Search

The immediate impacts of COVID-19 on K12 schooling are well known. Over nearly 18 months, students' academic performance and mental health deteriorated dramatically. This study aims to identify if and how the pandemic led to longer-term changes in core aspects of schooling. Drawing on semi-structured interviews with 31 teachers and administrators across 12 districts in two states, we find that schools today look quite different in several areas including the availability and use of instructional technology, instructional practice, parent-teacher communication, and the balance between academics and social-emotional well-being. We interpret these findings through the lens of institutional theory, and discuss implications of the changes for practitioners and policymakers.

This paper reports findings from a nationally representative survey of K-12 teachers in May 2023 that examines the potential long-term impacts of COVID-19 on public schooling. The findings suggest fundamental ways in which school operations, instructional practice and parent-teacher interaction have changed since the pandemic. Some changes seem promising; others suggest caution. While policymakers may not be able to directly influence some of the reported changes in the short run, monitoring the evolution of school practices (and their consequences for children) will position educational leaders to help teachers and students address the consequences of the COVID-19 pandemic going forward.

There's a rise in the employment of uncertified teachers, driven by acute staffing shortages and the flexibility offered by Texas' District of Innovation plans. This reliance on uncertified educators is raising alarms among educators and policymakers alike. Concerns are mounting over whether these teachers, often entering the classroom having never worked in public schools, are equipped to meet the demands of today's classrooms. This brief presents new evidence showing that uncertified teachers are linked to declines in student achievement. The study examines the prevalence, backgrounds, and impacts of uncertified teachers on student achievement and draws attention to significant differences within this diverse group of educators. As schools continue to hire uncertified teachers to address staffing shortages, knowing the backgrounds of these teachers and how they shape student achievement helps stakeholders prioritize transition points, such as hiring individuals with previous classroom experience in non-teaching roles. This brief outlines findings from the 2021-22 and 2022-23 school years and provides recommendations for policymakers and stakeholder groups.

Pulsar timing array observations have found evidence for an isotropic gravitational wave background with the Hellings-Downs angular correlations, expected from general relativity. This interpretation hinges on the measured shape of the angular correlations, which is predominately quadrupolar under general relativity. Here we explore a more flexible parameterization: we expand the angular correlations into a sum of Legendre polynomials and use a Bayesian analysis to constrain their coefficients with the 15-year pulsar timing data set collected by the North American Nanohertz Observatory for Gravitational Waves (NANOGrav). When including Legendre polynomials with multipoles $\ell \geq 2$, we only find a significant signal in the quadrupole with an amplitude consistent with general relativity and non-zero at the $\sim 95\%$ confidence level and a Bayes factor of 200. When we include multipoles $\ell \leq 1$, the Bayes factor evidence for quadrupole correlations decreases by more than an order of magnitude due to evidence for a monopolar signal at approximately 4 nHz which has also been noted in previous analyses of the NANOGrav 15-year data. Further work needs to be done in order to better characterize the properties of this monopolar signal and its effect on the evidence for quadrupolar angular correlations., Comment: 15 pages, 6 figures

We introduce Edify 3D, an advanced solution designed for high-quality 3D asset generation. Our method first synthesizes RGB and surface normal images of the described object at multiple viewpoints using a diffusion model. The multi-view observations are then used to reconstruct the shape, texture, and PBR materials of the object. Our method can generate high-quality 3D assets with detailed geometry, clean shape topologies, high-resolution textures, and materials within 2 minutes of runtime., Comment: Project website: https://research.nvidia.com/labs/dir/edify-3d

The detection of a stochastic gravitational wave background by pulsar timing arrays suggests the presence of a supermassive black hole binary population. Although the observed spectrum generally aligns with predictions from orbital evolution driven by gravitational wave emission in circular orbits, there is a discernible preference for a turnover at the lowest observed frequencies. This turnover could indicate a significant hardening phase, transitioning from early environmental influences to later stages predominantly influenced by gravitational wave emission. In the vicinity of these binaries, the ejection of stars or dark matter particles through gravitational three-body slingshots efficiently extracts orbital energy, leading to a low-frequency turnover in the spectrum. By analyzing the NANOGrav 15-year data, we assess how the gravitational wave spectrum depends on the initial inner galactic profile prior to disruption by binary ejections, accounting for a range of initial binary eccentricities. Our findings suggest a parsec-scale galactic center density around $10^6\,M_\odot/\textrm{pc}^3$ across most of the parameter space, offering insights into the environmental effects on black hole evolution and combined matter density near galaxy centers., Comment: 15 pages, 5 figures

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

This study addresses the burgeoning phenomenon of fully online alternative teacher certification programs (ACPs). In Texas where most teachers are prepared via ACPs, our research zeroes in on the proportion of teachers who are prepared fully online and the relative effectiveness of teacher preparation programs on student achievement and teacher retention. Using statewide longitudinal data from 2014-2023, our findings show that 1 in 4 of Texas students are being taught by teachers prepared fully online Students taught by teachers prepared online exhibit comparable levels of achievement to those taught by uncertified teachers, underperforming compared to students taught by teachers from other preparation pathways. Moreover, these teachers exhibit a markedly higher turnover rate. This study contributes to the ongoing discourse on teacher preparation quality, offering insights for policymakers and stakeholders.

