Search

A digital information explosion has transformed cities' residential and educational markets in ways that are still being uncovered. Although urban stratification scholars have increasingly scrutinized whether emerging digital platforms disrupt or reproduce longstanding segregation patterns, direct links between one theoretically important form of digital information--school quality data--and neighborhood and school segregation are rarely drawn. To clarify these dynamics, we leverage an exogenous digital information shock, in which the Los Angeles Times' website revealed measures of a particularly important school quality proxy--schools' value-added effectiveness--for nearly all elementary schools in the Los Angeles Unified School District. Results suggest that although the information shock had no detectable effects on residential sorting or neighborhood racial segregation, it did exert modest effects on school sorting--particularly for Latino and Asian students--albeit not in ways that materially diminished school racial segregation because the racial compositions of high and low value-added schools were broadly similar both before and after the information shock. We conclude that the urban stratification implications of digital information may be more nuanced than often appreciated, with effects shaped by racial heterogeneity in both constraints and preferences vis-à-vis specific types of information and operating through mechanisms beyond residential segregation.

One of the key challenges of detecting AI-generated images is spotting images that have been created by previously unseen generative models. We argue that the limited diversity of the training data is a major obstacle to addressing this problem, and we propose a new dataset that is significantly larger and more diverse than prior work. As part of creating this dataset, we systematically download thousands of text-to-image latent diffusion models and sample images from them. We also collect images from dozens of popular open source and commercial models. The resulting dataset contains 2.7M images that have been sampled from 4803 different models. These images collectively capture a wide range of scene content, generator architectures, and image processing settings. Using this dataset, we study the generalization abilities of fake image detectors. Our experiments suggest that detection performance improves as the number of models in the training set increases, even when these models have similar architectures. We also find that detection performance improves as the diversity of the models increases, and that our trained detectors generalize better than those trained on other datasets., Comment: 15 pages

New line lists for four isotopologues of nickel monohydride, $^{58}$NiH, $^{60}$NiH, $^{62}$NiH, and $^{58}$NiD are presented covering the wavenumber range $<10000$ cm$^{-1}$ ($\lambda > 1$ $\mu$m), $J$ up to 37.5 for transitions within and between the three lowest-lying electronic states, ${X}\,^{2}\Delta$, ${W}\,^{2}\Pi$, and ${V}\,^{2}\Sigma^{+}$. The line lists are applicable for temperatures up to 5000 K. The line lists calculations are based on a recent empirical NiH spectroscopic model [Havalyova et al. J. Quant. Spectrosc. Radiat. Transf., 272, 107800, (2021)] which is adapted for the variational nuclear-motion code Duo. The model consists of potential energy curves, spin-orbit coupling curves, electronic angular momentum curves, spin-rotation coupling curves, $\Lambda$-doubling correction curve for $^2\Pi$ states and Born-Oppenheimer breakdown (BOB) rotational correction curves. New ab initio dipole moment curves, scaled to match the experimental dipole moment of the ground state, are used to compute Einstein A coefficients. The BYOT line lists are included in the ExoMol database at www.exomol.com.

As our understanding of autism and ableism continues to increase, so does our understanding of ableist language towards autistic people. Such language poses a significant challenge in NLP research due to its subtle and context-dependent nature. Yet, detecting anti-autistic ableist language remains underexplored, with existing NLP tools often failing to capture its nuanced expressions. We present AUTALIC, the first benchmark dataset dedicated to the detection of anti-autistic ableist language in context, addressing a significant gap in the field. The dataset comprises 2,400 autism-related sentences collected from Reddit, accompanied by surrounding context, and is annotated by trained experts with backgrounds in neurodiversity. Our comprehensive evaluation reveals that current language models, including state-of-the-art LLMs, struggle to reliably identify anti-autistic ableism and align with human judgments, underscoring their limitations in this domain. We publicly release AUTALIC along with the individual annotations which serve as a valuable resource to researchers working on ableism, neurodiversity, and also studying disagreements in annotation tasks. This dataset serves as a crucial step towards developing more inclusive and context-aware NLP systems that better reflect diverse perspectives., Comment: 9 pages, 5 figures, 7 tables

Today's tactile sensors have a variety of different designs, making it challenging to develop general-purpose methods for processing touch signals. In this paper, we learn a unified representation that captures the shared information between different tactile sensors. Unlike current approaches that focus on reconstruction or task-specific supervision, we leverage contrastive learning to integrate tactile signals from two different sensors into a shared embedding space, using a dataset in which the same objects are probed with multiple sensors. We apply this approach to paired touch signals from GelSlim and Soft Bubble sensors. We show that our learned features provide strong pretraining for downstream pose estimation and classification tasks. We also show that our embedding enables models trained using one touch sensor to be deployed using another without additional training. Project details can be found at https://www.mmintlab.com/research/cttp/.

