Search

Topic modeling is a key method in text analysis, but existing approaches are limited by assuming one topic per document or fail to scale efficiently for large, noisy datasets of short texts. We introduce Semantic Component Analysis (SCA), a novel topic modeling technique that overcomes these limitations by discovering multiple, nuanced semantic components beyond a single topic in short texts which we accomplish by introducing a decomposition step to the clustering-based topic modeling framework. Evaluated on multiple Twitter datasets, SCA matches the state-of-the-art method BERTopic in coherence and diversity, while uncovering at least double the semantic components and maintaining a noise rate close to zero while staying scalable and effective across languages, including an underrepresented one., Comment: 5 pages, 3 figures, code: https://github.com/mainlp/semantic_components

We present a large-scale evaluation of 30 cognitive biases in 20 state-of-the-art large language models (LLMs) under various decision-making scenarios. Our contributions include a novel general-purpose test framework for reliable and large-scale generation of tests for LLMs, a benchmark dataset with 30,000 tests for detecting cognitive biases in LLMs, and a comprehensive assessment of the biases found in the 20 evaluated LLMs. Our work confirms and broadens previous findings suggesting the presence of cognitive biases in LLMs by reporting evidence of all 30 tested biases in at least some of the 20 LLMs. We publish our framework code to encourage future research on biases in LLMs: https://github.com/simonmalberg/cognitive-biases-in-llms

The NOvA collaboration reports cross-section measurements for $\nu_{\mu}$ charged-current interactions with low hadronic energy (maximum kinetic energy of 250 MeV for protons and 175 MeV for pions) in the NOvA Near Detector. The results are presented as a double-differential cross section as a function of the direct observables of the final-state muon kinematics. Results are also presented as a single-differential cross section as a function of the derived square of the four-momentum transfer, $Q^{2}$, and as a function of the derived neutrino energy. The data correspond to an accumulated 8.09$\times10^{20}$ protons-on-target (POT) in the neutrino mode of the NuMI beam, with a narrow band of neutrino energies peaked at 1.8 GeV. The analysis provides a sample of neutrino-nucleus interactions with an enhanced fraction of quasi-elastic and two-particle-two-hole (2p2h) interactions. This enhancement allows quantitative comparisons with various nuclear models. We find strong disagreement between data and theory-based models in various regions of the muon kinematic phase space, especially in the forward muon direction., Comment: 20 pages, 12 figures. The second version includes an additional citation and adds four previously missing authors

Double- and single-differential cross sections for inclusive charged-current neutrino-nucleus scattering are reported for the kinematic domain 0 to 2 GeV/c in three-momentum transfer and 0 to 2 GeV in available energy, at a mean muon-neutrino energy of 1.86 GeV. The measurements are based on an estimated 995,760 muon-neutrino CC interactions in the scintillator medium of the NOvA Near Detector. The subdomain populated by 2-particle-2-hole reactions is identified by the cross-section excess relative to predictions for neutrino-nucleus scattering that are constrained by a data control sample. Models for 2-particle-2- hole processes are rated by chi-square comparisons of the predicted-versus-measured muon-neutrino CC inclusive cross section over the full phase space and in the restricted subdomain. Shortfalls are observed in neutrino generator predictions obtained using the theory-based Val`encia and SuSAv2 2p2h models., Comment: 20 pages, 14 figures

Explanatory images play a pivotal role in accessible and easy-to-read (E2R) texts. However, the images available in online databases are not tailored toward the respective texts, and the creation of customized images is expensive. In this large-scale study, we investigated whether text-to-image generation models can close this gap by providing customizable images quickly and easily. We benchmarked seven, four open- and three closed-source, image generation models and provide an extensive evaluation of the resulting images. In addition, we performed a user study with people from the E2R target group to examine whether the images met their requirements. We find that some of the models show remarkable performance, but none of the models are ready to be used at a larger scale without human supervision. Our research is an important step toward facilitating the creation of accessible information for E2R creators and tailoring accessible images to the target group's needs., Comment: To be published at TSAR workshop 2024 (https://tsar-workshop.github.io/)