An optimal transport (OT) problem seeks to find the cheapest mapping between two distributions with equal total density, given the cost of transporting density from one place to another. Unbalanced OT allows for different total density in each distribution. This is the typical setting for precipitation forecast and observation data, when considering the densities as accumulated rainfall, or intensity. In this work, entropic unbalanced OT and its associated Sinkhorn divergence are examined as a spatial forecast verification method for precipitation data. It offers many attractive features, such as morphing one field into another, defining a distance between fields and providing feature based optimal assignment. It is found that the Sinkhorn divergence is robust against the common double penalty problem (a form of phase error), aligns with expert assessments of model performance, and allows for a variety of novel pictorial illustrations of error. It provides informative summary scores, and has few limitations to its application. Combined, these findings place unbalanced entropy regularised optimal transport and the Sinkhorn divergence as an informative method which follows geometric intuition., Comment: submitted to Meteorological Applications

We show the existence of a properly immersed translating solution to curve diffusion flow in the plane. Curve diffusion flow is a higher order version of curve shortening flow, namely \[ \left( \frac{dX}{dt}\right) ^{\perp}=-\kappa_{ss}N. \]

Despite the growing use of large language models (LLMs) for providing feedback, limited research has explored how to achieve high-quality feedback. This case study introduces an evaluation framework to assess different zero-shot prompt engineering methods. We varied the prompts systematically and analyzed the provided feedback on programming errors in R. The results suggest that prompts suggesting a stepwise procedure increase the precision, while omitting explicit specifications about which provided data to analyze improves error identification., Comment: 8 pages, 1 figure

Scaling has not yet been convincingly demonstrated for pure self-supervised learning from video. However, prior work has focused evaluations on semantic-related tasks $\unicode{x2013}$ action classification, ImageNet classification, etc. In this paper we focus on evaluating self-supervised learning on non-semantic vision tasks that are more spatial (3D) and temporal (+1D = 4D), such as camera pose estimation, point and object tracking, and depth estimation. We show that by learning from very large video datasets, masked auto-encoding (MAE) with transformer video models actually scales, consistently improving performance on these 4D tasks, as model size increases from 20M all the way to the largest by far reported self-supervised video model $\unicode{x2013}$ 22B parameters. Rigorous apples-to-apples comparison with many recent image and video models demonstrates the benefits of scaling 4D representations.

The suite of datasets commonly used to train and evaluate the mathematical capabilities of AI-based mathematical copilots (primarily large language models) exhibit several shortcomings. These limitations include a restricted scope of mathematical complexity, typically not exceeding lower undergraduate-level mathematics, binary rating protocols and other issues, which makes comprehensive proof-based evaluation suites difficult. We systematically explore these limitations and contend that enhancing the capabilities of large language models, or any forthcoming advancements in AI-based mathematical assistants (copilots or "thought partners"), necessitates a paradigm shift in the design of mathematical datasets and the evaluation criteria of mathematical ability: It is necessary to move away from result-based datasets (theorem statement to theorem proof) and convert the rich facets of mathematical research practice to data LLMs can train on. Examples of these are mathematical workflows (sequences of atomic, potentially subfield-dependent tasks that are often performed when creating new mathematics), which are an important part of the proof-discovery process. Additionally, we advocate for mathematical dataset developers to consider the concept of "motivated proof", introduced by G. P\'olya in 1949, which can serve as a blueprint for datasets that offer a better proof learning signal, alleviating some of the mentioned limitations. Lastly, we introduce math datasheets for datasets, extending the general, dataset-agnostic variants of datasheets: We provide a questionnaire designed specifically for math datasets that we urge dataset creators to include with their datasets. This will make creators aware of potential limitations of their datasets while at the same time making it easy for readers to assess it from the point of view of training and evaluating mathematical copilots., Comment: 40 pages

Robot learning is witnessing a significant increase in the size, diversity, and complexity of pre-collected datasets, mirroring trends in domains such as natural language processing and computer vision. Many robot learning methods treat such datasets as multi-task expert data and learn a multi-task, generalist policy by training broadly across them. Notably, while these generalist policies can improve the average performance across many tasks, the performance of generalist policies on any one task is often suboptimal due to negative transfer between partitions of the data, compared to task-specific specialist policies. In this work, we argue for the paradigm of training policies during deployment given the scenarios they encounter: rather than deploying pre-trained policies to unseen problems in a zero-shot manner, we non-parametrically retrieve and train models directly on relevant data at test time. Furthermore, we show that many robotics tasks share considerable amounts of low-level behaviors and that retrieval at the "sub"-trajectory granularity enables significantly improved data utilization, generalization, and robustness in adapting policies to novel problems. In contrast, existing full-trajectory retrieval methods tend to underutilize the data and miss out on shared cross-task content. This work proposes STRAP, a technique for leveraging pre-trained vision foundation models and dynamic time warping to retrieve sub-sequences of trajectories from large training corpora in a robust fashion. STRAP outperforms both prior retrieval algorithms and multi-task learning methods in simulated and real experiments, showing the ability to scale to much larger offline datasets in the real world as well as the ability to learn robust control policies with just a handful of real-world demonstrations., Comment: Project website at https://weirdlabuw.github.io/strap/

