Search

High quality annotations are increasingly a bottleneck in the explosively growing machine learning ecosystem. Scalable evaluation methods that avoid costly annotation have therefore become an important research ambition. Many hope to use strong existing models in lieu of costly labels to provide cheap model evaluations. Unfortunately, this method of using models as judges introduces biases, such as self-preferencing, that can distort model comparisons. An emerging family of debiasing tools promises to fix these issues by using a few high quality labels to debias a large number of model judgments. In this paper, we study how far such debiasing methods, in principle, can go. Our main result shows that when the judge is no more accurate than the evaluated model, no debiasing method can decrease the required amount of ground truth labels by more than half. Our result speaks to the severe limitations of the LLM-as-a-judge paradigm at the evaluation frontier where the goal is to assess newly released models that are possibly better than the judge. Through an empirical evaluation, we demonstrate that the sample size savings achievable in practice are even more modest than what our theoretical limit suggests. Along the way, our work provides new observations about debiasing methods for model evaluation, and points out promising avenues for future work., Comment: 22 pages, 5 figures

Chick sexing, the process of determining the gender of day-old chicks, is a critical task in the poultry industry due to the distinct roles that each gender plays in production. While effective traditional methods achieve high accuracy, color, and wing feather sexing is exclusive to specific breeds, and vent sexing is invasive and requires trained experts. To address these challenges, we propose a novel approach inspired by facial gender classification techniques in humans: facial chick sexing. This new method does not require expert knowledge and aims to reduce training time while enhancing animal welfare by minimizing chick manipulation. We develop a comprehensive system for training and inference that includes data collection, facial and keypoint detection, facial alignment, and classification. We evaluate our model on two sets of images: Cropped Full Face and Cropped Middle Face, both of which maintain essential facial features of the chick for further analysis. Our experiment demonstrates the promising viability, with a final accuracy of 81.89%, of this approach for future practices in chick sexing by making them more universally applicable.

Stage-IV galaxy surveys will measure correlations at small cosmological scales with high signal-to-noise ratio. One of the main challenges of extracting information from small scales is devising accurate models, as well as characterizing the theoretical uncertainties associated with any given model. In this work, we explore the mitigation of theoretical uncertainty due to nonlinear galaxy bias in the context of photometric 2x2-point analyses. We consider linear galaxy bias as the fiducial model and derive the contribution to the covariance matrix induced by neglected higher-order bias. We construct a covariance matrix for the theoretical error in galaxy clustering and galaxy-galaxy lensing using simulation-based relations that connect higher-order parameters to linear bias. To test this mitigation model, we apply the modified likelihood to 2x2-point analyses based on two sets of mock data vectors: (1) simulated data vectors, constructed from those same relations between bias parameters, and (2) data vectors based on the AbacusSummit simulation suite. We then compare the performance of the theoretical-error approach to the commonly employed scale cuts methodology. We find most theoretical-error configurations yield results equivalent to the scale cuts in terms of precision and accuracy, in some cases producing significantly stronger bounds on cosmological parameters. These results are independent of the maximum scale k_max in the analysis with theoretical error. The scenarios where linear bias supplemented by theoretical error is unable to recover unbiased cosmology are connected to inadequate modeling of the gg-gk covariance of theoretical error. In view of its removing the ambiguity in the choice of k_max, as well as the possibility of attaining higher precision than the usual scale cuts, we consider this method to be promising for analyses of LSS in upcoming photometric galaxy surveys., Comment: 30 pages, 9 figures

Mental-health therapy involves a complex conversation flow in which patients and therapists continuously negotiate what should be talked about next. For example, therapists might try to shift the conversation's direction to keep the therapeutic process on track and avoid stagnation, or patients might push the discussion towards issues they want to focus on. How do such patient and therapist redirections relate to the development and quality of their relationship? To answer this question, we introduce a probabilistic measure of the extent to which a certain utterance immediately redirects the flow of the conversation, accounting for both the intention and the actual realization of such a change. We apply this new measure to characterize the development of patient-therapist relationships over multiple sessions in a very large, widely-used online therapy platform. Our analysis reveals that (1) patient control of the conversation's direction generally increases relative to that of the therapist as their relationship progresses; and (2) patients who have less control in the first few sessions are significantly more likely to eventually express dissatisfaction with their therapist and terminate the relationship., Comment: To appear in the Proceedings of EMNLP (Findings) 2024. Code available at https://convokit.cornell.edu