Neural encoding, or neural representation, is a field in neuroscience that focuses on characterizing how information is encoded in the spiking activity of neurons. Currently, little is known about how sensory neurons can preserve information from multiple stimuli given their broad receptive fields. Multiplexing is a neural encoding theory that posits that neurons temporally switch between encoding various stimuli in their receptive field. Here, we construct a statistically falsifiable single-neuron model for multiplexing using a competition-based framework. The spike train models are constructed using drift-diffusion models, implying an integrate-and-fire framework to model the temporal dynamics of the membrane potential of the neuron. In addition to a multiplexing-specific model, we develop alternative models that represent alternative encoding theories (normalization, winner-take-all, subadditivity, etc.) with some level of abstraction. Using information criteria, we perform model comparison to determine whether the data favor multiplexing over alternative theories of neural encoding. Analysis of spike trains from the inferior colliculus of two macaque monkeys provides tenable evidence of multiplexing and offers new insight into the timescales at which switching occurs.

This paper uses topic modeling and bias measurement techniques to analyze and determine gender bias in English song lyrics. We utilize BERTopic to cluster 537,553 English songs into distinct topics and chart their development over time. Our analysis shows the thematic shift in song lyrics over the years, from themes of romance to the increasing sexualization of women in songs. We observe large amounts of profanity and misogynistic lyrics on various topics, especially in the overall biggest cluster. Furthermore, to analyze gender bias across topics and genres, we employ the Single Category Word Embedding Association Test (SC-WEAT) to compute bias scores for the word embeddings trained on the most popular topics as well as for each genre. We find that words related to intelligence and strength tend to show a male bias across genres, as opposed to appearance and weakness words, which are more female-biased; however, a closer look also reveals differences in biases across topics., Comment: Accepted and presented at the 17th International Conference on Social Computing, Behavioral-Cultural Modeling, & Prediction and Behavior Representation in Modeling and Simulation (see https://sbp-brims.org/2024/papers/working-papers/Chen_SBP-BRiMS2024_Final_31.pdf )

In the era dominated by information overload and its facilitation with Large Language Models (LLMs), the prevalence of misinformation poses a significant threat to public discourse and societal well-being. A critical concern at present involves the identification of machine-generated news. In this work, we take a significant step by introducing a benchmark dataset designed for neural news detection in four languages: English, Turkish, Hungarian, and Persian. The dataset incorporates outputs from multiple multilingual generators (in both, zero-shot and fine-tuned setups) such as BloomZ, LLaMa-2, Mistral, Mixtral, and GPT-4. Next, we experiment with a variety of classifiers, ranging from those based on linguistic features to advanced Transformer-based models and LLMs prompting. We present the detection results aiming to delve into the interpretablity and robustness of machine-generated texts detectors across all target languages., Comment: EMNLP 2024 NLP4PI Workshop

We use recent results from the theory of random matrices to improve instrumental variables estimation with many instruments. In settings where the first-stage parameters are dense, we show that Ridge lowers the implicit price of a bias adjustment. This comes along with improved (finite-sample) properties in the second stage regression. Our theoretical results nest existing results on bias approximation and bias adjustment. Moreover, it extends them to settings with more instruments than observations.

Advanced Artificial Intelligence (AI) systems, specifically large language models (LLMs), have the capability to generate not just misinformation, but also deceptive explanations that can justify and propagate false information and erode trust in the truth. We examined the impact of deceptive AI generated explanations on individuals' beliefs in a pre-registered online experiment with 23,840 observations from 1,192 participants. We found that in addition to being more persuasive than accurate and honest explanations, AI-generated deceptive explanations can significantly amplify belief in false news headlines and undermine true ones as compared to AI systems that simply classify the headline incorrectly as being true/false. Moreover, our results show that personal factors such as cognitive reflection and trust in AI do not necessarily protect individuals from these effects caused by deceptive AI generated explanations. Instead, our results show that the logical validity of AI generated deceptive explanations, that is whether the explanation has a causal effect on the truthfulness of the AI's classification, plays a critical role in countering their persuasiveness - with logically invalid explanations being deemed less credible. This underscores the importance of teaching logical reasoning and critical thinking skills to identify logically invalid arguments, fostering greater resilience against advanced AI-driven misinformation.