This summary of the second Terrestrial Very-Long-Baseline Atom Interferometry (TVLBAI) Workshop provides a comprehensive overview of our meeting held in London in April 2024, building on the initial discussions during the inaugural workshop held at CERN in March 2023. Like the summary of the first workshop, this document records a critical milestone for the international atom interferometry community. It documents our concerted efforts to evaluate progress, address emerging challenges, and refine strategic directions for future large-scale atom interferometry projects. Our commitment to collaboration is manifested by the integration of diverse expertise and the coordination of international resources, all aimed at advancing the frontiers of atom interferometry physics and technology, as set out in a Memorandum of Understanding signed by over 50 institutions., Comment: Summary of the second Terrestrial Very-Long-Baseline Atom Interferometry Workshop held at Imperial College London: https://indico.cern.ch/event/1369392/

The pseudogap in high-temperature superconducting cuprates is an exotic state of matter, displaying emerging Fermi arcs and a momentum-selective suppression of states upon cooling. We show how these phenomena are originating in the three-band Emery model by performing cutting-edge dynamical vertex approximation calculations for its normal state. For the hole-doped parent compound our results demonstrate the formation of a pseudogap due to short-ranged commensurate antiferromagnetic fluctuations. At larger doping values, progressively, incommensurate correlations and a metallic regime appear. Our results are in qualitative agreement with the normal state of cuprates, and, hence, represent a crucial step towards the uniform description of their phase diagrams within a single theoretical framework., Comment: 7 pages, 4 figures

We present DriveGPT, a scalable behavior model for autonomous driving. We model driving as a sequential decision making task, and learn a transformer model to predict future agent states as tokens in an autoregressive fashion. We scale up our model parameters and training data by multiple orders of magnitude, enabling us to explore the scaling properties in terms of dataset size, model parameters, and compute. We evaluate DriveGPT across different scales in a planning task, through both quantitative metrics and qualitative examples including closed-loop driving in complex real-world scenarios. In a separate prediction task, DriveGPT outperforms a state-of-the-art baseline and exhibits improved performance by pretraining on a large-scale dataset, further validating the benefits of data scaling., Comment: 14 pages, 16 figures, 9 tables, and 1 video link

The symmetry requirements for realizing unconventional compensated magnets with spin-polarized bands such as altermagnets have recently been uncovered. The most recent addition to this family of magnets is parity-odd or $p$-wave magnets. We demonstrate that $p$-wave magnets are perfectly compatible with superconductivity due to the spin polarization of their electron bands and that they induce unexpected spin transport phenomena. We first show that $p$-wave magnetism can coexist with conventional superconductivity regardless of the magnitude of the spin splitting. We then predict that $p$-wave magnets induce a charge-to-spin conversion, which can be strongly enhanced by the presence of superconductivity providing a way to probe the coexistence in experiments. Our results open a new avenue for material combinations with a synergetic relation between spintronics and superconductivity., Comment: 6 pages, 3 figures

In this work, we utilize new coil objectives for stellarator optimization with autodifferentiation, including pointwise and net coil-coil forces and torques. We use these methods to perform the first large-scale optimization of planar dipole coil arrays, since arrays of small and geometrically simple coils have been proposed to partially produce the 3D magnetic fields for stellarators, generate advantageous magnetic field perturbations in tokamaks, and provide active, real-time control capabilities. We perform an ablation study to show that minimizing the orientation and location of each coil may be essential to get coil forces, coil torques, and field errors to tolerable levels. We conclude by generating planar coil array solutions for three reactor-scale quasi-symmetric stellarators.

Two-dimensional optical spectroscopy experiments have shown that exciton transfer pathways in the Fenna-Matthews-Olson (FMO) photosynthetic complex differ drastically under reduced and oxidised conditions, suggesting a functional role for collective vibronic mechanisms that may be active in the reduced form but attenuated in the oxidised state. Higgins et al. [PNAS 118 (11) e2018240118 (2021)] used Redfield theory to link the experimental observations to altered exciton transfer rates due to oxidative onsite energy shifts that detune excitonic energy gaps from a specific vibrational frequency of the bacteriochlorophyll (BChl) a. Using a memory kernel formulation of the hierarchical equations of motion, we present non-perturbative estimations of transfer rates that yield a modified physical picture. Our findings indicate that onsite energy shifts alone cannot reproduce the observed rate changes in oxidative environments, either qualitatively or quantitatively. By systematically examining combined changes both in site energies and the local environment for the oxidised complex, while maintaining consistency with absorption spectra, our results suggest that vibronic tuning of transfer rates may indeed be active in the reduced complex. However, we achieve qualitative, but not quantitative, agreement with the experimentally measured rates. Our analysis indicates potential limitations of the FMO electronic Hamiltonian, which was originally derived by fitting spectra to second-order cumulant and Redfield theories. This suggests that reassessment of these electronic parameters with a non-perturbative scheme, or derived from first principles, is essential for a consistent and accurate understanding of exciton dynamics in FMO under varying redox conditions., Comment: 14 pages, 7 figures

