Search

A near-minimal instance of optical cooling is experimentally presented wherein the internal-state entropy of a single atom is reduced more than twofold by illuminating it with broadband, incoherent light. Since the rate of optical pumping by a thermal state increases monotonically with its temperature, the cooling power in this scenario increases with higher thermal occupation, an example of a phenomenon known as cooling by heating. In contrast to optical pumping by coherent, narrow-band laser light, here we perform the same task with fiber-coupled, broadband sunlight, the brightest laboratory-accessible source of continuous blackbody radiation., Comment: 4 pages

Linked micromaps were originally developed to display geographically indexed statistics in an intuitive way by connecting them to a sequence of small maps. The approach integrates several visualization design principles, such as small multiples, discrete color indexing, and ordering. Linked micromaps allow for other types of data displays that are connected to and conditional on geographic areas. Initial applications of micromaps used data from the National Cancer Institute and the Environmental Protection Agency. In this paper, we will show how linked micromaps can be used to better understand and explore relationships and distributions of statistics linked to US states and Washington, DC. We will compare linked micromaps with other popular data displays of geographic data, such as bubble maps, choropleth maps, and bar charts. We will illustrate how linked micromaps can be used for evidence-based decision-making using data from the Bureau of Labor Statistics, the Census Bureau, and the Economic Research Service. The presentations, R scripts, and the data sets used in this article are available here: https://github.com/wlmcensus/Joint-Statistical-Meetings-Presentation-2024. The work discussed in this article was presented at the Joint Statistical Meetings (JSM) 2024 and the American Association for Public Opinion Research (AAPOR) 2024 Annual Conference., Comment: 18 pages, 13 figures, JSM 2024 Conference Proceedings

Fun acts as a catalyst for learning by enhancing motivation, active engagement and knowledge retention. As social robots gain traction as educational tools, understanding how their unique affordances can be leveraged to cultivate fun becomes crucial. This research investigates the concept of fun in educational games involving social robots to support the design of REMind:a robot-mediated role-play game aimed at encouraging bystander intervention against peer bullying among children. To incorporate fun elements into design of REMind, we conducted a user-centered Research through Design (RtD) study with focus groups of children to gain a deeper understanding of their perceptions of fun. We analyzed children's ideas by using Framework Analysis and leveraging LeBlanc's Taxonomy of Game Pleasures and identified 28 elements of fun that can be incorporated into robot-mediated games. We present our observations, discuss their impact on REMind's design, and offer recommendations for designing fun educational games using social robots., Comment: This is a pre-print of a manuscript that was accepted to International Conference on Social Robotics 2024 (ICSR'24 + AI), 2024, which was held in Odense, Denmark

Hierarchical sorting is a fundamental task for programmable matter, inspired by the spontaneous formation of interfaces and membranes in nature. The task entails particles of different types, present in fixed densities, sorting into corresponding regions of a space that are themselves organized. By analyzing the Gibbs distribution of a general fixed-magnetization model of equilibrium statistical mechanics, we prove that particles moving stochastically according to local affinities solve the hierarchical sorting task. The analysis of fixed-magnetization models is notoriously difficult, and approaches that have led to recent breakthroughs in sampling the low-temperature regime only work in the variable-magnetization setting by default. To overcome this barrier, we introduce a new approach for comparing the partition functions of fixed- and variable-magnetization models. The core technique identifies a special class of configurations that contribute comparably to the two partition functions, which then serves as a bridge between the fixed- and variable-magnetization settings. Our main result is an estimate of the Gibbs distribution that unifies existing and new results for models at fixed magnetization, including the Ising, Potts, and Blume--Capel models, and leads to stochastic distributed algorithms for hierarchical sorting and other self-organizing tasks, like compression and separation.

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

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.

Recently, 3D Gaussian Splatting (3DGS) has emerged as a significant advancement in 3D scene reconstruction, attracting considerable attention due to its ability to recover high-fidelity details while maintaining low complexity. Despite the promising results achieved by 3DGS, its rendering performance is constrained by its dependence on costly non-commutative alpha-blending operations. These operations mandate complex view dependent sorting operations that introduce computational overhead, especially on the resource-constrained platforms such as mobile phones. In this paper, we propose Weighted Sum Rendering, which approximates alpha blending with weighted sums, thereby removing the need for sorting. This simplifies implementation, delivers superior performance, and eliminates the "popping" artifacts caused by sorting. Experimental results show that optimizing a generalized Gaussian splatting formulation to the new differentiable rendering yields competitive image quality. The method was implemented and tested in a mobile device GPU, achieving on average $1.23\times$ faster rendering.

We investigate the different transport mechanisms that can occur in atomically precise advanced-manufacturing (APAM) pn junction devices at cryogenic and room temperatures. We first elucidate the potential cause of the anomalous behaviors observed in the forward-bias response of these devices in recent cryogenic temperature measurements, which deviates from the theoretical response of a silicon Esaki diode. Specifically, the suppression of the tunneling current at low bias and the appearance of regular diode current at lower biases than theoretically expected for silicon. We find that the latter can be attributed to modifications in the electronic band structure within the $\delta$-layer regions, leading to band-gap narrowing induced by the high density of dopants. We also find that a combination of two sets of band-to-band tunneling (BTBT) parameters can qualitatively approximate the shape of the tunneling current at low bias. This can arise from band quantization and realignment due to the strong potential confinement in APAM-doped layers. Finally, we extend our analyses to room temperature operation, and we predict that trap-assisted tunneling (TAT) may become significant, leading to a complex superposition of BTBT and TAT transport mechanisms in the electrical measurements.