Chess has long been a testbed for AI's quest to match human intelligence, and in recent years, chess AI systems have surpassed the strongest humans at the game. However, these systems are not human-aligned; they are unable to match the skill levels of all human partners or model human-like behaviors beyond piece movement. In this paper, we introduce Allie, a chess-playing AI designed to bridge the gap between artificial and human intelligence in this classic game. Allie is trained on log sequences of real chess games to model the behaviors of human chess players across the skill spectrum, including non-move behaviors such as pondering times and resignations In offline evaluations, we find that Allie exhibits humanlike behavior: it outperforms the existing state-of-the-art in human chess move prediction and "ponders" at critical positions. The model learns to reliably assign reward at each game state, which can be used at inference as a reward function in a novel time-adaptive Monte-Carlo tree search (MCTS) procedure, where the amount of search depends on how long humans would think in the same positions. Adaptive search enables remarkable skill calibration; in a large-scale online evaluation against players with ratings from 1000 to 2600 Elo, our adaptive search method leads to a skill gap of only 49 Elo on average, substantially outperforming search-free and standard MCTS baselines. Against grandmaster-level (2500 Elo) opponents, Allie with adaptive search exhibits the strength of a fellow grandmaster, all while learning exclusively from humans.

We investigate the sensitivity of the 21cm power spectrum from cosmic dawn and the epoch of reionization to models of free-streaming dark radiation (parameterized through $N_{\rm eff}$) and interacting dark radiation-dark matter models (DM-DR). The latter models have gained attention for their potential in addressing recent cosmological tensions and structure formation challenges. We perform a Fisher matrix analysis under different assumptions regarding the astrophysical modeling, and forecast the sensitivity of HERA observations, combined with CMB data from Planck and the Simons Observatory (SO), to $N_{\rm eff}$ and DM-DR interaction modeled using the ETHOS framework assuming a constant scattering rate between the two components. Most importantly, we find that 21cm observations can improve the sensitivity to the DM-DR interaction rate by up to four order of magnitude compared to Planck and SO. Conversely, in the limit of low interaction rate (which asymptotically matches $N_{\rm eff}$), CMB data dominates the constraining power, but the inclusion of HERA data can provide a $\sim 20\%$ improvement in sensitivity over CMB data alone. Moreover, we find that HERA observations will be able to probe a region of the DM-DR interaction parameter space which is promising to explain the weak lensing amplitude `$S_8$' tension. Our results demonstrate the complementarity of 21cm and CMB data in exploring dark sector interactions., Comment: 25+13pages, 11 figures. Comments welcome!

Recent advancements in large language models have significantly expedited the process of generating front-end code. This allows users to rapidly prototype user interfaces and ideate through code, a process known as exploratory programming. However, existing LLM code-generation tools focus more on technical implementation details rather than finding the right design given a particular problem. We present DynEx, an LLM-based method for design exploration in accelerated exploratory programming. DynEx uses LLMs to guide users through a structured Design Matrix to explore the design space before dynamic iterative implementation. It also introduces a technique to self-invoke generative AI, enabling the creation of a diverse suite of applications. A user study of 10 experts found that DynEx increased design exploration and enabled the creation of more complex and varied prototypes compared to a Claude Artifact baseline. We conclude with a discussion of the implications of design exploration for exploratory programming., Comment: 18 pages, 7 figures

We present helium atom micro-diffraction as an ideal technique for characterization of 2D materials due to its ultimate surface sensitivity combined with sub-micron spatial resolution. Thermal energy neutral helium scatters from the valence electron density, 2-3A above the ionic cores of a surface, making the technique ideal for studying 2D materials, where other approaches can struggle due to small interaction cross-sections with few-layer samples. Sub-micron spatial resolution is key development in neutral atom scattering to allow measurements from device-scale samples. We present measurements of monolayer-substrate interactions, thermal expansion coefficients, the electron-phonon coupling constant and vacancy-type defect density on monolayer-MoS2. We also discuss extensions to the presented methods which can be immediately implemented on existing instruments to perform spatial mapping of these material properties., Comment: Draft version, 11 pages, 6 figures, 2 tables