A theoretical line list for $^{15}$NH$_3$ CoYuTe-15 is presented based on the empirical potential energy and ab initio dipole moments surfaces developed and used for the production of the ExoMol line list CoYuTe for $^{14}$NH$_3$. The ro-vibrational energy levels and wavefunctions are computed using the variational program TROVE. The line list ranges up to 10000 cm$^{-1}$ ($\lambda \geq 1$ $\mu$m) and contains 929 795 249 transitions between 1 269 961 states with $J\le 30$. The line list should be applicable for temperatures up to $\sim$1000 K. To improve the accuracy of the line positions, a set of experimentally-derived energy levels of $^{15}$NH$_3$ is produced using the MARVEL procedure. To this end, 37 experimental sources of the line positions of $^{15}$NH$_3$ available in the literature are collected, combined and systematised to produce a self-consistent spectroscopic network of 21095 $^{15}$NH$_3$ transitions covering 40 vibrational bands ranging up to 6818 cm$^{-1}$ and resulting in 2777 energy term values. These MARVEL energies are then used to replace the theoretical values in the CoYuTe-15 line list and also complemented by pseudo-MARVEL energies obtained by an isotopologue extrapolation using the previously reported MARVEL energies of the $^{14}$NH$_3$ parent isotopologue of ammonia. A list of 53856 high resolution transitions between MARVEL states and theoretical intensities is provided in the HITRAN format. Comparison with the recent experimental spectra of $^{15}$NH$_3$ illustrate the potential of the line list for detections and as an efficient assistant in spectroscopic assignments. The line list is available from www.exomol.com.

We investigate the Lyman-$\alpha$ (Ly$\alpha$) and Lyman continuum (LyC) properties of the Sunburst Arc, a $z=2.37$ gravitationally lensed galaxy with a multiply-imaged, compact region leaking LyC and a triple-peaked Ly$\alpha$ profile indicating direct Ly$\alpha$ escape. Non-LyC-leaking regions show a redshifted Ly$\alpha$ peak, a redshifted and central Ly$\alpha$ peak, or a triple-peaked Ly$\alpha$ profile. We measure the properties of the Ly$\alpha$ profile from different regions of the galaxy using $R\sim5000$ Magellan/MagE spectra. We compare the Ly$\alpha$ spectral properties to LyC and narrowband Ly$\alpha$ maps from Hubble Space Telescope (HST) imaging to explore the subgalactic Ly$\alpha-$LyC connection. We find strong correlations (Pearson correlation coefficient $r>0.6$) between the LyC escape fraction ($f_{\rm esc}^{\rm LyC}$) and Ly$\alpha$ (1) peak separation $v_{\rm{sep}}$, (2) ratio of the minimum flux density between the redshifted and blueshifted Ly$\alpha$ peaks to continuum flux density $f_{\rm{min}}/f_{\rm{cont}}$, and (3) equivalent width. We favor a complex \ion{H}{1} geometry to explain the Ly$\alpha$ profiles from non-LyC-leaking regions and suggest two \ion{H}{1} geometries that could diffuse and/or rescatter the central Ly$\alpha$ peak from the LyC-leaking region into our sightline across transverse distances of several hundred parsecs. Our results emphasize the complexity of Ly$\alpha$ radiative transfer and its sensitivity to the anisotropies of \ion{H}{1} gas on subgalactic scales. Large differences in the physical scales on which we observe spatially variable direct escape Ly$\alpha$, blueshifted Ly$\alpha$, and escaping LyC photons in the Sunburst Arc underscore the importance of resolving the physical scales that govern Ly$\alpha$ and LyC escape., Comment: Submitted to The Astrophysical Journal with revisions from the first referee report. Comments welcome

Cat qubits, a type of bosonic qubit encoded in a harmonic oscillator, can exhibit an exponential noise bias against bit-flip errors with increasing mean photon number. Here, we focus on cat qubits stabilized by two-photon dissipation, where pairs of photons are added and removed from a harmonic oscillator by an auxiliary, lossy buffer mode. This process requires a large loss rate and strong nonlinearities of the buffer mode that must not degrade the coherence and linearity of the oscillator. In this work, we show how to overcome this challenge by coloring the loss environment of the buffer mode with a multi-pole filter and optimizing the circuit to take into account additional inductances in the buffer mode. Using these techniques, we achieve near-ideal enhancement of cat-qubit bit-flip times with increasing photon number, reaching over $0.1$ seconds with a mean photon number of only $4$. Concurrently, our cat qubit remains highly phase coherent, with phase-flip times corresponding to an effective lifetime of $T_{1,\text{eff}} \simeq 70$ $\mu$s, comparable with the bare oscillator lifetime. We achieve this performance even in the presence of an ancilla transmon, used for reading out the cat qubit states, by engineering a tunable oscillator-ancilla dispersive coupling. Furthermore, the low nonlinearity of the harmonic oscillator mode allows us to perform pulsed cat-qubit stabilization, an important control primitive, where the stabilization can remain off for a significant fraction (e.g., two thirds) of a $3~\mathrm{\mu s}$ cycle without degrading bit-flip times. These advances are important for the realization of scalable error-correction with cat qubits, where large noise bias and low phase-flip error rate enable the use of hardware-efficient outer error-correcting codes., Comment: Comments welcome!