We study linear contracts for combinatorial problems in multi-agent settings. In this problem, a principal designs a linear contract with several agents, each of whom can decide to take a costly action or not. The principal observes only the outcome of the agents' collective actions, not the actions themselves, and obtains a reward from this outcome. Agents that take an action incur a cost, and so naturally agents require a fraction of the principal's reward as an incentive for taking their action. The principal needs to decide what fraction of their reward to give to each agent so that the principal's expected utility is maximized. Our focus is on the case when the agents are vertices in a graph and the principal's reward corresponds to the number of edges between agents who take their costly action. This case represents the natural scenario when an action of each agent complements actions of other agents though collaborations. Recently, Deo-Campo Vuong et.al. showed that for this problem it is impossible to provide any finite multiplicative approximation or additive FPTAS unless $\mathcal{P} = \mathcal{NP}$. On a positive note, the authors provided an additive PTAS for the case when all agents have the same cost. They asked whether an additive PTAS can be obtained for the general case, i.e when agents potentially have different costs. We answer this open question in positive.

We present spectra of the supernova (SN) impostor AT 2016blu spanning over a decade. This transient exhibits quasiperiodic outbursts with a $\sim$113 d period, likely triggered by periastron encounters in an eccentric binary system where the primary star is a luminous blue variable (LBV). The overall spectrum remains fairly consistent during quiescence and eruptions, with subtle changes in line-profile shapes and other details. Some narrow emission features indicate contamination from a nearby H II region in the host galaxy, NGC 4559. Broader H$\alpha$ profiles exhibit Lorentzian shapes with full width at half-maximum intensity (FWHM) values that vary significantly, showing no correlation with photometric outbursts or the 113 d phase. At some epochs, H$\alpha$ exhibits asymmetric profiles with a stronger redshifted wing, while broad and sometimes multicomponent P Cygni absorption features occasionally appear, but are again uncorrelated with brightness or phase. These P Cygni absorptions have high velocities compared to the FWHM of the H$\alpha$ emission line, perhaps suggesting that the absorption component is not in the LBV's wind, but is instead associated with a companion. The lack of phase dependence in line-profile changes may point to interaction between a companion and a variable or inhomogeneous primary wind, in an orbit with only mild eccentricity. Recent photometric data indicate that AT 2016blu experienced its 20th outburst around May/June 2023, as predicted based on its period. This type of quasiperiodic LBV remains poorly understood, but its spectra and erratic light curve resemble some pre-SN outbursts like those of SN 2009ip., Comment: 21 pages, 17 figures, and 1 table, Submitted to MNRAS

It has recently been realized that magnetic systems with coplanar magnetic order that are invariant under the combined operation of time-reversal and translation with half a unit cell feature energy bands with a symmetry-protected $p$-wave spin polarization. Such $p$-wave magnets are a sought-after spin analogy of unconventional triplet superconducting pairing and show promise for use in spintronics. Metallic helimagnets provide a realization of $p$-wave magnetism, but such order often occurs in systems lacking inversion symmetry so that Rashba spin-orbit interactions can be prominent. An important question is therefore how the magnitude and the existence of $p$-wave spin polarization is affected by Rashba spin-orbit interaction. Here, we prove that while the $p$-wave symmetry of the spin-polarized bands is strictly speaking removed by spin-orbit interactions in helimagnets unless the period of the helix is fine-tuned, the actual quantitative deviation from $p$-wave symmetry is extremely weak unless the period of the helix is only a few lattice sites. Thereafter, we show that the $p$-wave magnetism becomes completely robust in pairs of antiferromagnetically coupled helices. More precisely, the $p$-wave spin-polarization of the bands then appears regardless of the periodicity and regardless of the strength of the spin-orbit interactions. This shows that antiferromagnetically coupled helimagnetic chains produce robust $p$-wave spin polarization free of fine-tuning requirements, making them attractive for potential spintronic applications.

This paper presents the SareQuant project, which aims to evolve the Basque NREN (National Research and Education Networks) into a quantum-based communication infrastructure. SareQuant focuses on the network design and on the integration of quantum technologies into real-world scenarios and applications. Therefore, this paper provides insights into the opportunities and challenges regarding the integration of quantum technologies, thus paving the way for a secure and advanced Quantum Internet.

The development of quantum circuits based on hybrid superconductor-semiconductor Josephson junctions holds promise for exploring their mesoscopic physics and for building novel superconducting devices. The gate-tunable superconducting transmon qubit (gatemon) is the paradigmatic example of such a superconducting circuit. However, gatemons typically suffer from unstable and hysteretic qubit frequencies with respect to the applied gate voltage and reduced coherence times. Here we develop methods for characterizing these challenges in gatemons and deploy these methods to compare the impact of shunt capacitor designs on gatemon performance. Our results indicate a strong frequency- and design-dependent behavior of the qubit stability, hysteresis, and dephasing times. Moreover, we achieve highly reliable tuning of the qubit frequency with 1 MHz precision over a range of several GHz, along with improved stability in grounded gatemons compared to gatemons with a floating capacitor design., Comment: 13 pages, 12 figures