Sulfur vacancy defects mediate a wide range of optoelectronic properties in MoS2, with precise control of defect density allowing for tuneable optoelectronic devices. However, accurate measurement of defect density in monolayer and few-layer samples poses a challenge due to their small scattering cross-sections to photon or electron probes. Conventional lab-based techniques such as Raman and photoluminescence can infer approximate defect density in micro-scale samples via optoelectronic properties, but they require validation using stoichiometric beam-line XPS. We introduce an ultra-low energy (~64 meV) and non-intrusive lab-based technique to quantify the surface defect density in micron-scale monolayer MoS2. Here we show that a recently developed technique, helium atom micro-diffraction (referred to as scanning helium microscopy (SHeM) in literature), can be used to directly measure vacancy-type defect density in 2D materials by performing atom diffraction from a microscopic spot. SHeM uses a neutral, inert, and thermal energy probe of helium-4 atoms to measure ordered and disordered atom-surface scattering allowing the level of surface order to be inferred. The presented method enables rapid, non-damaging, and material-agnostic lab-based quantification of defect density in 2D materials, a crucial step towards the wider adoption of 2D semiconductors in devices., Comment: 15 pages, 7 figures

We present the results of a James Webb Space Telescope (JWST) NIRCam and NIRSpec investigation into the young massive star cluster (YMC) population of NGC 3256, the most cluster-rich luminous infrared galaxy (LIRG) in the Great Observatories All Sky LIRG Survey. We detect 3061 compact YMC candidates with a $S/N \geq 3$ at F150W, F200W, and F335M. Based on yggdrasil stellar population models, we identify 116/3061 sources with F150W - F200W $> 0.47$ and F200W - F355M $> -1.37$ colors suggesting they are young (t $\leq 5$ Myr), dusty ($A_{V} = 5 - 15$), and massive ($M_{\odot} > 10^{5}$). This increases the sample of dust-enshrouded YMCs detected in this system by an order of magnitude relative to previous HST studies. With NIRSpec IFU pointings centered on the northern and southern nucleus, we extract the Pa$\alpha$ and 3.3$\mu$m PAH equivalent widths for 8 bright and isolated YMCs. Variations in both the F200W - F335M color and 3.3$\mu$m PAH emission with the Pa$\alpha$ line strength suggest a rapid dust clearing ($< 3 - 4$ Myr) for the emerging YMCs in the nuclei of NGC 3256. Finally, with both the age and dust emission accurately measured we use yggdrasil to derive the color excess (E(B - V)) for all 8 YMCs. We demonstrate that YMCs with strong 3.3$\mu$m PAH emission (F200W - F335M $> 0$) correspond to sources with E(B - V) $> 3$, which are typically missed in UV-optical studies. This underscores the importance of deep near-infrared imaging for finding and characterizing these very young and dust-embedded sources., Comment: 18 pages, 7 figures, 1 table; Accepted for publication in ApJL

The Matrix Profile (MP), a versatile tool for time series data mining, has been shown effective in time series anomaly detection (TSAD). This paper delves into the problem of anomaly detection in multidimensional time series, a common occurrence in real-world applications. For instance, in a manufacturing factory, multiple sensors installed across the site collect time-varying data for analysis. The Matrix Profile, named for its role in profiling the matrix storing pairwise distance between subsequences of univariate time series, becomes complex in multidimensional scenarios. If the input univariate time series has n subsequences, the pairwise distance matrix is a n x n matrix. In a multidimensional time series with d dimensions, the pairwise distance information must be stored in a n x n x d tensor. In this paper, we first analyze different strategies for condensing this tensor into a profile vector. We then investigate the potential of extending the MP to efficiently find k-nearest neighbors for anomaly detection. Finally, we benchmark the multidimensional MP against 19 baseline methods on 119 multidimensional TSAD datasets. The experiments covers three learning setups: unsupervised, supervised, and semi-supervised. MP is the only method that consistently delivers high performance across all setups.

We present the CO(1-0) maps of 28 infrared-bright galaxies from the Great Observatories All-Sky Luminous Infrared Galaxy Survey (GOALS) taken with the Combined Array for Research in Millimeter Astronomy (CARMA). We detect 100GHz continuum in 16 of 28 galaxies, which trace both active galactic nuclei (AGNs) and compact star-forming cores. The GOALS galaxies show a variety of molecular gas morphologies, though in the majority of cases, the average velocity fields show a gradient consistent with rotation. We fit the full continuum SEDs of each of the source using either MAGPHYS or SED3FIT (if there are signs of an AGN) to derive the total stellar mass, dust mass, and star formation rates of each object. We adopt a value determined from luminous and ultraluminous infrared galaxies (LIRGs and ULIRGs) of $\alpha_{\rm CO}=1.5^{+1.3}_{-0.8}~M_\odot$ (K km s$^{-1}$ pc$^2)^{-1}$, which leads to more physical values for $f_{\rm mol}$ and the gas-to-dust ratio. Mergers tend to have the highest gas-to-dust ratios. We assume the cospatiality of the molecular gas and star formation, and plot the sample on the Schmidt-Kennicutt relation, we find that they preferentially lie above the line set by normal star-forming galaxies. This hyper-efficiency is likely due to the increased turbulence in these systems, which decreases the freefall time compared to star-forming galaxies, leading to "enhanced" star formation efficiency. Line wings are present in a non-negligible subsample (11/28) of the CARMA GOALS sources and are likely due to outflows driven by AGNs or star formation, gas inflows, or additional decoupled gas components., Comment: 29 pages, 4 tables, 11 figures, Accepted by the Astrophysical Journal