Due to intensive genetic selection for rapid growth rates and high broiler yields in recent years, the global poultry industry has faced a challenging problem in the form of woody breast (WB) conditions. This condition has caused significant economic losses as high as $200 million annually, and the root cause of WB has yet to be identified. Human palpation is the most common method of distinguishing a WB from others. However, this method is time-consuming and subjective. Hyperspectral imaging (HSI) combined with machine learning algorithms can evaluate the WB conditions of fillets in a non-invasive, objective, and high-throughput manner. In this study, 250 raw chicken breast fillet samples (normal, mild, severe) were taken, and spatially heterogeneous hardness distribution was first considered when designing HSI processing models. The study not only classified the WB levels from HSI but also built a regression model to correlate the spectral information with sample hardness data. To achieve a satisfactory classification and regression model, a neural network architecture search (NAS) enabled a wide-deep neural network model named NAS-WD, which was developed. In NAS-WD, NAS was first used to automatically optimize the network architecture and hyperparameters. The classification results show that NAS-WD can classify the three WB levels with an overall accuracy of 95%, outperforming the traditional machine learning model, and the regression correlation between the spectral data and hardness was 0.75, which performs significantly better than traditional regression models.

We present a simple, self-supervised approach to the Tracking Any Point (TAP) problem. We train a global matching transformer to find cycle consistent tracks through video via contrastive random walks, using the transformer's attention-based global matching to define the transition matrices for a random walk on a space-time graph. The ability to perform "all pairs" comparisons between points allows the model to obtain high spatial precision and to obtain a strong contrastive learning signal, while avoiding many of the complexities of recent approaches (such as coarse-to-fine matching). To do this, we propose a number of design decisions that allow global matching architectures to be trained through self-supervision using cycle consistency. For example, we identify that transformer-based methods are sensitive to shortcut solutions, and propose a data augmentation scheme to address them. Our method achieves strong performance on the TapVid benchmarks, outperforming previous self-supervised tracking methods, such as DIFT, and is competitive with several supervised methods., Comment: ECCV 2024. Project link: https://ayshrv.com/gmrw . Code: https://github.com/ayshrv/gmrw/

Speech sounds convey a great deal of information about the scenes, resulting in a variety of effects ranging from reverberation to additional ambient sounds. In this paper, we manipulate input speech to sound as though it was recorded within a different scene, given an audio-visual conditional example recorded from that scene. Our model learns through self-supervision, taking advantage of the fact that natural video contains recurring sound events and textures. We extract an audio clip from a video and apply speech enhancement. We then train a latent diffusion model to recover the original speech, using another audio-visual clip taken from elsewhere in the video as a conditional hint. Through this process, the model learns to transfer the conditional example's sound properties to the input speech. We show that our model can be successfully trained using unlabeled, in-the-wild videos, and that an additional visual signal can improve its sound prediction abilities. Please see our project webpage for video results: https://tinglok.netlify.app/files/avsoundscape/, Comment: ECCV 2024

Modern incarnations of tactile sensors produce high-dimensional raw sensory feedback such as images, making it challenging to efficiently store, process, and generalize across sensors. To address these concerns, we introduce a novel implicit function representation for tactile sensor feedback. Rather than directly using raw tactile images, we propose neural implicit functions trained to reconstruct the tactile dataset, producing compact representations that capture the underlying structure of the sensory inputs. These representations offer several advantages over their raw counterparts: they are compact, enable probabilistically interpretable inference, and facilitate generalization across different sensors. We demonstrate the efficacy of this representation on the downstream task of in-hand object pose estimation, achieving improved performance over image-based methods while simplifying downstream models. We release code, demos and datasets at https://www.mmintlab.com/tactile-functasets.

Large Language Models (LLMs) are powerful tools with the potential to benefit society immensely, yet, they have demonstrated biases that perpetuate societal inequalities. Despite significant advancements in bias mitigation techniques using data augmentation, zero-shot prompting, and model fine-tuning, biases continuously persist, including subtle biases that may elude human detection. Recent research has shown a growing interest in multi-LLM approaches, which have been demonstrated to be effective in improving the quality of reasoning and factuality in LLMs. Building on this approach, we propose a novel multi-LLM debiasing framework aimed at reducing bias in LLMs. Our work is the first to introduce and evaluate two distinct approaches within this framework for debiasing LLMs: a centralized method, where the conversation is facilitated by a single central LLM, and a decentralized method, where all models communicate directly. Our findings reveal that our multi-LLM framework significantly reduces bias in LLMs, outperforming the baseline method across several social groups.