Large Language Models (LLMs) are widely deployed in real-world applications, yet little is known about their training dynamics at the token level. Evaluation typically relies on aggregated training loss, measured at the batch level, which overlooks subtle per-token biases arising from (i) varying token-level dynamics and (ii) structural biases introduced by hyperparameters. While weight decay is commonly used to stabilize training, we reveal that it silently introduces performance biases detectable only at the token level. In fact, we empirically show across different dataset sizes, model architectures and sizes ranging from 270M to 3B parameters that as weight decay increases, low-frequency tokens are disproportionately depreciated. This is particularly concerning, as these neglected low-frequency tokens represent the vast majority of the token distribution in most languages, calling for novel regularization techniques that ensure fairness across all available tokens.

Low-Rank Adaptation (LoRA) is the bread and butter of Large Language Model (LLM) finetuning. LoRA learns an additive low-rank perturbation, $AB$, of a pretrained matrix parameter $W$ to align the model to a new task or dataset with $W+AB$. We identify three core limitations to LoRA for finetuning--a setting that employs limited amount of data and training steps. First, LoRA employs Dropout to prevent overfitting. We prove that Dropout is only suitable for long training episodes but fails to converge to a reliable regularizer for short training episodes. Second, LoRA's initialization of $B$ at $0$ creates a slow training dynamic between $A$ and $B$. That dynamic is also exacerbated by Dropout that further slows the escape from $0$ for $B$ which is particularly harmful for short training episodes. Third, the scaling factor multiplying each LoRA additive perturbation creates ``short-sighted'' interactions between the LoRA modules of different layers. Motivated by principled analysis of those limitations, we find an elegant solution: a Dropout-free, scaling-free, LoRA with Adaptive Learning rate--coined ALLoRA. By scaling the per sample and per parameter gradients with a coefficient inversely proportional to parameters' $\ell_2$ norm, ALLoRA alleviates those three limitations. As a by-product, ALLoRA removes two hyper-parameters from LoRA: the scaling factor and the dropout rate. Empirical results show that ALLoRA admits better accuracy than LoRA on various settings, including against recent LoRA variants such as Weight-Decomposed Low-Rank Adaptation (DoRA). Ablation studies show our solution is the optimal in a family of weight-dependent / output-dependent approaches on various LLMs including the latest Llama3.

In this work, we investigate the topological properties of knotted defects in smectic liquid crystals. Our story begins with screw dislocations, whose radial surface structure can be smoothly accommodated on $S^3$ for fibred knots by using the corresponding knot fibration. To understand how a smectic texture may take on a screw dislocation in the shape of a knot without a fibration, we study first knotted edge defects. Unlike screw defects, knotted edge dislocations force singular points in the system for any non-trivial knot. We provide a lower bound on the number of such point defects required for a given edge dislocation knot and draw an analogy between the point defect structure of knotted edge dislocations and that of focal conic domains. By showing that edge dislocations, too, are sensitive to knot fibredness, we reinterpret the so-called Morse-Novikov points required for non-fibred screw dislocation knots as analogous smectic defects. Our methods are then applied to $+1/2$ and negative-charge disclinations in the smectic phase, furthering the analogy between knotted smectic defects and focal conic domains and uncovering an intricate relationship between point and line defects in smectic liquid crystals. The connection between smectic defects and knot theory not only unravels the uniquely topological knotting of smectic defects but also provides a mathematical and experimental playground for modern questions in knot and Morse-Novikov theory.

Accurate anomaly detection is critical in vision-based infrastructure inspection, where it helps prevent costly failures and enhances safety. Self-Supervised Learning (SSL) offers a promising approach by learning robust representations from unlabeled data. However, its application in anomaly detection remains underexplored. This paper addresses this gap by providing a comprehensive evaluation of SSL methods for real-world anomaly detection, focusing on sewer infrastructure. Using the Sewer-ML dataset, we evaluate lightweight models such as ViT-Tiny and ResNet-18 across SSL frameworks, including BYOL, Barlow Twins, SimCLR, DINO, and MAE, under varying class imbalance levels. Through 250 experiments, we rigorously assess the performance of these SSL methods to ensure a robust and comprehensive evaluation. Our findings highlight the superiority of joint-embedding methods like SimCLR and Barlow Twins over reconstruction-based approaches such as MAE, which struggle to maintain performance under class imbalance. Furthermore, we find that the SSL model choice is more critical than the backbone architecture. Additionally, we emphasize the need for better label-free assessments of SSL representations, as current methods like RankMe fail to adequately evaluate representation quality, making cross-validation without labels infeasible. Despite the remaining performance gap between SSL and supervised models, these findings highlight the potential of SSL to enhance anomaly detection, paving the way for further research in this underexplored area of SSL applications.