This paper introduces a novel approach to illumination manipulation in diffusion models, addressing the gap in conditional image generation with a focus on lighting conditions. We conceptualize the diffusion model as a black-box image render and strategically decompose its energy function in alignment with the image formation model. Our method effectively separates and controls illumination-related properties during the generative process. It generates images with realistic illumination effects, including cast shadow, soft shadow, and inter-reflections. Remarkably, it achieves this without the necessity for learning intrinsic decomposition, finding directions in latent space, or undergoing additional training with new datasets.

For a positive and invertible linear operator $T$ acting on a $C^*$-algebra, we give necessary and sufficient criteria for the inverse operator $T^{-1}$ to be positive, too. Moreover, a simple counterexample shows that $T^{-1}$ need not be positive even if $T$ is unital and its spectrum is contained in the unit circle., Comment: 16 pages

Stimuli-responsive macromolecules display large conformational changes during their dynamics, sometimes switching states, and are an integral property for the development of soft functional materials. Here, we introduce a mean-field dynamical density functional theory (DDFT) for a model of responsive colloids (RCs) to study the nonequilibrium dynamics of a colloidal dispersion in time-dependent external fields, with a focus on the coupling of translational and conformational dynamics during their relaxation. Specifically, we consider soft Gaussian particles with a bimodal size distribution between two confining walls with time-dependent (switching-on and off) external gravitational and osmotic fields. We find a rich relaxation behavior of the systems in excellent agreement with particle-based Brownian dynamics (BD) computer simulations. In particular, we find time-asymmetric relaxations of integrated observables (wall pressures, mean size, liquid center-of-mass) for activation/deactivation of external potentials, respectively, which are tuneable by the ratio of translational and conformational diffusion time scales. Our work paves thus the way for studying the nonequilibrium relaxation dynamics of complex soft matter with multiple degrees of freedom and hierarchical relaxations., Comment: 14 pages, 8 figures

(abridged) The Atacama Large Aperture Submillimeter Telescope (AtLAST) is undergoing a design study for a large (50 meter) single-dish submm-wavelength Ritchey-Chr\'etien telescope to be located 5050 meters above sea level in the Atacama Desert in northern Chile. It will allow for observations covering 30 to 950 GHz. Observing at such high frequencies with a 50~m primary mirror will be challenging, and has never been attempted thus far. This observational capability demands exquisite control of systematics to ensure a reliable beam shape, and to mitigate the expected sidelobe levels. Among them, critical issues that large telescopes like AtLAST need to deal with are introduced by the panel gap pattern, the secondary mirror supporting struts, mirror deformations produced by thermal and gravitational effects, and Ruze scattering due to surface roughness. Proprietary software such as TICRA-Tools allows for full-wave, complex-field physical optics simulations taking into account these features. Such calculations can be computationally expensive since the mirror surfaces are gridded (meshed) into a fine array in which each element is treated as a current source. If the telescope size is large and the wavelengths are short this may lead to very long running times. Here we present a set of physical optics results that allow us to estimate the performance of the telescope in terms of beam shape, directivity, sidelobes level and stray light. We also discuss how we addressed the computational challenges, and provide caveats on how to shorten the run times. Above all, we conclude that the scattering effects from the gaps and tertiary support structure are minimal, and subdominant to the Ruze scattering., Comment: Submitted to SPIE Proceedings