We introduce AIRduct, a new array-based intermediate representation designed to support generating efficient code for interactive programs employing multiple cryptographic mechanisms. AIRduct is intended as an IR for the Viaduct compiler, which can synthesize secure, distributed programs with an extensible suite of cryptography. Therefore, AIRduct supports an extensible variety of cryptographic mechanisms, including MPC and ZKP.

Baryonic Acoustic Oscillation (BAO) data from the Dark Energy Spectroscopic Instrument (DESI), in combination with Cosmic Microwave Background (CMB) data and Type Ia Supernovae (SN) luminosity distances, suggests a dynamical evolution of the dark energy equation of state with a phantom phase ($w < -1$) in the past when the so-called $w_0w_a$ parametrization $w(a) = w_0 + w_a(1-a)$ is assumed. In this work, we investigate more general dark energy models that also allow a phantom equation of state. We consider three cases: an equation of state with a transition feature, a model-agnostic equation of state with constant values in chosen redshift bins, and a k-essence model. Since the dark energy equation of state is correlated with neutrino masses, we reassess constraints on the neutrino mass sum focusing on the model-agnostic equation of state. We find that the combination of DESI BAO with Planck 2018 CMB data and SN data from Pantheon, Pantheon+, or Union3 is consistent with an oscillatory dark energy equation of state, while a monotonic behavior is preferred by the DESY5 SN data. Performing model comparison techniques, we find that the $w_0w_a$ parametrization remains the simplest dark energy model that can provide a better fit to DESI BAO, CMB, and all SN datasets than $\Lambda$CDM. Constraints on the neutrino mass sum assuming dynamical dark energy are relaxed compared to $\Lambda$CDM and we show that these constraints are tighter in the model-agnostic case relative to $w_0w_a$ model by $70\%-90\%$., Comment: 36 pages, 12 figures, 4 tables

Cold New Early Dark Energy (Cold NEDE) addresses the Hubble tension through a triggered vacuum phase transition in the dark sector. In this paper, we constrain a phenomenological fluid model using recent cosmic microwave background likelihoods based on Planck NPIPE data alongside baryonic acoustic oscillations and supernovae data from Pantheon+. Exploiting the enhanced constraining power of the datasets, we introduce and study an extended version of the NEDE fluid model in which the equation of state parameter $w_\mathrm{NEDE}$, characterizing the post-phase transition fluid, is allowed to evolve with non-vanishing derivatives ${d}w_\mathrm{NEDE}/d\ln a$ and ${d^2}w_\mathrm{NEDE}/{d}(\ln a)^2$. Our results indicate that data is compatible with a rather simple time dependence that could arise from a mixture of radiation and a stiff fluid. With the updated datasets, the base and extended models still show a significant reduction of the DMAP tension from $6.3 \sigma$ in $\Lambda$CDM down to $3.5\sigma$ with a small simultaneous reduction of the $S_8$ tension, slightly improving over recent findings for the axion-like early dark energy model., Comment: 21 pages, 5 figures

Gait asymmetry, a consequence of various neurological or physical conditions such as aging and stroke, detrimentally impacts bipedal locomotion, causing biomechanical alterations, increasing the risk of falls and reducing quality of life. Addressing this critical issue, this paper introduces a novel diagnostic method for gait symmetry analysis through the use of an assistive robotic Smart Walker equipped with an innovative asymmetry detection scheme. This method analyzes sensor measurements capturing the interaction torque between user and walker. By applying a seasonal-trend decomposition tool, we isolate gait-specific patterns within these data, allowing for the estimation of stride durations and calculation of a symmetry index. Through experiments involving 5 experimenters, we demonstrate the Smart Walker's capability in detecting and quantifying gait asymmetry by achieving an accuracy of 84.9% in identifying asymmetric cases in a controlled testing environment. Further analysis explores the classification of these asymmetries based on their underlying causes, providing valuable insights for gait assessment. The results underscore the potential of the device as a precise, ready-to-use monitoring tool for personalized rehabilitation, facilitating targeted interventions for enhanced patient outcomes., Comment: 7 pages, 5 figures, accepted for the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Abu Dhabi, UAE, 2024