The 3D reconstruction of the surgical field in minimally invasive endoscopic surgery has posed a formidable challenge when using conventional monocular endoscopes. Existing 3D reconstruction methodologies are frequently encumbered by suboptimal accuracy and limited generalization capabilities. In this study, we introduce an innovative pipeline using Neural Radiance Fields (NeRF) for 3D reconstruction. Our approach utilizes a preliminary NeRF reconstruction that yields a coarse model, then creates a binocular scene within the reconstructed environment, which derives an initial depth map via stereo vision. This initial depth map serves as depth supervision for subsequent NeRF iterations, progressively refining the 3D reconstruction with enhanced accuracy. The binocular depth is iteratively recalculated, with the refinement process continuing until the depth map converges, and exhibits negligible variations. Through this recursive process, high-fidelity depth maps are generated from monocular endoscopic video of a realistic cranial phantom. By repeated measures of the final 3D reconstruction compared to X-ray computed tomography, all differences of relevant clinical distances result in sub-millimeter accuracy.

The Gemini Planet Imager (GPI) is a high contrast imaging instrument that aims to detect and characterize extrasolar planets. GPI is being upgraded to GPI 2.0, with several subsystems receiving a re-design to improve its contrast. To enable observations on fainter targets and increase performance on brighter ones, one of the upgrades is to the adaptive optics system. The current Shack-Hartmann wavefront sensor (WFS) is being replaced by a pyramid WFS with an low-noise electron multiplying CCD (EMCCD). EMCCDs are detectors capable of counting single photon events at high speed and high sensitivity. In this work, we characterize the performance of the HN\"u 240 EMCCD from N\"uv\"u Cameras, which was custom-built for GPI 2.0. Through our performance evaluation we found that the operating mode of the camera had to be changed from inverted-mode (IMO) to non-inverted mode (NIMO) in order to improve charge diffusion features found in the detector's images. Here, we characterize the EMCCD's noise contributors (readout noise, clock-induced charges, dark current) and linearity tests (EM gain, exposure time) before and after the switch to NIMO., Comment: Proceeding of the SPIE Astronomical Telescopes+Instrumentation. 14 pages, 15 figures

This paper evaluates the performance of the one-shot Semi-Persistent Scheduling (SPS) mechanism in Cellular Vehicle-to-Everything (C-V2X) networks under Denial-of-Service (DoS) smart attack scenarios. The study focuses on the impact of these attacks on key performance metrics, including Packet Delivery Ratio (PDR), Inter-Packet Gap (IPG), and Age of Information (AoI). Through extensive Monte Carlo simulations, we demonstrate that the one-shot mechanism significantly enhances network resilience by mitigating the adverse effects of smart DoS attacks. The findings reveal that while the one-shot mechanism improves the PDR and reduces the IPG and AoI tail values, its effectiveness diminishes slightly in high-density vehicular environments. Nevertheless, the one-shot mechanism proves to be a robust solution for maintaining the stability and reliability of C-V2X communications under adversarial conditions.

As highlighted in the National Spectrum Strategy, Dynamic Spectrum Access (DSA) is key for enabling 6G networks to meet the increasing demand for spectrum from various, heterogeneous emerging applications. In this paper, we consider heterogeneous wireless networks with multiple 6G base stations (BS) and a limited number of frequency bands available for transmission. Each BS is associated with a geographical location, a coverage area, and a bandwidth requirement. We assume that clients/UEs are within the corresponding BS's coverage area. To avoid interference, we impose that BSs with overlapping coverage areas must use different frequency bands. We address the challenging problem of efficiently allocating contiguous frequency bands to BSs while avoiding interference. Specifically, we define performance metrics that capture the feasibility of the frequency allocation task, the number of BSs that can be allocated within the limited frequency bands, and the amount of resources utilized by the network. Then, we consider five different DSA algorithms that prioritize BSs based on different features - one of these algorithms is known in the graph theory literature as Welsh-Powell graph colouring algorithm - and compare their performance using extensive simulations. Our results show that DSA algorithms that attempt to maximize the chances of obtaining a feasible frequency allocation - which have been widely studied in the literature - tend to under-perform in all other metrics.

Observations of the 21 cm signal face significant challenges due to bright astrophysical foregrounds that are several orders of magnitude higher than the brightness of the hydrogen line, along with various systematics. Successful 21 cm experiments require accurate calibration and foreground mitigation. Errors introduced during the calibration process such as systematics, can disrupt the intrinsic frequency smoothness of the foregrounds, leading to power leakage into the Epoch of Reionisation (EoR) window. Therefore, it is essential to develop strategies to effectively address these challenges. In this work, we adopt a stringent approach to identify and address suspected systematics, including malfunctioning antennas, frequency channels corrupted by radio frequency interference (RFI), and other dominant effects. We implement a statistical framework that utilises various data products from the data processing pipeline to derive specific criteria and filters. These criteria and filters are applied at intermediate stages to mitigate systematic propagation from the early stages of data processing. Our analysis focuses on observations from the Murchison Widefield Array (MWA) Phase I configuration. Out of the observations processed by the pipeline, our approach selects 18%, totalling 58 hours, that exhibit fewer systematic effects. The successful selection of observations with reduced systematic dominance enhances our confidence in achieving 21 cm measurements., Comment: 18 pages, 17 figures, 2 tables (accepted for publication in PASA)

X-ray diffraction gratings play an essential role in high-resolution spectroscopy of astrophysical phenomena. We present some scientific highlights from the X-ray grating spectrometers (XGS) on board of the Chandra and XMM/Newton missions, XGS optical design, and the basic physics of grating diffraction geometry and efficiency. We review design, fabrication, and performance of the currently orbiting transmission and reflection grating elements, followed by descriptions of the state-of-the art of more advanced grating technologies that promise orders-of-magnitude improvements in XGS performance, especially in combination with advanced X-ray telescope mirrors. A few key science questions that require new grating technology are posed, and powerful future mission concepts and recent and approved missions are presented., Comment: 40 pages, 12 figures. To be submitted to Springer for publication in the ISSI Scientific Reports series