We consider the challenge of black-box optimization within hybrid discrete-continuous and variable-length spaces, a problem that arises in various applications, such as decision tree learning and symbolic regression. We propose DisCo-DSO (Discrete-Continuous Deep Symbolic Optimization), a novel approach that uses a generative model to learn a joint distribution over discrete and continuous design variables to sample new hybrid designs. In contrast to standard decoupled approaches, in which the discrete and continuous variables are optimized separately, our joint optimization approach uses fewer objective function evaluations, is robust against non-differentiable objectives, and learns from prior samples to guide the search, leading to significant improvement in performance and sample efficiency. Our experiments on a diverse set of optimization tasks demonstrate that the advantages of DisCo-DSO become increasingly evident as the complexity of the problem increases. In particular, we illustrate DisCo-DSO's superiority over the state-of-the-art methods for interpretable reinforcement learning with decision trees., Comment: Accepted at AAAI-25

An organism that is newly introduced into an existing population has a survival probability that is dependent on both the population density of its environment and the competition it experiences with the members of that population. Expanding populations naturally form regions of high and low density, and simultaneously experience ecological interactions both internally and at the boundary of their range. For this reason, systems of expanding populations are ideal for studying the combination of density and ecological effects. Conservation ecologists have been studying the ability of an invasive species to establish for some time, attributing success to both ecological and spatial factors. Similar behaviors have been observed in spatially structured cell populations, such as those found in cancerous tumors and bacterial biofilms. In these scenarios, novel organisms may be the introduction of a new mutation or bacterial species with some form of drug resistance, leading to the possibility of treatment failure. In order to gain insight into the relationship between population density and ecological interactions, we study an expanding population of interacting wild-type cells and mutant cells. We simulate these interactions in time and study the spatially dependent probability for a mutant to survive or to take over the front of the population wave (gene surfing). Additionally, we develop a mathematical model that describes this survival probability and find agreement when the payoff for the mutant is positive (corresponding to cooperation, exploitation, or commensalism). By knowing the types of interactions, our model provides insight into the spatial distribution of survival probability. Conversely, given a spatial distribution of survival probabilities, our model provides insight into the types of interactions that were involved to generate it., Comment: 21 pages, 5 figures, 3 tables

Near-field thermophotovoltaics (NFTPV) systems have significant potential for waste heat recovery applications, with both high theoretical efficiency and power density, up to 40% and $11 \ \mathrm{W/cm^{2}}$ at 900 K. Yet experimental demonstrations have only achieved up to 14% efficiency and modest power densities (i.e., $0.75 \ \mathrm{W/cm^{2}}$). While experiments have recently started to focus on photovoltaic (PV) cells custom-made for NFTPV, most work still relies on conventional doped silicon radiators. In this work, we design an optimized NFTPV radiator for an indium arsenide-based system and, in the process, investigate models for the permittivity of InAs in the context of NFTPV. Based on existing measurements of InAs absorption, we find that the traditional Drude model overestimates free carrier absorption in InAs. We replace the Drude portion of the InAs dielectric function with a revised model derived from ionized impurity scattering. Using this revised model, we maximize the spectral efficiency and power density of a NFTPV system by optimizing the spectral coupling between a radiator and an InAs PV cell. We find that when the radiator and the PV cell are both made of InAs, a nearly threefold improvement of spectral efficiency is possible compared to a traditional silicon radiator with the same InAs cell. This enhancement reduces subgap thermal transfer while maintaining power output., Comment: 11 pages, 9 figures

Ram pressure stripped galaxies are rare cases of environmental evolution in action. However, our ability to understand these galaxies is limited by the small number of identified galaxies experiencing ram pressure stripping (RPS). Our aim is to explore the efficacy of citizen science classifications in identifying ram pressure stripped galaxies, and use this to aid in motivating new samples of ram pressure stripped candidates. We compile a sample of over 200 known ram pressure stripped galaxies from existing literature, with morphological classifications obtained from Galaxy Zoo. We compare these galaxies with magnitude and redshift-matched comparison cluster and field galaxies. Additionally, we create a sample of SDSS cluster galaxies, with morphological classifications similar to known ram pressure stripped galaxies, and compare the fraction of potential new RPS candidates against control samples. We find that ram pressure stripped galaxies exhibit a higher proportion of odd and irregular morphological classifications compared to field and cluster comparison samples. This trend is particularly pronounced in galaxies displaying strong optical ram pressure stripping features, but absent from galaxies with only radio tails. We find that SDSS galaxies with Galaxy Zoo classifications consistent with the known RPS galaxies have a higher fraction of visible ram pressure stripping features ($19\%$) compared with other cluster galaxies ($12\%$) when classified by experts. We identify 101 new ram pressure stripping candidate galaxies through these expert classifications. We demonstrate that indirect morphological classifications from citizen science projects can increase the efficiency in which new stripping candidates are found. Projects such as Galaxy Zoo can aid in the identification of ram pressure stripped galaxies that are key to understanding galaxy evolution in clusters., Comment: 20 pages, 11 figures, accepted for publication in Astronomy and Astrophysics