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

Today's touch sensors come in many shapes and sizes. This has made it challenging to develop general-purpose touch processing methods since models are generally tied to one specific sensor design. We address this problem by performing cross-modal prediction between touch sensors: given the tactile signal from one sensor, we use a generative model to estimate how the same physical contact would be perceived by another sensor. This allows us to apply sensor-specific methods to the generated signal. We implement this idea by training a diffusion model to translate between the popular GelSlim and Soft Bubble sensors. As a downstream task, we perform in-hand object pose estimation using GelSlim sensors while using an algorithm that operates only on Soft Bubble signals. The dataset, the code, and additional details can be found at https://www.mmintlab.com/research/touch2touch/.

We present the latest results from the Chicago Carnegie Hubble Program (CCHP) to measure the Hubble constant using data from the James Webb Space Telescope (JWST). This program is based upon three independent methods: (1) Tip of the Red Giant Branch (TRGB) stars, (2) JAGB (J-Region Asymptotic Giant Branch) stars, and (3) Cepheids. Our program includes 10 nearby galaxies, each hosting Type Ia supernovae, suitable for measuring the Hubble constant (Ho). It also includes NGC 4258, which has a geometric distance, setting the zero point for all three methods. The JWST observations have significantly higher signal-to-noise and finer angular resolution than previous observations with the Hubble Space Telescope (HST). We find three independent values of Ho = 69.85 +/- 1.75 (stat) +/- 1.54 (sys) for the TRGB, Ho = 67.96 +/- 1.85 (stat) +/- 1.90 (sys) for the JAGB, and Ho = 72.05 +/- 1.86 (stat) +/- 3.10 (sys) km/s/Mpc for Cepheids. Tying into supernovae, and combining these methods adopting a flat prior, yields our current estimate of Ho = 69.96 +/- 1.05 (stat) +/- 1.12 (sys) km/s/Mpc. The distances measured using the TRGB and the JAGB method agree at the 1% level, but differ from the Cepheid distances at the 2.5-4% level. The value of Ho based on these two methods with JWST data alone is Ho = 69.03 +/- 1.75 (total error) km/sec/Mpc. These numbers are consistent with the current standard Lambda CDM model, without the need for the inclusion of additional new physics. Future JWST data will be required to increase the precision and accuracy of the local distance scale., Comment: 61 pages, 20 figures

The J-region asymptotic giant branch (JAGB) method is a new standard candle based on the constant luminosities of carbon-rich asymptotic giant branch stars in the J band. The JAGB method is independent of the Cepheid and TRGB distance indicators. Therefore, we can leverage it to both cross-check Cepheid and TRGB distances for systematic errors and use it to measure an independent local Hubble constant. The JAGB method also boasts a number of advantages in measuring distances relative to the TRGB and Cepheids, several of which are especially amplified when combined with JWST's revolutionary resolving power. First, JAGB stars are 1 mag brighter in the NIR than the TRGB, and can be discovered from single-epoch NIR photometry unlike Cepheids which require congruent optical imaging in at least 12 epochs. Thus, JAGB stars can be used to measure significantly farther distances than both the TRGB stars and Cepheids using the same amount of observing time. Further advantages include: JAGB stars are easily identified solely via their colors and magnitudes, dust extinction is reduced in near-infrared observations, and JAGB stars are ubiquitous in all galaxies with intermediate-age populations. In this paper, we present a novel algorithm that identifies the optimal location in a galaxy for applying the JAGB method, so as to minimize effects from crowding. We then deploy this algorithm in JWST NIRCam imaging of seven SN Ia host galaxies to measure their JAGB distances, undertaking a completely blind analysis. The zero-point of this JAGB distance scale is set in the water mega-maser galaxy NGC 4258. In our CCHP overview paper Freedman et al. (2024), we apply the JAGB distances measured in this paper to the Carnegie Supernova Program (CSP) SNe Ia sample, measuring a Hubble constant of H0 = 67.96 +/- 1.85 (stat) +/- 1.90 (sys) km/s/Mpc., Comment: 24 pages, 10 figures, 5 tables, submitted to AAS Journals, v2 now includes systematic uncertainty in abstract

