the "Dragonfly StarChaser" which carries out image stabilization corrections in the telephoto lens, to enable hour-long exposures to be taken. In addition, we introduced new instrumentation such as a film to cover the optics to keep the filters clean. Here we describe the updated design of the complete 120-lens array, and the implementation of the instrumentation described above. Additionally, we present updated characterization of the cameras and filter transmission for the full array. Finally, we reflect on the construction and commissioning process of the complete 120-lens array Dragonfly Spectral Line Mapper, and remark on the feasibility of a larger 1000-lens array., Comment: 12 pages, 9 figures, SPIE conference proceedings

Language models (LMs) have been shown to behave unexpectedly post-deployment. For example, new jailbreaks continually arise, allowing model misuse, despite extensive red-teaming and adversarial training from developers. Given most model queries are unproblematic and frequent retraining results in unstable user experience, methods for mitigation of worst-case behavior should be targeted. One such method is classifying inputs as potentially problematic, then selectively applying steering vectors on these problematic inputs, i.e. adding particular vectors to model hidden states. However, steering vectors can also negatively affect model performance, which will be an issue on cases where the classifier was incorrect. We present KL-then-steer (KTS), a technique that decreases the side effects of steering while retaining its benefits, by first training a model to minimize Kullback-Leibler (KL) divergence between a steered and unsteered model on benign inputs, then steering the model that has undergone this training. Our best method prevents 44% of jailbreak attacks compared to the original Llama-2-chat-7B model while maintaining helpfulness (as measured by MT-Bench) on benign requests almost on par with the original LM. To demonstrate the generality and transferability of our method beyond jailbreaks, we show that our KTS model can be steered to reduce bias towards user-suggested answers on TruthfulQA. Code is available: https://github.com/AsaCooperStickland/kl-then-steer.

In reinforcement learning, specification gaming occurs when AI systems learn undesired behaviors that are highly rewarded due to misspecified training goals. Specification gaming can range from simple behaviors like sycophancy to sophisticated and pernicious behaviors like reward-tampering, where a model directly modifies its own reward mechanism. However, these more pernicious behaviors may be too complex to be discovered via exploration. In this paper, we study whether Large Language Model (LLM) assistants which find easily discovered forms of specification gaming will generalize to perform rarer and more blatant forms, up to and including reward-tampering. We construct a curriculum of increasingly sophisticated gameable environments and find that training on early-curriculum environments leads to more specification gaming on remaining environments. Strikingly, a small but non-negligible proportion of the time, LLM assistants trained on the full curriculum generalize zero-shot to directly rewriting their own reward function. Retraining an LLM not to game early-curriculum environments mitigates, but does not eliminate, reward-tampering in later environments. Moreover, adding harmlessness training to our gameable environments does not prevent reward-tampering. These results demonstrate that LLMs can generalize from common forms of specification gaming to more pernicious reward tampering and that such behavior may be nontrivial to remove., Comment: Make it easier to find samples from the model, and highlight that our operational definition of reward tampering has false positives where the model attempts to complete the task honestly but edits the reward. Add paragraph to conclusion to this effect, and add sentence to figure 1 to this effect

Interferometric experiments designed to detect the highly redshifted 21-cm signal from neutral hydrogen are producing increasingly stringent constraints on the 21-cm power spectrum, but some k-modes remain systematics-dominated. Mutual coupling is a major systematic that must be overcome in order to detect the 21-cm signal, and simulations that reproduce effects seen in the data can guide strategies for mitigating mutual coupling. In this paper, we analyse 12 nights of data from the Hydrogen Epoch of Reionization Array and compare the data against simulations that include a computationally efficient and physically motivated semi-analytic treatment of mutual coupling. We find that simulated coupling features qualitatively agree with coupling features in the data; however, coupling features in the data are brighter than the simulated features, indicating the presence of additional coupling mechanisms not captured by our model. We explore the use of fringe-rate filters as mutual coupling mitigation tools and use our simulations to investigate the effects of mutual coupling on a simulated cosmological 21-cm power spectrum in a "worst case" scenario where the foregrounds are particularly bright. We find that mutual coupling contaminates a large portion of the "EoR Window", and the contamination is several orders-of-magnitude larger than our simulated cosmic signal across a wide range of cosmological Fourier modes. While our fiducial fringe-rate filtering strategy reduces mutual coupling by roughly a factor of 100 in power, a non-negligible amount of coupling cannot be excised with fringe-rate filters, so more sophisticated mitigation strategies are required., Comment: 19 pages, 12 figures, submitted to MNRAS

PLATO (PLAnetary Transits and Oscillations of stars) is ESA's M3 mission designed to detect and characterise extrasolar planets and perform asteroseismic monitoring of a large number of stars. PLATO will detect small planets (down to <2 R_(Earth)) around bright stars (<11 mag), including terrestrial planets in the habitable zone of solar-like stars. With the complement of radial velocity observations from the ground, planets will be characterised for their radius, mass, and age with high accuracy (5 %, 10 %, 10 % for an Earth-Sun combination respectively). PLATO will provide us with a large-scale catalogue of well-characterised small planets up to intermediate orbital periods, relevant for a meaningful comparison to planet formation theories and to better understand planet evolution. It will make possible comparative exoplanetology to place our Solar System planets in a broader context. In parallel, PLATO will study (host) stars using asteroseismology, allowing us to determine the stellar properties with high accuracy, substantially enhancing our knowledge of stellar structure and evolution. The payload instrument consists of 26 cameras with 12cm aperture each. For at least four years, the mission will perform high-precision photometric measurements. Here we review the science objectives, present PLATO's target samples and fields, provide an overview of expected core science performance as well as a description of the instrument and the mission profile at the beginning of the serial production of the flight cameras. PLATO is scheduled for a launch date end 2026. This overview therefore provides a summary of the mission to the community in preparation of the upcoming operational phases.