The purpose of this short note is to show the interplay between math outreach and conducting original research, in particular how each can build off the other., Comment: 9 pages, 5 figures

Recent latent-space monitoring techniques have shown promise as defenses against LLM attacks. These defenses act as scanners that seek to detect harmful activations before they lead to undesirable actions. This prompts the question: Can models execute harmful behavior via inconspicuous latent states? Here, we study such obfuscated activations. We show that state-of-the-art latent-space defenses -- including sparse autoencoders, representation probing, and latent OOD detection -- are all vulnerable to obfuscated activations. For example, against probes trained to classify harmfulness, our attacks can often reduce recall from 100% to 0% while retaining a 90% jailbreaking rate. However, obfuscation has limits: we find that on a complex task (writing SQL code), obfuscation reduces model performance. Together, our results demonstrate that neural activations are highly malleable: we can reshape activation patterns in a variety of ways, often while preserving a network's behavior. This poses a fundamental challenge to latent-space defenses., Comment: Project page: https://obfuscated-activations.github.io/

Computational modeling of aerodynamics is a key problem in aerospace engineering, often involving flows interacting with solid objects such as airfoils. Deep surrogate models have emerged as purely data-driven approaches that learn direct mappings from simulation conditions to solutions based on either simulation or experimental data. Here, we consider modeling of incompressible flows over solid objects, wherein geometric structures are a key factor in determining aerodynamics. To effectively incorporate geometries, we propose a message passing scheme that efficiently and expressively integrates the airfoil shape with the mesh representation. Under this framework, we first obtain a representation of the geometry in the form of a latent graph on the airfoil surface. We subsequently propagate this representation to all collocation points through message passing on a directed, bipartite graph. We demonstrate that this framework supports efficient training by downsampling the solution mesh while avoiding distribution shifts at test time when evaluated on the full mesh. To enable our model to be able to distinguish between distinct spatial regimes of dynamics relative to the airfoil, we represent mesh points in both a leading edge and trailing edge coordinate system. We further enhance the expressiveness of our coordinate system representations by embedding our hybrid Polar-Cartesian coordinates using sinusoidal and spherical harmonics bases. We additionally find that a change of basis to canonicalize input representations with respect to inlet velocity substantially improves generalization. Altogether, these design choices lead to a purely data-driven machine learning framework known as GeoMPNN, which won the Best Student Submission award at the NeurIPS 2024 ML4CFD Competition, placing 4th overall. Our code is publicly available as part of the AIRS library (https://github.com/divelab/AIRS).

Interpretability methods seek to understand language model representations, yet the outputs of most such methods -- circuits, vectors, scalars -- are not immediately human-interpretable. In response, we introduce LatentQA, the task of answering open-ended questions about model activations in natural language. Towards solving LatentQA, we propose Latent Interpretation Tuning (LIT), which finetunes a decoder LLM on a dataset of activations and associated question-answer pairs, similar to how visual instruction tuning trains on question-answer pairs associated with images. We use the decoder for diverse reading applications, such as extracting relational knowledge from representations or uncovering system prompts governing model behavior. Our decoder also specifies a differentiable loss that we use to control models, such as debiasing models on stereotyped sentences and controlling the sentiment of generations. Finally, we extend LatentQA to reveal harmful model capabilities, such as generating recipes for bioweapons and code for hacking., Comment: Project page is at https://latentqa.github.io

Quantum physics can be extended into the complex domain by considering non-Hermitian Hamiltonians that are $\mathcal{PT}$-symmetric. These exhibit exceptional points (EPs) where the eigenspectrum changes from purely real to purely imaginary values and have useful properties enabling applications such as accelerated entanglement generation and the delay of the sudden death of entanglement in noisy systems. An interesting question is whether similar beneficial effects can be achieved away from EPs, since this would extend the available parameter space and make experiments more accessible. We investigate this by considering the more general case of pseudo-Hermitian Hamiltonians where two-mode squeezing interactions are incorporated into a $\mathcal{PT}$-symmetric optomechanical system. The addition of squeezing is motivated by an attempt to extend the lifetime of the system's entanglement. We derive analytic expressions for the entanglement dynamics under noise-free conditions and present numerical simulations that include the effects of noise. Although we find that the two-mode squeezing interactions do not generally preserve the initial entanglement, rich dynamics are observed in both the pseudo-Hermitian and $\mathcal{PT}$-symmetric cases, including the sudden death and revival of entanglement under certain conditions. We find that the sudden disappearance of entanglement can be mitigated at EPs (similar to $\mathcal{PT}$-symmetric systems) but also show that the revival of entanglement is quite robust to thermal noise in a group of parameters away from the EPs. Our study extends our understanding of non-Hermitian systems and opens a new perspective for the development of quantum devices in non-Hermitian systems even away from EPs.

Benchmarking is one of the key ways in which we can gain insight into the strengths and weaknesses of optimization algorithms. In sampling-based optimization, considering the anytime behavior of an algorithm can provide valuable insights for further developments. In the context of multi-objective optimization, this anytime perspective is not as widely adopted as in the single-objective context. In this paper, we propose a new software tool which uses principles from unbounded archiving as a logging structure. This leads to a clearer separation between experimental design and subsequent analysis decisions. We integrate this approach as a new Python module into the IOHprofiler framework and demonstrate the benefits of this approach by showcasing the ability to change indicators, aggregations, and ranking procedures during the analysis pipeline.