Wide-separation lensed quasars (WSLQs) are a rare class of strongly lensed quasars, magnified by foreground massive galaxy clusters, with typically large magnifications of the multiple quasar images. They are a relatively unexplored opportunity for detailed study of quasar host galaxies. The current small sample of known WSLQs has a median redshift of $z\approx 2.1$, larger than most other samples of quasar host galaxies studied to date. Here, we derive precise constraints on the properties of six WSLQs and their host galaxies, using parametric surface brightness fitting, measurements of quasar emission lines, and stellar population synthesis of host galaxies in six WSLQ systems. Our results, with significant uncertainty, indicate that these six hosts are a mixture of star-forming and quiescent galaxies. To probe for co-evolution between AGNs and host galaxies, we model the offset from the `local' ($z=0$) $M_{\rm{BH}}\unicode{x2013}M_\star$ relation as a simple power-law in redshift. Accounting for selection effects, a WSLQ-based model for evolution in the $M_{\rm{BH}}\unicode{x2013}M_\star$ relation has a power-law index of $\gamma_M=-0.42\pm0.31$, consistent with no evolution. Compared to several literature samples, which mostly probe unlensed quasars at $z<2$, the WSLQ sample shows less evolution from the local relation, at $\sim 4\sigma$. We find that selection affects and choices of $M_{\rm{BH}}$ calibration are the most important systematics in these comparisons. Given that we resolve host galaxy flux confidently even from the ground in some instances, our work demonstrates that WSLQs and highly magnified AGNs are exceptional systems for future AGN$\unicode{x2013}$host co-evolution studies., Comment: Submitted to ApJ. 25 pages + 7-page appendix, 12+4 figures. Key results are shown starting with Figure 6. Comments welcome

Obtaining microscopic structure-property relationships for grain boundaries are challenging because of the complex atomic structures that underlie their behavior. This has led to recent efforts to obtain these relationships with machine learning, but representing a grain boundary structure in a manner suitable for machine learning is not a trivial task. There are three key steps common to property prediction in grain boundaries and other variable-sized atom clustered structures. These are: (1) describe the atomic structure as a feature matrix, (2) transform the variable-sized feature matrices of different structures to a fixed length common to all structures, and (3) apply machine learning to predict properties from the transformed feature matrices. We examine these feature engineering steps to understand how they impact the accuracy of grain boundary energy predictions. A database of over 7000 grain boundaries serves to evaluate the different feature engineering combinations. We also examine how these combination of engineered features provide interpretability, or the ability to extract insightful physics from the obtained structure-property relationships., Comment: 18 pages, 8 figures

Non-stationary systems are found throughout the world, from climate patterns under the influence of variation in carbon dioxide concentration, to brain dynamics driven by ascending neuromodulation. Accordingly, there is a need for methods to analyze non-stationary processes, and yet most time-series analysis methods that are used in practice, on important problems across science and industry, make the simplifying assumption of stationarity. One important problem in the analysis of non-stationary systems is the problem class that we refer to as Parameter Inference from a Non-stationary Unknown Process (PINUP). Given an observed time series, this involves inferring the parameters that drive non-stationarity of the time series, without requiring knowledge or inference of a mathematical model of the underlying system. Here we review and unify a diverse literature of algorithms for PINUP. We formulate the problem, and categorize the various algorithmic contributions. This synthesis will allow researchers to identify gaps in the literature and will enable systematic comparisons of different methods. We also demonstrate that the most common systems that existing methods are tested on - notably the non-stationary Lorenz process and logistic map - are surprisingly easy to perform well on using simple statistical features like windowed mean and variance, undermining the practice of using good performance on these systems as evidence of algorithmic performance. We then identify more challenging problems that many existing methods perform poorly on and which can be used to drive methodological advances in the field. Our results unify disjoint scientific contributions to analyzing non-stationary systems and suggest new directions for progress on the PINUP problem and the broader study of non-stationary phenomena.

Boasting a 6.5m mirror in space, JWST can increase by several times the number of supernovae (SNe) to which a redshift-independent distance has been measured with a precision distance indicator (e.g., TRGB or Cepheids); the limited number of such SN calibrators currently dominates the uncertainty budget in distance ladder Hubble constant (H0) experiments. JWST/NIRCAM imaging of the Virgo Cluster galaxy NGC4536 is used here to preview JWST program GO-1995, which aims to measure H0 using three stellar distance indicators (Cepheids, TRGB, JAGB/carbon stars). Each population of distance indicator was here successfully detected -- with sufficiently large number statistics, well-measured fluxes, and characteristic distributions consistent with ingoing expectations -- so as to confirm that we can acquire distances from each method precise to about 0.05mag (statistical uncertainty only). We leverage overlapping HST imaging to identify TRGB stars, cross-match them with the JWST photometry, and present a preliminary constraint on the slope of the TRGB's F115W-(F115W}-F444W) relation equal to -0.99 +/- 0.16 mag/mag. This slope is consistent with prior slope measurements in the similar 2MASS J-band, as well as with predictions from the BASTI isochrone suite. We use the new TRGB slope estimate to flatten the two-dimensional TRGB feature and measure a (blinded) TRGB distance relative to a set of fiducial TRGB colors, intended to represent the absolute fiducial calibrations expected from geometric anchors such as NGC4258 and the Magellanic Clouds. In doing so, we empirically demonstrate that the TRGB can be used as a standardizable candle at the IR wavelengths accessible with JWST., Comment: Revised version after submission to AAS journals; 20 pages, 12 figures; Fig. 1 compressed to reduce file size