Self-supervised Pretrained Models (PTMs) have demonstrated remarkable performance in computer vision and natural language processing tasks. These successes have prompted researchers to design PTMs for time series data. In our experiments, most self-supervised time series PTMs were surpassed by simple supervised models. We hypothesize this undesired phenomenon may be caused by data scarcity. In response, we test six time series generation methods, use the generated data in pretraining in lieu of the real data, and examine the effects on classification performance. Our results indicate that replacing a real-data pretraining set with a greater volume of only generated samples produces noticeable improvement., Comment: To appear in CIKM 2024 as a short paper; the version here is the self-contained version that includes the non-mandatory supplementary material available on the paper's companion website

We show that the Morse boundary of a Morse local-to-global group is $\sigma$-compact. Moreover, we show that the converse holds for small cancellation groups. As an application, we show that the Morse boundary of a non-hyperbolic, Morse local-to-global group that has contraction does not admit a non-trivial stationary measure. In fact, we show that any stationary measure on a geodesic boundary of such a groups needs to assign measure zero to the Morse boundary. Unlike previous results, we do not need any assumptions on the stationary measures considered., Comment: 36 pages, 9 figures. Comments are welcome!

The `cosmic calibration tension' is a $> 5\sigma$ discrepancy between the cosmological distance ladder built from baryonic acoustic oscillations (BAO) calibrated by the Planck/$\Lambda$CDM sound horizon ($r_s$) and Type Ia supernovae (SN1a) calibrated instead with the S$H_0$ES absolute magnitude, assuming the distance-duality relationship (DDR) holds. In this work, we emphasize the consequences of this tension beyond the value of the Hubble constant $H_0$, and the implications for physics beyond $\Lambda$CDM. Of utmost importance, it implies a larger physical matter density $\omega_m\equiv \Omega_m h^2$, as both the fractional matter density $\Omega_m$ and $h\equiv H_0/100$ km/s/Mpc are well constrained from late-time data. New physics in the pre-recombination era must thus be able to decrease $r_s$ while either reducing the value of $\Omega_m$, or increasing the value of $\omega_m$. Assuming a $\Lambda$CDM-like primordial power spectrum, this necessarily results in an increase in the clustering amplitude $\sigma_8$. Deviations from $\Lambda$CDM in the late-time expansion history cannot resolve the calibrator tension but can help relax the required shifts to the matter density and $\sigma_8$: it is in that sense that a combination of early and late-time new physics may help alleviate the tension. More precisely, models that modify the pre-recombination expansion history can accommodate the increase in $\omega_m$ without the need for additional modifications. It is those models which only affect recombination that require additional deviations at late-times to be successful. Hence, the `cosmic calibration tension' points either to a targeted modification of the pre-recombination expansion history, or to a broader change affecting multiple cosmic epochs., Comment: 12+6pages, 12 figures. Comments welcome!

Cosmological neutrino mass bounds are becoming increasingly stringent. The latest limit within $\Lambda$CDM from Planck 2018+ACT lensing+DESI is $\sum m_\nu < 0.072\,{\rm eV}$ at 95% CL, very close to the minimum possible sum of neutrino masses ($\sum m_\nu > 0.06\,{\rm eV}$), hinting at vanishing or even ''negative'' cosmological neutrino masses. In this context, it is urgent to carefully evaluate the origin of these cosmological constraints. In this paper, we investigate the robustness of these results in three ways: i) we check the role of potential anomalies in Planck CMB and DESI BAO data; ii) we compare the results for frequentist and Bayesian techniques, as very close to physical boundaries subtleties in the derivation and interpretation of constraints can arise; iii) we investigate how deviations from $\Lambda$CDM, potentially alleviating these anomalies, can alter the constraints. From a profile likelihood analysis, we derive constraints in agreement at the $\sim 10\%$ level with Bayesian posteriors. We find that the weak preference for negative neutrino masses is mostly present for Planck 18 data, affected by the well-known `lensing anomaly'. It disappears when the new Planck 2020 HiLLiPoP is used, leading to significantly weaker constraints. Additionally, the pull towards negative masses in DESI data stems from the $z=0.7$ bin, which is in $\sim 3\sigma$ tension with Planck expectations. Without these outliers, and in combination with HiLLiPoP, the bound relaxes to $\sum m_\nu < 0.11\,{\rm eV}$ at 95% CL. The recent preference for dynamical dark energy alleviates this tension and further weakens the bound. As we are at the dawn of a neutrino mass discovery from cosmology, it will be very exciting to see if this trend is confirmed by future data., Comment: 17 pages + Appendices. Comments welcome