This work presents the INBD network proposed by Gillert et al. in CVPR-2023 and studies its application for delineating tree rings in RGB images of Pinus taeda cross sections captured by a smartphone (UruDendro dataset), which are images with different characteristics from the ones used to train the method. The INBD network operates in two stages: first, it segments the background, pith, and ring boundaries. In the second stage, the image is transformed into polar coordinates, and ring boundaries are iteratively segmented from the pith to the bark. Both stages are based on the U-Net architecture. The method achieves an F-Score of 77.5, a mAR of 0.540, and an ARAND of 0.205 on the evaluation set. The code for the experiments is available at https://github.com/hmarichal93/mlbrief_inbd., Comment: Submitted to IPOL ad an MLBriefs paper

Local interactions of uncoordinated individuals produce the collective behaviors of many biological systems, inspiring much of the current research in programmable matter. A striking example is the spontaneous assembly of fire ants into "bridges" comprising their own bodies to traverse obstacles and reach sources of food. Experiments and simulations suggest that, remarkably, these ants always form one bridge -- instead of multiple, competing bridges -- despite a lack of central coordination. We argue that the reliable formation of a single bridge does not require sophistication on behalf of the individuals by provably reproducing this behavior in a self-organizing particle system. We show that the formation of a single bridge by the particles is a statistical inevitability of their preferences to move in a particular direction, such as toward a food source, and their preference for more neighbors. Two parameters, $\eta$ and $\beta$, reflect the strengths of these preferences and determine the Gibbs stationary measure of the corresponding particle system's Markov chain dynamics. We show that a single bridge almost certainly forms when $\eta$ and $\beta$ are sufficiently large. Our proof introduces an auxiliary Markov chain, called an "occupancy chain," that captures only the significant, global changes to the system. Through the occupancy chain, we abstract away information about the motion of individual particles, but we gain a more direct means of analyzing their collective behavior. Such abstractions provide a promising new direction for understanding many other systems of programmable matter.

GQ Lup B is one of the few substellar companions with a detected cicumplanetary disk, or CPD. Observations of the CPD suggest the presence of a cavity, possibly formed by an exo-satellite. Using the Keck Planet Imager and Characterizer (KPIC), a high contrast imaging suite that feeds a high resolution spectrograph (1.9-2.5 microns, R$\sim$35,000), we present the first dedicated radial velocity (RV) observations around a high-contrast, directly imaged substellar companion, GQ Lup B, to search for exo-satellites. Over 11 epochs, we find a best and median RV error of 400-1000 m/s, most likely limited by systematic fringing in the spectra due to transmissive optics within KPIC. With this RV precision, KPIC is sensitive to exomoons 0.6-2.8% the mass of GQ Lup B ($\sim 30 M_{\text{Jup}}$) at separations between the Roche limit and $65 R_{\text{Jup}}$, or the extent of the cavity inferred within the CPD detected around GQ Lup B. Using simulations of HISPEC, a high resolution infrared spectrograph planned to debut at W.M. Keck Observatory in 2026, we estimate future exomoon sensitivity to increase by over an order of magnitude, providing sensitivity to less massive satellites potentially formed within the CPD itself. Additionally, we run simulations to estimate the amount of material that different masses of satellites could clear in a CPD to create the observed cavity. We find satellite-to-planet mass ratios of $q > 2 \times 10^{-4}$ can create observable cavities and report a maximum cavity size of $\sim 51 \, R_{\text{Jup}}$ carved from a satellite., Comment: 15 pages, 5 figures

The Keck Planet Imager and Characterizer (KPIC) combines high contrast imaging with high resolution spectroscopy (R$\sim$35,000 in K band) to study directly imaged exoplanets and brown dwarfs in unprecedented detail. KPIC aims to spectrally characterize substellar companions through measurements of planetary radial velocities, spins, and atmospheric composition. Currently, the dominant source of systematic noise for KPIC is fringing, or oscillations in the spectrum as a function of wavelength. The fringing signal can dominate residuals by up to 10% of the continuum for high S/N exposures, preventing accurate wavelength calibration, retrieval of atmospheric parameters, and detection of planets with flux ratios less than 1% of the host star. To combat contamination from fringing, we first identify its three unique sources and adopt a physically informed model of Fabry-P\'{e}rot cavities to apply to post-processed data. We find this strategy can effectively model the fringing in observations of A0V/F0V stars, reducing the residual systematics caused by fringing by a factor of 2. Next, we wedge two of the transmissive optics internal to KPIC to eliminate two sources of fringing and confirm the third source as the entrance window to the spectrograph. Finally, we apply our previous model of the Fabry-P\'{e}rot cavity to new data taken with the wedged optics to reduce the amplitude of the residuals by a factor of 10., Comment: 13 pages, 6 figures

Knitting turns a one dimensional yarn into a highly ramified three-dimensional structure. As a method of additive manufacturing, it holds promise for a new class of lightweight, ultrastrong materials. Here we present a purely geometric model to predict the three-dimensional folding of knitted fabrics made only of the two traditional stitches, knit and purl., Comment: 7 pages, 3 figures, the whole megillah