Domain adaptive semantic segmentation aims to generate accurate and dense predictions for an unlabeled target domain by leveraging a supervised model trained on a labeled source domain. The prevalent self-training approach involves retraining the dense discriminative classifier of $p(class|pixel feature)$ using the pseudo-labels from the target domain. While many methods focus on mitigating the issue of noisy pseudo-labels, they often overlook the underlying data distribution p(pixel feature|class) in both the source and target domains. To address this limitation, we propose the multi-prototype Gaussian-Mixture-based (ProtoGMM) model, which incorporates the GMM into contrastive losses to perform guided contrastive learning. Contrastive losses are commonly executed in the literature using memory banks, which can lead to class biases due to underrepresented classes. Furthermore, memory banks often have fixed capacities, potentially restricting the model's ability to capture diverse representations of the target/source domains. An alternative approach is to use global class prototypes (i.e. averaged features per category). However, the global prototypes are based on the unimodal distribution assumption per class, disregarding within-class variation. To address these challenges, we propose the ProtoGMM model. This novel approach involves estimating the underlying multi-prototype source distribution by utilizing the GMM on the feature space of the source samples. The components of the GMM model act as representative prototypes. To achieve increased intra-class semantic similarity, decreased inter-class similarity, and domain alignment between the source and target domains, we employ multi-prototype contrastive learning between source distribution and target samples. The experiments show the effectiveness of our method on UDA benchmarks.

We analyze an optical atomic clock using two-photon $5S_{1/2} \rightarrow 4D_J$ transitions in rubidium. Four one- and two-color excitation schemes to probe the fine-structure states $4D_{3/2}$ and $4D_{5/2}$ are considered in detail. We compare key characteristics of Rb $4D_J$ and $5D_{5/2}$ two-photon clocks. The $4D_J$ clock features a high signal-to-noise ratio due to two-photon decay at favorable wavelengths, low dc electric and magnetic susceptibilities, and minimal black-body shifts. Ac Stark shifts from the clock interrogation lasers are compensated by two-color Rabi-frequency matching. We identify a "magic" wavelength near 1060~nm, which allows for in-trap, Doppler-free clock-transition interrogation with lattice-trapped cold atoms. From our analysis of clock statistics and systematics, we project a quantum-noise-limited relative clock stability at the $10^{-13}/\sqrt{\tau(s)}$-level, with integration time $\tau$ in seconds, and a relative accuracy of $\sim 10^{-13}$. We describe a potential architecture for implementing the proposed clock using a single telecom clock laser at 1550~nm, which is conducive to optical communication and long-distance clock comparisons. Our work could be of interest in efforts to realize small and portable Rb clocks and in high-precision measurements of atomic properties of Rb $4D_J$-states., Comment: A.D. and C.O. contributed equally to this work; 15 pages, 3 figures, 1 table

The ExoMol database (www.exomol.com) provides molecular data for spectroscopic studies of hot atmospheres. These data are widely used to model atmospheres of exoplanets, cool stars and other astronomical objects, as well as a variety of terrestrial applications. The 2024 data release reports the current status of the database which contains recommended line lists for 91 molecules and 224 isotopologues giving a total of almost 10$^{12}$ individual transitions. New features of the database include extensive "MARVELization" of line lists to allow them to be used for high resolutions studies, extension of several line lists to ultraviolet wavelengths, provision of photodissociation cross sections and extended provision of broadening parameters. Some of the in-house data specifications have been rewritten in JSON and moved to conformity with other international standards. Data products, including specific heats, a database of lifetimes for plasma studies, and the ExoMolHR web app which allows exclusively high resolution data to be extracted, are discussed.

We present a computational framework that transforms single images into 3D physical objects. The visual geometry of a physical object in an image is determined by three orthogonal attributes: mechanical properties, external forces, and rest-shape geometry. Existing single-view 3D reconstruction methods often overlook this underlying composition, presuming rigidity or neglecting external forces. Consequently, the reconstructed objects fail to withstand real-world physical forces, resulting in instability or undesirable deformation -- diverging from their intended designs as depicted in the image. Our optimization framework addresses this by embedding physical compatibility into the reconstruction process. We explicitly decompose the three physical attributes and link them through static equilibrium, which serves as a hard constraint, ensuring that the optimized physical shapes exhibit desired physical behaviors. Evaluations on a dataset collected from Objaverse demonstrate that our framework consistently enhances the physical realism of 3D models over existing methods. The utility of our framework extends to practical applications in dynamic simulations and 3D printing, where adherence to physical compatibility is paramount.