Discovering and tracking communities in time-varying networks is an important task in network science, motivated by applications in fields ranging from neuroscience to sociology. In this work, we characterize the celebrated family of spectral methods for static clustering in terms of the low-rank approximation of high-dimensional node embeddings. From this perspective, it becomes natural to view the evolving community detection problem as one of subspace tracking on the Grassmann manifold. While the resulting optimization problem is nonconvex, we adopt a block majorize-minimize Riemannian optimization scheme to learn the Grassmann geodesic which best fits the data. Our framework generalizes any static spectral community detection approach and leads to algorithms achieving favorable performance on synthetic and real temporal networks, including those that are weighted, signed, directed, mixed-membership, multiview, hierarchical, cocommunity-structured, bipartite, or some combination thereof. We demonstrate how to specifically cast a wide variety of methods into our framework, and demonstrate greatly improved dynamic community detection results in all cases., Comment: 34 pages, 13 figures

1D superconductor-semiconductor nanowires are leading candidates for topological quantum computation due to their ability to host Majorana zero modes (MZMs). However, standard methods for identifying MZMs are often inadequate, particularly in the presence of disorder, where many properties considered to be heralds of MZMs are often generated by trivial disorder induced Andreev bound states. Recent works clearly indicate the need for developing new techniques for identifying and diagnosing MZMs. In this study, we utilize a Vision Transformer-based neural network to predict using tunnel conductance measurements, both whether a device manifests a topological MZMs phase, and also to map out the entire topological phase diagram. We show the ability of our method up to arbitrary confidence ($P>0.9998$) classify a device as possessing a non-trivial MZM-carrying topological phase for a wide variety of disorder parameters. Further, within these regimes we demonstrate an ability to predict from conductance measurements alternative (to the extensively used scattering-matrix-invariant topological indicator) Majorana indicators based on local density of states (LDOS). This work serves as a significant advance offering a step towards the practical realization of Majorana-based quantum devices, enabling a deep-learning understanding of the topological properties in disordered nanowire systems. We validate our method using extensive simulated Majorana results in the presence of disorder, and suggest using this technique for the analysis of experimental data in superconductor-semiconductor hybrid Majorana platforms., Comment: 21 pages, 12 Figures

In hydrogen-rich atmospheres with low mean molecular weight (MMW), an air parcel containing a higher-molecular-weight condensible can be negatively buoyant even if its temperature is higher than the surrounding environment. This should fundamentally alter the dynamics of moist convection, but the low-MMW regime has previously been explored primarily via one-dimensional theories that cannot capture the complexity of moist turbulence. Here, we use a three-dimensional cloud-resolving model to simulate moist convection in atmospheres with a wide range of background MMW, and confirm that a humidity threshold for buoyancy reversal first derived by Guillot (1995) coincides with an abrupt change in tropospheric structure. Crossing the "Guillot threshold" in near-surface humidity causes the dry (subcloud) boundary layer to collapse and be replaced by a very cloudy layer with a temperature lapse rate that exceeds the dry adiabatic rate. Simulations with reduced surface moisture availability in the lower atmosphere feature a deeper dry subcloud layer, which allows the superadiabatic cloud layer to remain aloft. Our simulations support a potentially observable systematic trend toward increased cloudiness for atmospheres with near-surface moisture concentrations above the Guillot threshold. This should apply to \ce{H2O} and potentially to other condensible species on hotter worlds. We also find evidence for episodic convective activity and associated variability in cloud cover in some of our low-MMW simulations, which should be investigated further with global-scale simulations.

Determining which rocky exoplanets have atmospheres, and why, is a key goal for JWST. So far, emission observations of individual rocky exoplanets orbiting M stars (M-Earths) have not provided definitive evidence for atmospheres. Here, we synthesize emission data for M-Earths and find a trend in measured brightness temperature (ratioed to its theoretical maximum value) as a function of instellation. However, the statistical evidence of this trend is dependent on the choice of stellar model and we consider its identification tentative. We show that this trend can be explained by either the onset of thin/tenuous atmospheres on colder worlds, or a population of bare rocks with stronger space weathering and/or coarser regolith on closer-in worlds. Such grain coarsening may be caused by sintering near the melting point of rock or frequent volcanic resurfacing. We also find that fresh, fine-grained surfaces can serve as a false positive to the detection of moderate atmospheric heat redistribution. However, we argue that such surfaces are unlikely given the ubiquity of space weathering in the Solar System and the low albedo of Solar System airless bodies. Furthermore, we highlight considerations when testing rocky planet hypotheses at the population level, including the choice of instrument, stellar modeling, and how brightness temperatures are derived. Emission data from a larger sample of M-Earths will be able to confirm or reject this tentative trend and diagnose its cause., Comment: Submitted to AAS Journals, comments welcome

The performance of modern reinforcement learning algorithms critically relies on tuning ever-increasing numbers of hyperparameters. Often, small changes in a hyperparameter can lead to drastic changes in performance, and different environments require very different hyperparameter settings to achieve state-of-the-art performance reported in the literature. We currently lack a scalable and widely accepted approach to characterizing these complex interactions. This work proposes a new empirical methodology for studying, comparing, and quantifying the sensitivity of an algorithm's performance to hyperparameter tuning for a given set of environments. We then demonstrate the utility of this methodology by assessing the hyperparameter sensitivity of several commonly used normalization variants of PPO. The results suggest that several algorithmic performance improvements may, in fact, be a result of an increased reliance on hyperparameter tuning.