The main motivation of the present work is to investigate the asymptotic behavior as $\kappa \to 0+$ of multiradial Schramm-Loewner evolution, SLE$_\kappa$. We show that this process with the common parameterization satisfies a finite-time large deviation principle (LDP) in the Hausdorff metric with non-negative rate function, the multiradial Loewner energy. We also characterize the large-time behavior of curves with finite energy and zero energy (whose driving functions correspond to the trigonometric Calogero-Moser system). The first half of this article is of independent interest regardless of SLE theory. It is devoted to proving a finite-time LDP for Dyson Brownian motion on the circle for a fixed number $n$ of particles as the coupling parameter $\beta=8/\kappa$ tends to $\infty$. To our knowledge, in the literature large deviations of Dyson Brownian motion has only been considered for fixed $\beta$ and as $n$ tends to $\infty$. While the non-Lipschitz drift precludes the application of the Freidlin-Wentzell theorem, we show that the rate function has the same form as in Freidlin-Wentzell theory for diffusions with uniformly Lipschitz drift. In the second half of this article, we turn to proving an LDP for multiradial SLE$_\kappa$. Here, the main technical difficulty is that the SLE$_\kappa$ curves have a common target point, preventing the usual configurational, or global, approach. Instead, we make careful use of the contraction principle from the LDP for Dyson Brownian motion (proven in the first part of the article), combined with topological results in Loewner theory: we show that finite-energy multiradial Loewner hulls are always disjoint unions of simple curves, except possibly at their common endpoint. A key to this is obtained from a derivative estimate for the radial Loewner map in terms of the energy of its driving function.

Large language models (LLMs) are remarkably good at writing code. A particularly valuable case of human-LLM collaboration is code-based UI prototyping, a method for creating interactive prototypes that allows users to view and fully engage with a user interface. We conduct a formative study of GPT Pilot, a leading LLM-generated code-prototyping system, and find that its inflexibility towards change once development has started leads to weaknesses in failure prevention and dynamic planning; it closely resembles the linear workflow of the waterfall model. We introduce DIDUP, a system for code-based UI prototyping that follows an iterative spiral model, which takes changes and iterations that come up during the development process into account. We propose three novel mechanisms for LLM-generated code-prototyping systems: (1) adaptive planning, where plans should be dynamic and reflect changes during implementation, (2) code injection, where the system should write a minimal amount of code and inject it instead of rewriting code so users have a better mental model of the code evolution, and (3) lightweight state management, a simplified version of source control so users can quickly revert to different working states. Together, this enables users to rapidly develop and iterate on prototypes., Comment: 5 pages, 3 figures

Atmospheric neutral density is a crucial component to accurately predict and track the motion of satellites. During periods of elevated solar and geomagnetic activity atmospheric neutral density becomes highly variable and dynamic. This variability and enhanced dynamics make it difficult to accurately model neutral density leading to increased errors which propagate from neutral density models through to orbit propagation models. In this paper we investigate the dynamics of neutral density during geomagnetic storms. We use a combination of solar and geomagnetic variables to develop three Random Forest machine learning models of neutral density. These models are based on (1) slow solar indices, (2) high cadence solar irradiance, and (3) combined high-cadence solar irradiance and geomagnetic indices. Each model is validated using an out-of-sample dataset using analysis of residuals and typical metrics. During quiet-times, all three models perform well; however, during geomagnetic storms, the combined high cadence solar irradiance/geomagnetic model performs significantly better than the models based solely on solar activity. The combined model capturing an additional 10\% in the variability of density and having an error up to six times smaller during geomagnetic storms then the solar models. Overall, this work demonstrates the importance of including geomagnetic activity in the modeling of atmospheric density and serves as a proof of concept for using machine learning algorithms to model, and in the future forecast atmospheric density for operational use., Comment: Submitted for publication to Space Weather

We study the distribution of consecutive sums of two squares in arithmetic progressions. We show that for any odd squarefree modulus $q$, any two reduced congruence classes $a_1$ and $a_2$ mod $q$, and any $r_1,r_2 \ge 1$, a positive density of sums of two squares begin a chain of $r_1$ consecutive sums of two squares, all of which are $a_1$ mod $q$, followed immediately by a chain of $r_2$ consecutive sums of two squares, all of which are $a_2$ mod $q$. This is an analog of the result of Maynard for the sequence of primes, showing that for any reduced congruence class $a$ mod $q$ and for any $r \ge 1$, a positive density of primes begin a sequence of $r$ consecutive primes, all of which are $a$ mod $q$.