Spectrograms are 2D representations of sound that look very different from the images found in our visual world. And natural images, when played as spectrograms, make unnatural sounds. In this paper, we show that it is possible to synthesize spectrograms that simultaneously look like natural images and sound like natural audio. We call these visual spectrograms images that sound. Our approach is simple and zero-shot, and it leverages pre-trained text-to-image and text-to-spectrogram diffusion models that operate in a shared latent space. During the reverse process, we denoise noisy latents with both the audio and image diffusion models in parallel, resulting in a sample that is likely under both models. Through quantitative evaluations and perceptual studies, we find that our method successfully generates spectrograms that align with a desired audio prompt while also taking the visual appearance of a desired image prompt. Please see our project page for video results: https://ificl.github.io/images-that-sound/, Comment: Accepted to NeurIPS 2024. Project site: https://ificl.github.io/images-that-sound/

We propose a simple strategy for masking image patches during visual-language contrastive learning that improves the quality of the learned representations and the training speed. During each iteration of training, we randomly mask clusters of visually similar image patches, as measured by their raw pixel intensities. This provides an extra learning signal, beyond the contrastive training itself, since it forces a model to predict words for masked visual structures solely from context. It also speeds up training by reducing the amount of data used in each image. We evaluate the effectiveness of our model by pre-training on a number of benchmarks, finding that it outperforms other masking strategies, such as FLIP, on the quality of the learned representation., Comment: CVPR 2024, Project page: https://zxp46.github.io/cluster-masking/ , Code: https://github.com/Zi-hao-Wei/Efficient-Vision-Language-Pre-training-by-Cluster-Masking

"Godzilla" is a peculiar object within the gravitationally lensed Sunburst Arc at $z=2.37$. Despite being very bright, it appears in only one of the twelve lensed images of the source galaxy, and shows exotic spectroscopic properties not found elsewhere in the galaxy. We use JWST's unique combination of spatial resolution and spectroscopic sensitivity to provide a unified, coherent explanation of the physical nature of Godzilla. We measure fluxes and kinematic properties of rest-optical emission lines in Godzilla and surrounding regions. Using standard line ratio-based diagnostic methods in combination with NIRCam imaging and ground based rest-UV spectra, we characterize Godzilla and its surroundings. We find that Godzilla is most likely an extremely magnified, non-erupting LBV star with dense gas condensations in close proximity. Among around 60 detected lines, we find a cascade of strong O I lines pumped by intense Ly$\beta$ emission, as well as Ly$\alpha$-pumped rest-optical Fe II lines, reminiscent of the Weigelt blobs in the local LBV star Eta Carinae. Godzilla is surrounded by dusty, inhomogeneous gas common to massive, evolved stars. Spectra and images of Godzilla and adjacent objects and the detection of a low-surface brightness foreground galaxy in the NIRCam data support the interpretation that Godzilla is a stellar-scale object extremely magnified by alignment with lensing caustics. To explain the dusty surroundings, strong [Ne III] and line kinematics simultaneously, we argue that Godzilla is a post-eruption LBV accompanied by a hotter companion and/or gas condensations exposed to more intense radiation compared to the Weigelt blobs. We expect periodic spectroscopic variations if Godzilla is a binary system. If Godzilla is confirmed to be an LBV star, it expands the distance to the furthest known LBV from a dozen Mpc to several Gpc., Comment: 18 pages, 16 figures. Submitted to A&A

We measure the gas-phase abundances of the elements He, N, O, Ne, S, Ar, and Fe in the Lyman-continuum emitting region of the Sunburst Arc, a highly magnified galaxy at redshift $z=2.37$. We detect the temperature-sensitive auroral lines [SII]$\lambda\lambda4069,4076$, [OII]$\lambda\lambda7320,7330$, [SIII]$\lambda6312$, [OIII]$\lambda4363$, and [NeIII]$\lambda3343$ in a stacked spectrum of 5 multiple images of the Lyman-continuum emitter (LCE), from which we directly measure the electron temperature in the low, intermediate, and high ionization zones. We also detect the density-sensitive doublets of [OII]$\lambda\lambda3727,3729$, [SII]$\lambda\lambda6717,6731$, and [ArIV]$\lambda\lambda4713,4741$, which constrain the density in both the low- and high-ionization gas. With these temperature and density measurements, we measure gas-phase abundances with similar rigor as studies of local galaxies. We measure a gas-phase metallicity for the LCE of $12+\log(\textrm{O}/\textrm{H}) = 7.97 \pm 0.05$, and find an enhanced nitrogen abundance $\log(\textrm{N}/\textrm{O}) = -0.65^{+0.16}_{-0.25}$. This nitrogen abundance is consistent with enrichment from a population of Wolf-Rayet stars, additional signatures of which are reported in a companion paper. Abundances of sulfur, argon, neon, and iron are consistent with local low-metallicity HII regions and low-redshift galaxies. This study represents the most complete chemical abundance analysis of a galaxy at Cosmic Noon to date, which enables direct comparisons between local HII regions and those in the distant universe., Comment: 15 pages, 4 figures, 3 tables. Submitted to ApJ