Aerosol interactions with clouds represent a significant uncertainty in our understanding of the Earth system. Deep convective clouds may respond to aerosol perturbations in several ways that have proven difficult to elucidate with observations. Here, we leverage the two busiest maritime shipping lanes in the world, which emit aerosol particles and their precursors into an otherwise relatively clean tropical marine boundary layer, to make headway on the influence of aerosol on deep convective clouds. The recent seven-fold change in allowable fuel sulfur by the International Maritime Organization allows us to test the sensitivity of the lightning to changes in ship plume aerosol size distributions. We find that, across a range of atmospheric thermodynamic conditions, the previously documented enhancement of lightning over the shipping lanes has fallen by over 40%. The enhancement is therefore at least partially aerosol-mediated, a conclusion that is supported by observations of droplet number at cloud base, which show a similar decline over the shipping lane. These results have fundamental implications for our understanding of aerosol-cloud interactions, suggesting that deep convective clouds are impacted by the aerosol number distribution in the remote marine environment.

Significant research efforts have been made to scale and improve vision-language model (VLM) training approaches. Yet, with an ever-growing number of benchmarks, researchers are tasked with the heavy burden of implementing each protocol, bearing a non-trivial computational cost, and making sense of how all these benchmarks translate into meaningful axes of progress. To facilitate a systematic evaluation of VLM progress, we introduce UniBench: a unified implementation of 50+ VLM benchmarks spanning a comprehensive range of carefully categorized capabilities from object recognition to spatial awareness, counting, and much more. We showcase the utility of UniBench for measuring progress by evaluating nearly 60 publicly available vision-language models, trained on scales of up to 12.8B samples. We find that while scaling training data or model size can boost many vision-language model capabilities, scaling offers little benefit for reasoning or relations. Surprisingly, we also discover today's best VLMs struggle on simple digit recognition and counting tasks, e.g. MNIST, which much simpler networks can solve. Where scale falls short, we find that more precise interventions, such as data quality or tailored-learning objectives offer more promise. For practitioners, we also offer guidance on selecting a suitable VLM for a given application. Finally, we release an easy-to-run UniBench code-base with the full set of 50+ benchmarks and comparisons across 59 models as well as a distilled, representative set of benchmarks that runs in 5 minutes on a single GPU.

In this paper, we overview one promising avenue of progress at the mathematical foundation of deep learning: the connection between deep networks and function approximation by affine splines (continuous piecewise linear functions in multiple dimensions). In particular, we will overview work over the past decade on understanding certain geometrical properties of a deep network's affine spline mapping, in particular how it tessellates its input space. As we will see, the affine spline connection and geometrical viewpoint provide a powerful portal through which to view, analyze, and improve the inner workings of a deep network.

The combined influence of urban heat islands, climate change, and extreme temperature events are increasingly impacting transit travelers, especially vulnerable populations such as older adults, people with disabilities, and those with chronic diseases. Previous studies have generally attempted to address this issue at either the micro- or macro-level, but each approach presents different limitations in modeling the impacts on transit trips. Other research proposes a meso-level approach to address some of these gaps, but the use of additive exposure calculation and spatial shortest path routing poses constraints meso-modeling accuracy. This study introduces HeatPath Analyzer, a framework to assess the exposure of transit riders to extreme temperatures, using TransitSim 4.0 to generate second-by-second spatio-temporal trip trajectories, the traveler activity profiles, and thermal comfort levels along the entire journey. The approach uses heat stress combines the standards proposed by the NWS and CDC to estimate cumulative exposure for transit riders, with specific parameters tailored to the elderly and people with disabilities. The framework assesses the influence of extreme heat and winter chill. A case study in Atlanta, GA, reveals that 10.2% of trips on an average summer weekday in 2019 were at risk of extreme heat. The results uncover exposure disparities across different transit trip mode segments, and across mitigation-based and adaptation-based strategies. While the mitigation-based strategy highlights high-exposure segments such as long ingress and egress, adaptation should be prioritized toward the middle or second half of the trip when a traveler is waiting for transit or transferring between routes. A comparison between the traditional additive approach and the dynamic approach presented also shows significant disparities, which, if overlooked, can mislead policy decisions., Comment: 44 pages, 1 table, 8 figures

Transit is a crucial mode of transportation, especially in urban areas and for urban and rural disadvantaged communities. Because extreme temperatures often pose threats to the elderly, members of the disability community, and other vulnerable populations, this study seeks to understand the level of influence that extreme temperatures may have on transit users across different demographic groups. In this case study for Atlanta, GA, heat stress is predicted for 2019 transit riders (using transit rider activity survey data) and for three future climate scenarios, SSP245, SSP370, and SSP585, into the year 2100. The HeatPath Analyzer and TransitSim 4.0 models were applied to predict cumulative heat exposure and trip-level risk for 35,999 trip equivalents for an average Atlanta area weekday in the summer of 2019. The analyses show that under 2019 weather conditions, 8.33% of summer trips were estimated to be conducted under extreme heat. With the projected future climate conditions, the percentage of trips under extreme heat risk grows steadily. By 2100, 37.1%, 56.1%, and 76.4% are projected to be under extreme heat risk for scenarios SSP245, SSP370, and SSP585, respectively. Under current weather conditions, Atlanta transit riders that own no vehicles and transit riders that are African American are disproportionately influenced by extreme heat. The disparity between these two groups and other groups of transit riders becomes wider as climate change continues to exacerbate. The findings of the study highlight an urgent need to implement heat mitigation and adaptation strategies in urban transit networks., Comment: 24 pages, 1 table, 4 figures