Botnets could autonomously infect, propagate, communicate and coordinate with other members in the botnet, enabling cybercriminals to exploit the cumulative computing and bandwidth of its bots to facilitate cybercrime. Traditional detection methods are becoming increasingly unsuitable against various network-based detection evasion methods. These techniques ultimately render signature-based fingerprinting detection infeasible and thus this research explores the application of network flow-based behavioural modelling to facilitate the binary classification of bot network activity, whereby the detection is independent of underlying communications architectures, ports, protocols and payload-based detection evasion mechanisms. A comparative evaluation of various machine learning classification methods is conducted, to precisely determine the average accuracy of each classifier on bot datasets like CTU-13, ISOT 2010 and ISCX 2014. Additionally, hyperparameter tuning using Genetic Algorithm (GA), aiming to efficiently converge to the fittest hyperparameter set for each dataset was done. The bioinspired optimisation of Random Forest (RF) with GA achieved an average accuracy of 99.85% when it was tested against the three datasets. The model was then developed into a software product. The YouTube link of the project and demo of the software developed: https://youtu.be/gNQjC91VtOI, Comment: 24 pages

The RNA World hypothesis predicts that self-replicating RNAs evolved before DNA genomes and coded proteins. Despite widespread support for the RNA World, self-replicating RNAs have yet to be identified in a natural context, leaving a key 'missing link' for this explanation of the origin of life. Inspired by recent work showing that condensates of charged polymers can create electrochemical gradients capable of catalyzing hydrolysis, we consider a catalytic RNA condensate as a candidate for the self-replicating RNA. We develop a theoretical framework where an RNA condensate formed by the spontaneous demixing of disordered RNA sequences undergoes self-replicative amplification. Our theory addresses two central problems in the origins of life: (i) the origin of compartmentalization and (ii) the error threshold for the accuracy of templated replication. We show that many of the needed properties of this self-replicating RNA condensate have been realized experimentally in recent studies and can be formalized within a standard polymer physics framework. Specifically, we propose that short, low-complexity RNA polymers formed catalytic condensates capable of templated RNA polymerization. Because the condensate properties depend on the RNA sequences, RNAs that formed condensates with improved polymerization and demixing capacity would be amplified, leading to a 'condensate chain reaction' and evolution by natural selection. We believe this prediction could be tested with current experimental and theoretical tools. Furthermore, we note that the extant nucleolus appears to satisfy many of the requirements of an evolutionary relic for the model we propose. More generally, we suggest that future work on the origin of life would benefit from condensate-centric biophysical models of RNA evolution., Comment: 12 pages, 2 figures, 1 table

This paper investigates the suitability of frontier Large Language Models (LLMs) for Q&A interactions in science centres, with the aim of boosting visitor engagement while maintaining factual accuracy. Using a dataset of questions collected from the National Space Centre in Leicester (UK), we evaluated responses generated by three leading models: OpenAI's GPT-4, Claude 3.5 Sonnet, and Google Gemini 1.5. Each model was prompted for both standard and creative responses tailored to an 8-year-old audience, and these responses were assessed by space science experts based on accuracy, engagement, clarity, novelty, and deviation from expected answers. The results revealed a trade-off between creativity and accuracy, with Claude outperforming GPT and Gemini in both maintaining clarity and engaging young audiences, even when asked to generate more creative responses. Nonetheless, experts observed that higher novelty was generally associated with reduced factual reliability across all models. This study highlights the potential of LLMs in educational settings, emphasizing the need for careful prompt engineering to balance engagement with scientific rigor., Comment: 19 pages, 2 figures, 10 tables

We measure the two-dimensional elastic modulus $E^\text{2D}$ of atomically clean defect-engineered graphene with a known defect distribution and density in correlated ultra-high vacuum experiments. The vacancies are introduced via low-energy (< 200 eV) Ar ion irradiation and the atomic structure is obtained via semi-autonomous scanning transmission electron microscopy and image analysis. Based on atomic force microscopy nanoindentation measurements, a decrease of $E^\text{2D}$ from 286 to 158 N/m is observed when measuring the same graphene membrane before and after an ion irradiation-induced vacancy density of $1.0\times 10^{13}$ cm$^{-2}$. This decrease is significantly greater than what is predicted by most theoretical studies and in stark contrast to some measurements presented in the literature. With the assistance of atomistic simulations, we show that this softening is mostly due to corrugations caused by local strain at vacancies with two or more missing atoms, while the influence of single vacancies is negligible. We further demonstrate that the opposite effect can be measured when surface contamination is not removed before defect engineering, Comment: 16 pages, 10 figures