Many applications of RCTs involve the presence of multiple treatment administrators -- from field experiments to online advertising -- that compete for the subjects' attention. In the face of competition, estimating a causal effect becomes difficult, as the position at which a subject sees a treatment influences their response, and thus the treatment effect. In this paper, we build a game-theoretic model of agents who wish to estimate causal effects in the presence of competition, through a bidding system and a utility function that minimizes estimation error. Our main technical result establishes an approximation with a tractable objective that maximizes the sample value obtained through strategically allocating budget on subjects. This allows us to find an equilibrium in our model: we show that the tractable objective has a pure Nash equilibrium, and that any Nash equilibrium is an approximate equilibrium for our general objective that minimizes estimation error under broad conditions. Conceptually, our work successfully combines elements from causal inference and game theory to shed light on the equilibrium behavior of experimentation under competition., Comment: 37 pages, 3 figures, accepted at ICML'24

Quantum computing, a field utilizing the principles of quantum mechanics, promises great advancements across various industries. This survey paper is focused on the burgeoning intersection of quantum computing and intelligent transportation systems, exploring its potential to transform areas such as traffic optimization, logistics, routing, and autonomous vehicles. By examining current research efforts, challenges, and future directions, this survey aims to provide a comprehensive overview of how quantum computing could affect the future of transportation.

To investigate the short-term variability and determine the size of the optical continuum emitting size of intermediate-mass black holes (IMBHs), we carried out high-cadence, multi-band photometric monitoring of a Seyfert 1 galaxy J0249-0815 across two nights, together with a one-night single-band preliminary test. The presence of the broad Ha component in our target was confirmed by recent Palomar/P200 spectroscopic observations, 23 years after Sloan Digital Sky Survey, ruling out the supernovae origin of the broad Ha line. The photometric experiment was primarily conducted utilizing four-channel imagers MuSCAT 3 & 4 mounted on 2-meter telescopes within the Las Cumbres Observatory Global Telescope Network. Despite the expectation of variability, we observed no significant variation (<1.4%) on timescales of 6-10 hours. This non-detection is likely due to substantial host galaxy light diluting the subtle AGN variability. Dual-band preliminary tests and tailored simulations may enhance the possibility of detecting variability and lag in future IMBH reverberation campaigns., Comment: 14 pages, 6 figures, submitted to ApJ, comments welcome

The ability to build and leverage world models is essential for a general-purpose AI agent. Testing such capabilities is hard, in part because the building blocks of world models are ill-defined. We present Elements of World Knowledge (EWOK), a framework for evaluating world modeling in language models by testing their ability to use knowledge of a concept to match a target text with a plausible/implausible context. EWOK targets specific concepts from multiple knowledge domains known to be vital for world modeling in humans. Domains range from social interactions (help/hinder) to spatial relations (left/right). Both, contexts and targets are minimal pairs. Objects, agents, and locations in the items can be flexibly filled in enabling easy generation of multiple controlled datasets. We then introduce EWOK-CORE-1.0, a dataset of 4,374 items covering 11 world knowledge domains. We evaluate 20 openweights large language models (1.3B--70B parameters) across a battery of evaluation paradigms along with a human norming study comprising 12,480 measurements. The overall performance of all tested models is worse than human performance, with results varying drastically across domains. These data highlight simple cases where even large models fail and present rich avenues for targeted research on LLM world modeling capabilities., Comment: 21 pages (11 main), 7 figures. Authors Anna Ivanova, Aalok Sathe, Benjamin Lipkin contributed equally

Animated logos are a compelling and ubiquitous way individuals and brands represent themselves online. Manually authoring these logos can require significant artistic skill and effort. To help novice designers animate logos, design tools currently offer templates and animation presets. However, these solutions can be limited in their expressive range. Large language models have the potential to help novice designers create animated logos by generating animation code that is tailored to their content. In this paper, we introduce LogoMotion, an LLM-based system that takes in a layered document and generates animated logos through visually-grounded program synthesis. We introduce techniques to create an HTML representation of a canvas, identify primary and secondary elements, synthesize animation code, and visually debug animation errors. When compared with an industry standard tool, we find that LogoMotion produces animations that are more content-aware and are on par in terms of quality. We conclude with a discussion of the implications of LLM-generated animation for motion design.