Young, self-luminous super-Jovian companions discovered by direct imaging provide a challenging test of planet formation and evolution theories. By spectroscopically characterizing the atmospheric compositions of these super-Jupiters, we can constrain their formation histories. Here we present studies of the recently discovered HIP 99770 b, a 16 MJup high-contrast companion on a 17 au orbit, using the fiber-fed high-resolution spectrograph KPIC (R~35,000) on the Keck II telescope. Our K-band observations led to detections of H2O and CO in the atmosphere of HIP 99770 b. We carried out free retrieval analyses using petitRADTRANS to measure its chemical abundances, including the metallicity and C/O ratio, projected rotation velocity (vsini), and radial velocity (RV). We found that the companion's atmosphere has C/O=0.55(-0.04/+0.06) and [M/H]=0.26(-0.23/+0.24) (1{\sigma} confidence intervals), values consistent with those of the Sun and with a companion formation via gravitational instability or core accretion. The projected rotation velocity < 7.8 km/s is small relative to other directly imaged companions with similar masses and ages. This may imply a near pole-on orientation or effective magnetic braking by a circumplanetary disk. In addition, we added the companion-to-primary relative RV measurement to the orbital fitting and obtained updated constraints on orbital parameters. Detailed characterization of super-Jovian companions within 20 au like HIP 99770 b is critical for understanding the formation histories of this population., Comment: 18 pages, 5 figures, accepted to AJ

With the increasing demand for wireless services, spectrum management agencies and service providers (SPs) are seeking more flexible mechanisms for spectrum sharing to accommodate this growth. Such mechanisms impact the market dynamics of competitive SPs. Prior market models of spectrum sharing largely focus on scenarios where competing SPs had identical coverage areas. We depart from this and consider a scenario in which two competing SPs have overlapping but distinct coverage areas. We study the resulting competition using a Cournot model. Our findings reveal that with limited shared bandwidth, SPs might avoid overlapping areas to prevent potential losses due to interference. Sometimes SPs can strategically cooperate by agreeing not to provide service in the overlapping areas and, surprisingly, customers might also benefit from such cooperation under certain circumstances. Overall, market outcomes exhibit complex behaviors that are influenced by the sizes of coverage areas and the bandwidth of the shared spectrum.

Learning good representations involves capturing the diverse ways in which data samples relate. Contrastive loss - an objective matching related samples - underlies methods from self-supervised to multimodal learning. Contrastive losses, however, can be viewed more broadly as modifying a similarity graph to indicate how samples should relate in the embedding space. This view reveals a shortcoming in contrastive learning: the similarity graph is binary, as only one sample is the related positive sample. Crucially, similarities \textit{across} samples are ignored. Based on this observation, we revise the standard contrastive loss to explicitly encode how a sample relates to others. We experiment with this new objective, called $\mathbb{X}$-Sample Contrastive, to train vision models based on similarities in class or text caption descriptions. Our study spans three scales: ImageNet-1k with 1 million, CC3M with 3 million, and CC12M with 12 million samples. The representations learned via our objective outperform both contrastive self-supervised and vision-language models trained on the same data across a range of tasks. When training on CC12M, we outperform CLIP by $0.6\%$ on both ImageNet and ImageNet Real. Our objective appears to work particularly well in lower-data regimes, with gains over CLIP of $16.8\%$ on ImageNet and $18.1\%$ on ImageNet Real when training with CC3M. Finally, our objective seems to encourage the model to learn representations that separate objects from their attributes and backgrounds, with gains of $3.3$-$5.6$\% over CLIP on ImageNet9. We hope the proposed solution takes a small step towards developing richer learning objectives for understanding sample relations in foundation models.

The evolution of smart home technologies, particularly agentic ones such as conversational agents, robots, and virtual avatars, is reshaping our understanding of home and domestic life. This shift highlights the complexities of modern domestic life, with the household landscape now featuring diverse cohabiting units like co-housing and communal living arrangements. These agentic technologies present specific design challenges and opportunities as they become integrated into everyday routines and activities. Our workshop envisions smart homes as dynamic, user-shaped spaces, focusing on the integration of these technologies into daily life. We aim to explore how these technologies transform household dynamics, especially through boundary fluidity, by uniting researchers and practitioners from fields such as design, sociology, and ethnography. Together, we will develop a taxonomy of challenges and opportunities, providing a structured perspective on the integration of agentic technologies and their impact on contemporary living arrangements., Comment: 13 pages, no figures, ECSCW 2024 Workshop Proposal

Smectic liquid crystals can be viewed as model systems for lamellar structures for which there has been extensive theoretical development. We demonstrate that a nonlinear energy description is required with respect to the usual Landau-de Gennes elasticity in order to explain the observed layer spacing of highly curved smectic layers. Using X-ray diffraction we have quantitatively determined the dilation of bent layers distorted by antagonistic anchoring (as high as 1.8% of dilation for the most bent smectic layers) and accurately described it by the minimal nonlinear expression for energy. We observe a 1{\deg} tilt of planar layers that are connected to the curved layers. This value is consistent with simple energetic calculations, demonstrating how the bending energy impacts the overall structure of a thin distorted smectic film. Finally, we show that combined X-ray measurements and theoretical modeling allow for the quantitative determination of the number of curved smectic layers and of the resulting thickness of the dilated region with unprecedented precision., Comment: 6 pages with 6 figures