The Atacama Large Aperture Submillimeter Telescope (AtLAST) aims to be the premier next generation large diameter (50-meter) single dish observatory capable of observations across the millimeter/sub-millimeter spectrum, from 30 to 950~GHz. The large primary mirror diameter, the 2-degree field of view and its large 4.7-meter focal surface give AtLAST a high throughput (aperture size times field of view) and grasp (throughput times spectral reach), with the ability to illuminate $>\mathcal{O}(10^7)$ detectors. The optical design concept for AtLAST consists of a numerically optimized two-mirror Ritchey-Chr\'etien system with an additional flat folding mirror, which enables a quick selection among its planned six instrument positions. We present the optical design concept and discuss the expected optical performance of AtLAST. We then present design concepts that can be implemented in the receiver and instrument optics to correct for astigmatism and mitigate the high degree of curvature of the focal surface in order to recover significant fractions of the geometric field of view at sub-millimeter wavelengths.

The high level of photorealism in state-of-the-art diffusion models like Midjourney, Stable Diffusion, and Firefly makes it difficult for untrained humans to distinguish between real photographs and AI-generated images. To address this problem, we designed a guide to help readers develop a more critical eye toward identifying artifacts, inconsistencies, and implausibilities that often appear in AI-generated images. The guide is organized into five categories of artifacts and implausibilities: anatomical, stylistic, functional, violations of physics, and sociocultural. For this guide, we generated 138 images with diffusion models, curated 9 images from social media, and curated 42 real photographs. These images showcase the kinds of cues that prompt suspicion towards the possibility an image is AI-generated and why it is often difficult to draw conclusions about an image's provenance without any context beyond the pixels in an image. Human-perceptible artifacts are not always present in AI-generated images, but this guide reveals artifacts and implausibilities that often emerge. By drawing attention to these kinds of artifacts and implausibilities, we aim to better equip people to distinguish AI-generated images from real photographs in the future., Comment: 54 pages, 189 Figures

The Atacama Large Aperture Submm Telescope (AtLAST) is a concept for a novel 50-meter class single-dish telescope operating at sub-millimeter and millimeter wavelengths (30-950 GHz). The telescope will provide an unprecedentedly wide field of view (FoV) of 1-2 degree diameter with a large receiver cabin housing six major instruments in Nasmyth and Cassegrain positions. The high observing frequencies, combined with the scanning operation movements with up to 3 deg/second, place high demands on the accuracy and stability of the optical and structural components. The design features the introduction of a rocking chair type mount with an iso-statically decoupled main reflector backup structure and an active main reflector surface with a high precision metrology system. The planned site location is in the Chilean Atacama Desert at approximately 5050 meters above sea level, near Llano de Chajnantor. This paper gives an overview of the optical, structural, and mechanical design concepts. It explains the flow-down from key science requirements to technical design decisions as well as showing design analogies from other existing large radio, (sub-)mm, and optical telescopes., Comment: 20 pages. Submitted to the SPIE proceedings

The study of growth-induced surface wrinkling in constrained bilayers comprising a thin film attached to a thick substrate is a canonical model for understanding pattern formation in many biological systems. While the bilayer model has received much prior attention, the nonlinear behaviour for arrangements with similar film and substrate properties, or substrate growth that outpaces film growth, remains poorly understood. This paper therefore focuses on these cases in which the substrate's elasticity dominates surface wrinkling. We study the critical states, and the initial and advanced post-critical behaviour of growing bilayers with film-to-substrate modulus ratios in the region of $2.5$--$50$, and cases where the substrate grows faster than the film. Based on nonlinear elasticity, we formulate analytical models for linear buckling analyses and asymptotic projections around the critical point, and use finite element (FE) models coupled to continuation and branch-switching algorithms to uncover the deep post-critical regime. It is shown that a rapidly growing substrate may change the critical mode from film-governed sinusoidal wrinkling to substrate-governed Biot wrinkling depending on the stiffness ratio and growth ratio. We present a phase change diagram of the post-critical modal landscape split into sinusoidal wrinkling, period doubling, period quadrupling, and creasing regimes in terms of the stiffness ratio and growth ratio. While the post-critical regime of film- and substrate-dominated bilayers (either in terms of dominant elasticity or growth rate) is governed by sinusoidal wrinkling and Biot creasing, respectively, the intermediate regions allow for period doubling and quadrupling bifurcations. Finally, we demonstrate the existence of multi-stability in the advanced post-buckling regimes for growing bilayers where growth in the substrate surpasses that of the film., Comment: 32 pages, 9 figures