The success of employment is highly related to a job seeker's capability of communicating and collaborating with others. While leveraging one's network during the job-seeking process is intuitive to the neurotypical, this can be challenging for people with autism. Recent empirical findings have started to show how facilitating collaboration between people with autism and their social surroundings through new design can improve their chances of employment. This work aims to provide actionable guidelines and conceptual frameworks that future researchers and practitioners can apply to improve collaborative design for job-seekers with autism. Built upon the literature on past technological interventions built for supporting job-seekers with autism, we define three major research challenges of (1) communication support, (2) employment stage-wise support, and (3) group work support. For each challenge, we review the current state-of-the-art practices and possible future solutions. We then suggest future designs that can provide breakthroughs from the interdisciplinary lens of human-AI collaboration, health services, group work, accessibility computing, and natural language processing.

Motivated by the quantum description of gauge theories, we study the cosmological effects of relaxing the Hamiltonian and momentum constraints in general relativity and Gauss' law in electromagnetism. We show that the unconstrained theories have new source terms that mimic a pressureless dust and a charge density that only follows geodesics. The source terms may be the simplest explanation for dark matter and generically predict a charged component. We comment that discovery of such terms would rule out inflation and be a direct probe of the initial conditions of the universe., Comment: 6 pages

Graph Transformers have garnered significant attention for learning graph-structured data, thanks to their superb ability to capture long-range dependencies among nodes. However, the quadratic space and time complexity hinders the scalability of Graph Transformers, particularly for large-scale recommendation. Here we propose an efficient Masked Graph Transformer, named MGFormer, capable of capturing all-pair interactions among nodes with a linear complexity. To achieve this, we treat all user/item nodes as independent tokens, enhance them with positional embeddings, and feed them into a kernelized attention module. Additionally, we incorporate learnable relative degree information to appropriately reweigh the attentions. Experimental results show the superior performance of our MGFormer, even with a single attention layer.

Recent advances in multimodal large language models (LLMs) have made it easier to rapidly prototype AI-powered features, especially for mobile use cases. However, gathering early, mobile-situated user feedback on these AI prototypes remains challenging. The broad scope and flexibility of LLMs means that, for a given use-case-specific prototype, there is a crucial need to understand the wide range of in-the-wild input users are likely to provide and their in-context expectations for the AI's behavior. To explore the concept of in situ AI prototyping and testing, we created MobileMaker: a platform that enables designers to rapidly create and test mobile AI prototypes directly on devices. This tool also enables testers to make on-device, in-the-field revisions of prototypes using natural language. In an exploratory study with 16 participants, we explored how user feedback on prototypes created with MobileMaker compares to that of existing prototyping tools (e.g., Figma, prompt editors). Our findings suggest that MobileMaker prototypes enabled more serendipitous discovery of: model input edge cases, discrepancies between AI's and user's in-context interpretation of the task, and contextual signals missed by the AI. Furthermore, we learned that while the ability to make in-the-wild revisions led users to feel more fulfilled as active participants in the design process, it might also constrain their feedback to the subset of changes perceived as more actionable or implementable by the prototyping tool.

The majority of visual SLAM systems are not robust in dynamic scenarios. The ones that deal with dynamic objects in the scenes usually rely on deep-learning-based methods to detect and filter these objects. However, these methods cannot deal with unknown moving objects. This work presents Panoptic-SLAM, an open-source visual SLAM system robust to dynamic environments, even in the presence of unknown objects. It uses panoptic segmentation to filter dynamic objects from the scene during the state estimation process. Panoptic-SLAM is based on ORB-SLAM3, a state-of-the-art SLAM system for static environments. The implementation was tested using real-world datasets and compared with several state-of-the-art systems from the literature, including DynaSLAM, DS-SLAM, SaD-SLAM, PVO and FusingPanoptic. For example, Panoptic-SLAM is on average four times more accurate than PVO, the most recent panoptic-based approach for visual SLAM. Also, experiments were performed using a quadruped robot with an RGB-D camera to test the applicability of our method in real-world scenarios. The tests were validated by a ground-truth created with a motion capture system.