The Simons Observatory (SO) is a cosmic microwave background experiment composed of three 0.42 m Small Aperture Telescopes (SATs) and one 6 m Large Aperture Telescope (LAT) in the Atacama Desert of Chile. The Large Aperture Telescope Receiver (LATR) was integrated into the LAT in August 2023; however, because mirrors were not yet installed, the LATR optical chain was capped at the 4K stage. In this dark configuration we are able to characterize many elements of the instrument without contributions from atmospheric noise. Here we show this noise is below the required upper limit and its features are well described with a simple noise model. Maps produced using this noise model have properties that are in good agreement with the white noise levels of our dark data. Additionally, we show that our nominal scan strategy has a minimal effect on the noise when compared to the noise when the telescope is stationary

We undertake a comprehensive investigation into the distribution of insitu stars within Milky Way-like galaxies, leveraging TNG50 simulations and comparing their predictions with data from the H3 survey. Our analysis reveals that 28% of galaxies demonstrate reasonable agreement with H3, while only 12% exhibit excellent alignment in their profiles, regardless of the specific spatial cut employed to define insitu stars. To uncover the underlying factors contributing to deviations between TNG50 and H3 distributions, we scrutinize correlation coefficients among internal drivers(e.g., virial radius, star formation rate [SFR]) and merger-related parameters (such as the effective mass-ratio, mean distance, average redshift, total number of mergers, average spin-ratio and maximum spin alignment between merging galaxies). Notably, we identify significant correlations between deviations from observational data and key parameters such as the median slope of virial radius, mean SFR values, and the rate of SFR change across different redshift scans. Furthermore, positive correlations emerge between deviations from observational data and parameters related to galaxy mergers. We validate these correlations using the Random Forest Regression method. Our findings underscore the invaluable insights provided by the H3 survey in unraveling the cosmic history of galaxies akin to the Milky Way, thereby advancing our understanding of galactic evolution and shedding light on the formation and evolution of Milky Way-like galaxies in cosmological simulations., Comment: 20 pages, 13 figures

We obtained New Horizons LORRI images to measure the cosmic optical background (COB) intensity integrated over $0.4\lesssim\lambda\lesssim0.9{~\rm\mu m}.$ The survey comprises 16 high Galactic-latitude fields selected to minimize scattered diffuse Galactic light (DGL) from the Milky Way galaxy, as well as scattered light from bright stars. This work supersedes an earlier analysis based on observations of one of the present fields. Isolating the COB contribution to the raw total sky levels measured in the fields requires subtracting the remaining scattered light from bright stars and galaxies, intensity from faint stars within the fields fainter than the photometric detection-limit, and the DGL foreground. DGL is estimated from Planck HFI $350 {~\rm\mu m}$ and $550 {~\rm\mu m}$ intensities, using a new self-calibrated indicator based on the 16 fields augmented with eight additional DGL calibration fields obtained as part of the survey. The survey yields a highly significant detection ($6.8\sigma$) of the COB at ${\rm 11.16\pm 1.65~(1.47~sys,~0.75~ran) ~nW ~m^{-2} ~sr^{-1}}$ at the LORRI pivot wavelength of 0.608 $\mu$m. The estimated integrated intensity from background galaxies, ${\rm 8.17\pm 1.18 ~nW ~m^{-2} ~sr^{-1}},$ can account for the great majority of this signal. The rest of the COB signal, ${\rm 2.99\pm2.03~ (1.75~sys,~1.03~ran) ~nW ~m^{-2} ~sr^{-1}},$ is formally classified as anomalous intensity but is not significantly different from zero. The simplest interpretation is that the COB is completely due to galaxies., Comment: Accepted for publication in the Astrophysical Journal; 30 pages, 14 figures. v3 has updated version of fig 14 to show total error value for Symons et al. 2023 result

Although the sequential tsunami scenario detection framework was validated in our previous work, several tasks remain to be resolved from a practical point of view. This study aims to evaluate the performance of the previous tsunami scenario detection framework using a diverse database consisting of complex fault rupture patterns with heterogeneous slip distributions. Specifically, we compare the effectiveness of scenario superposition to that of the previous most likely scenario detection method. Additionally, how the length of the observation time window influences the accuracy of both methods is analyzed. We utilize an existing database comprising 1771 tsunami scenarios targeting the city Westport (WA, U.S.), which includes synthetic wave height records and inundation distributions as the result of fault rupture in the Cascadia subduction zone. The heterogeneous patterns of slips used in the database increase the diversity of the scenarios and thus make it a proper database for evaluating the performance of scenario superposition. To assess the performance, we consider various observation time windows shorter than 15 minutes and divide the database into five testing and learning sets. The evaluation accuracy of the maximum offshore wave, inundation depth, and its distribution is analyzed to examine the advantages of the scenario superposition method over the previous method. We introduce the dynamic time warping (DTW) method as an additional benchmark and compare its results to that of the Bayesian scenario detection method., Comment: 15 pages, 12 figures