The Simons Observatory will map the temperature and polarization over half of the sky, at millimeter wavelengths in six spectral bands from the Atacama Desert in Chile. These data will provide new insights into the genesis, content, and history of our Universe; the astrophysics of galaxies and galaxy clusters; objects in our solar system; and time-varying astrophysical phenomena. This ambitious new instrument suite, initially comprising three 0.5 m small-aperture telescopes and one 6 m large aperture telescope, is designed using a common combination of new technologies and new implementations to realize an observatory significantly more capable than the previous generation. In this paper, we present the pre-deployment performance of the first mid-frequency "optics tube" which will be fielded on the large aperture telescope with sensitivity to the 90 and 150 GHz spectral bands. This optics tube contains lenses, filters, detectors, and readout components, all of which operate at cryogenic temperatures. It is one of seven that form the core of the large aperture telescope receiver in its initial deployment. We describe this optics tube, including details of comprehensive testing methods, new techniques for beam and passband characterization, and its measured performance. The performance metrics include beams, optical efficiency, passbands, and forecasts for the on-sky performance of the system. We forecast a sensitivity that exceeds the requirements of the large aperture telescope with greater than 30% margin in each spectral band, and predict that the instrument will realize diffraction-limited performance and the expected detector passbands.

The Simons Observatory (SO) is a cosmic microwave background (CMB) survey experiment that includes small-aperture telescopes (SATs) observing from an altitude of 5,200 m in the Atacama Desert in Chile. The SO SATs will cover six spectral bands between 27 and 280 GHz to search for primordial B-modes to a sensitivity of $\sigma(r)=0.002$, with quantified systematic errors well below this value. Each SAT is a self-contained cryogenic telescope with a 35$^\circ$ field of view, 42 cm diameter optical aperture, 40 K half-wave plate, 1 K refractive optics, and $<0.1$ K focal plane that holds $>12,000$ TES detectors. We describe the nominal design of the SATs and present details about the integration and testing for one operating at 93 and 145 GHz.

The task of toxicity detection is still a relevant task, especially in the context of safe and fair LMs development. Nevertheless, labeled binary toxicity classification corpora are not available for all languages, which is understandable given the resource-intensive nature of the annotation process. Ukrainian, in particular, is among the languages lacking such resources. To our knowledge, there has been no existing toxicity classification corpus in Ukrainian. In this study, we aim to fill this gap by investigating cross-lingual knowledge transfer techniques and creating labeled corpora by: (i)~translating from an English corpus, (ii)~filtering toxic samples using keywords, and (iii)~annotating with crowdsourcing. We compare LLMs prompting and other cross-lingual transfer approaches with and without fine-tuning offering insights into the most robust and efficient baselines., Comment: Accepted to WOAH, NAACL, 2024. arXiv admin note: text overlap with arXiv:2404.02043

The first stars might have been fast rotators. This would have important consequences for their radiative, mechanical and chemical feedback. We discuss the impact of fast initial rotation on the evolution of massive Population III models and on their nitrogen and oxygen stellar yields. We explore the evolution of Population III stars with initial masses in the range of 9Msol < Mini < 120Msol starting with an initial rotation on the Zero Age Main Sequence equal to 70% of the critical one. We find that with the physics of rotation considered here, our rapidly-rotating Population III stellar models do not follow a homogeneous evolution. They lose very little mass in case mechanical winds are switched on when the surface rotation becomes equal or larger than the critical velocity. Impact on the ionising flux appears modest when compared to moderately-rotating models. Fast rotation favours, in models with initial masses above ~20Msol, the appearance of a very extended intermediate convective zone around the H-burning shell during the core He-burning phase. This shell has important consequences on the sizes of the He- and CO-cores and thus impacts the final fate of stars. Moreover, it has a strong impact on nucleosynthesis boosting the production of primary 14N. Fast initial rotation impacts significantly the chemical feedback of Population III stars. Observations of extremely metal-poor stars and/or starbursting regions are essential to provide constraints on the properties of the first stars., Comment: Accepted in A&A. Pages 17. Figures 15. Tables 2