We present a scene representation, which we call a tactile-augmented radiance field (TaRF), that brings vision and touch into a shared 3D space. This representation can be used to estimate the visual and tactile signals for a given 3D position within a scene. We capture a scene's TaRF from a collection of photos and sparsely sampled touch probes. Our approach makes use of two insights: (i) common vision-based touch sensors are built on ordinary cameras and thus can be registered to images using methods from multi-view geometry, and (ii) visually and structurally similar regions of a scene share the same tactile features. We use these insights to register touch signals to a captured visual scene, and to train a conditional diffusion model that, provided with an RGB-D image rendered from a neural radiance field, generates its corresponding tactile signal. To evaluate our approach, we collect a dataset of TaRFs. This dataset contains more touch samples than previous real-world datasets, and it provides spatially aligned visual signals for each captured touch signal. We demonstrate the accuracy of our cross-modal generative model and the utility of the captured visual-tactile data on several downstream tasks. Project page: https://dou-yiming.github.io/TaRF, Comment: CVPR 2024, Project page: https://dou-yiming.github.io/TaRF, Code: https://github.com/Dou-Yiming/TaRF/

Changing clinical algorithms to remove race adjustment has been proposed and implemented for multiple health conditions. Removing race adjustment from estimated glomerular filtration rate (eGFR) equations may reduce disparities in chronic kidney disease (CKD), but has not been studied in clinical practice after implementation. Here, we assessed whether implementing an eGFR equation (CKD-EPI 2021) without adjustment for Black or African American race modified quarterly rates of nephrology referrals and visits within a single healthcare system, Stanford Health Care (SHC). Our cohort study analyzed 547,194 adult patients aged 21 and older who had at least one recorded serum creatinine or serum cystatin C between January 1, 2019 and September 1, 2023. During the study period, implementation of CKD-EPI 2021 did not modify rates of quarterly nephrology referrals in those documented as Black or African American or in the overall cohort. After adjusting for capacity at SHC nephrology clinics, estimated rates of nephrology referrals and visits with CKD-EPI 2021 were 34 (95% CI 29, 39) and 188 (175, 201) per 10,000 patients documented as Black or African American. If race adjustment had not been removed, estimated rates were nearly identical: 38 (95% CI: 28, 53) and 189 (165, 218) per 10,000 patients. Changes to the eGFR equation are likely insufficient to achieve health equity in CKD care decision-making as many other structural inequities remain., Comment: Accepted to Conference on Health, Inference, and Learning (CHIL) 2024

Characterizing and predicting the training performance of modern machine learning (ML) workloads on compute systems with compute and communication spread between CPUs, GPUs, and network devices is not only the key to optimization and planning but also a complex goal to achieve. The primary challenges include the complexity of synchronization and load balancing between CPUs and GPUs, the variance in input data distribution, and the use of different communication devices and topologies (e.g., NVLink, PCIe, network cards) that connect multiple compute devices, coupled with the desire for flexible training configurations. Built on top of our prior work for single-GPU platforms, we address these challenges and enable multi-GPU performance modeling by incorporating (1) data-distribution-aware performance models for embedding table lookup, and (2) data movement prediction of communication collectives, into our upgraded performance modeling pipeline equipped with inter-and intra-rank synchronization for ML workloads trained on multi-GPU platforms. Beyond accurately predicting the per-iteration training time of DLRM models with random configurations with a geomean error of 5.21% on two multi-GPU platforms, our prediction pipeline generalizes well to other types of ML workloads, such as Transformer-based NLP models with a geomean error of 3.00%. Moreover, even without actually running ML workloads like DLRMs on the hardware, it is capable of generating insights such as quickly selecting the fastest embedding table sharding configuration (with a success rate of 85%)., Comment: 12 pages, 11 figures, 4 tables

Given a factorization of an image into a sum of linear components, we present a zero-shot method to control each individual component through diffusion model sampling. For example, we can decompose an image into low and high spatial frequencies and condition these components on different text prompts. This produces hybrid images, which change appearance depending on viewing distance. By decomposing an image into three frequency subbands, we can generate hybrid images with three prompts. We also use a decomposition into grayscale and color components to produce images whose appearance changes when they are viewed in grayscale, a phenomena that naturally occurs under dim lighting. And we explore a decomposition by a motion blur kernel, which produces images that change appearance under motion blurring. Our method works by denoising with a composite noise estimate, built from the components of noise estimates conditioned on different prompts. We also show that for certain decompositions, our method recovers prior approaches to compositional generation and spatial control. Finally, we show that we can extend our approach to generate hybrid images from real images. We do this by holding one component fixed and generating the remaining components, effectively solving an inverse problem.