The Simons Observatory (SO) is a Cosmic Microwave Background experiment located in the Atacama Desert in Chile. SO consists of three small aperture telescopes (SATs) and one large aperture telescope (LAT) with a total of 60,000 detectors in six frequency bands. As an observatory, SO encompasses hundreds of hardware components simultaneously running at different readout rates, all separate from its 60,000 detectors on-sky and their metadata. We provide an overview of commissioning SO's data acquisition software system for SAT-MF1, the first SAT deployed to the Atacama site. Additionally, we share insights from deploying data access software for all four telescopes, detailing how performance limitations affected data loading and quality investigations, which led to site-compatible software improvements., Comment: 15 pages, 9 figures, 3 tables. Presented at SPIE Astronomical Telescopes + Instrumentation 2024 Second submission: fix typos

The Keck Planet Imager and Characterizer (KPIC), a series of upgrades to the Keck II Adaptive Optics System and Instrument Suite, aims to demonstrate high-resolution spectroscopy of faint exoplanets that are spatially resolved from their host stars. In this paper, we measure KPIC's sensitivity to companions as a function of separation (i.e., the contrast curve) using on-sky data collected over four years of operation. We show that KPIC is able to reach contrasts of $1.3 \times 10^{-4}$ at 90 mas and $9.2 \times 10^{-6}$ at 420 mas separation from the star, and that KPIC can reach planet-level sensitivities at angular separations within the inner working angle of coronagraphic instruments such as GPI and SPHERE. KPIC is also able to achieve more extreme contrasts than other medium-/high-resolution spectrographs that are not as optimized for high-contrast performance. We decompose the KPIC performance budget into individual noise terms and discuss limiting factors. The fringing that results from combining a high-contrast imaging system with a high-resolution spectrograph is identified as an important source of systematic noise. After mitigation and correction, KPIC is able to reach within a factor of 2 of the photon noise limit at separations < 200 mas. At large separations, KPIC is limited by the background noise performance of NIRSPEC., Comment: 16 pages, 6 figures, submitted to the proceedings of SPIE Astronomical Telescopes + Instrumentation 2024, 13096-69

Deep Learning is often depicted as a trio of data-architecture-loss. Yet, recent Self Supervised Learning (SSL) solutions have introduced numerous additional design choices, e.g., a projector network, positive views, or teacher-student networks. These additions pose two challenges. First, they limit the impact of theoretical studies that often fail to incorporate all those intertwined designs. Second, they slow-down the deployment of SSL methods to new domains as numerous hyper-parameters need to be carefully tuned. In this study, we bring forward the surprising observation that--at least for pretraining datasets of up to a few hundred thousands samples--the additional designs introduced by SSL do not contribute to the quality of the learned representations. That finding not only provides legitimacy to existing theoretical studies, but also simplifies the practitioner's path to SSL deployment in numerous small and medium scale settings. Our finding answers a long-lasting question: the often-experienced sensitivity to training settings and hyper-parameters encountered in SSL come from their design, rather than the absence of supervised guidance.

We develop Scalable Latent Exploration Score (ScaLES) to mitigate over-exploration in Latent Space Optimization (LSO), a popular method for solving black-box discrete optimization problems. LSO utilizes continuous optimization within the latent space of a Variational Autoencoder (VAE) and is known to be susceptible to over-exploration, which manifests in unrealistic solutions that reduce its practicality. ScaLES is an exact and theoretically motivated method leveraging the trained decoder's approximation of the data distribution. ScaLES can be calculated with any existing decoder, e.g. from a VAE, without additional training, architectural changes, or access to the training data. Our evaluation across five LSO benchmark tasks and three VAE architectures demonstrates that ScaLES enhances the quality of the solutions while maintaining high objective values, leading to improvements over existing solutions. We believe that new avenues to LSO will be opened by ScaLES ability to identify out of distribution areas, differentiability, and computational tractability. Open source code for ScaLES is available at https://github.com/OmerRonen/scales.

We report the detection of seven new galaxy candidates with redshift $z$ $>$ 11 within the JWST Advanced Deep Extragalactic Survey (JADES) GOODS-S and GOODS-N fields. These new candidates are detected through meticulous analysis of NIRCam photometry in eight filters spanning a wavelength range of 0.8-5.0 $\mu$m. Photometric redshifts of these galaxy candidates are independently measured utilizing spectral energy distribution (SED) fitting techniques using \texttt{EAZY} and \texttt{BAGPIPES} codes, followed by visual scrutiny. Two of these galaxy candidates are located in GOODS-S field, while the remaining five galaxies are located in GOODS-N field. Our analysis reveals that the stellar masses of these galaxies typically range from log $M_{\ast}$/$M_{\odot}$ = 7.75--8.75. Futhermore, these galaxies are typically young with their mass-weighted ages spanning from 80 to 240 Myr. Their specific star formation rates (sSFR), quantified as $\log (\text{sSFR}/\text{Gyr}$), are measured to vary between $\sim 0.95$ to 1.46. These new galaxy candidates offer a robust sample for probing the physical properties of galaxies within the first few hundred Myr of the history of the Universe. We also analyze the relationship between star formation rate (SFR) and stellar mass ($M_\ast$) within our sample. Using linear regression, our analysis yields a slope of $0.71 \pm 0.12$, which we then compare with results from previous studies. Continued investigation through spectroscopic analysis using JWST/NIRSpec is needed to spectroscopically confirm these high-redshift galaxy candidates and investigate further into their physical properties. We plan to follow up on these candidates with future NIRSpec observations., Comment: 12 pages, 10 figures, Submitted to MNRAS