Text simplification seeks to improve readability while retaining the original content and meaning. Our study investigates whether pre-trained classifiers also maintain such coherence by comparing their predictions on both original and simplified inputs. We conduct experiments using 11 pre-trained models, including BERT and OpenAI's GPT 3.5, across six datasets spanning three languages. Additionally, we conduct a detailed analysis of the correlation between prediction change rates and simplification types/strengths. Our findings reveal alarming inconsistencies across all languages and models. If not promptly addressed, simplified inputs can be easily exploited to craft zero-iteration model-agnostic adversarial attacks with success rates of up to 50%, Comment: Published at DeTermIt! Workshop at LREC-COLING 2024

Despite the extensive amount of labeled datasets in the NLP text classification field, the persistent imbalance in data availability across various languages remains evident. Ukrainian, in particular, stands as a language that still can benefit from the continued refinement of cross-lingual methodologies. Due to our knowledge, there is a tremendous lack of Ukrainian corpora for typical text classification tasks. In this work, we leverage the state-of-the-art advances in NLP, exploring cross-lingual knowledge transfer methods avoiding manual data curation: large multilingual encoders and translation systems, LLMs, and language adapters. We test the approaches on three text classification tasks -- toxicity classification, formality classification, and natural language inference -- providing the "recipe" for the optimal setups.

The Simons Array (SA) project is a ground-based Cosmic Microwave Background (CMB) polarization experiment. The SA observes the sky using three telescopes, and POLARBEAR-2A (PB-2A) is the receiver system on the first telescope. For the ground-based experiment, atmospheric fluctuation is the primary noise source that could cause polarization leakage. In the PB-2A receiver system, a continuously rotating half-wave plate (HWP) is used to mitigate the polarization leakage. However, due to the rapid modulation of the polarization signal, the uncertainty in the time constant of the detector results in an uncertainty in the polarization angle. For PB-2A, the time constant of each bolometer needs to be calibrated at the sub-millisecond level to avoid introducing bias to the polarization signal. We have developed a new calibrator system that can be used to calibrate the time constants of the detectors. In this study, we present the design of the calibration system and the preliminary results of the time constant calibration for PB-2A., Comment: Proceedings of the 15th Asia Pacific Physics Conference (APPC15)

Benchmarking plays a pivotal role in assessing and enhancing the performance of compact deep learning models designed for execution on resource-constrained devices, such as microcontrollers. Our study introduces a novel, entirely artificially generated benchmarking dataset tailored for speech recognition, representing a core challenge in the field of tiny deep learning. SpokeN-100 consists of spoken numbers from 0 to 99 spoken by 32 different speakers in four different languages, namely English, Mandarin, German and French, resulting in 12,800 audio samples. We determine auditory features and use UMAP (Uniform Manifold Approximation and Projection for Dimension Reduction) as a dimensionality reduction method to show the diversity and richness of the dataset. To highlight the use case of the dataset, we introduce two benchmark tasks: given an audio sample, classify (i) the used language and/or (ii) the spoken number. We optimized state-of-the-art deep neural networks and performed an evolutionary neural architecture search to find tiny architectures optimized for the 32-bit ARM Cortex-M4 nRF52840 microcontroller. Our results represent the first benchmark data achieved for SpokeN-100., Comment: Accepted as a full paper by the tinyML Research Symposium 2024

This Letter reports a search for charge-parity (CP) symmetry violating non-standard interactions (NSI) of neutrinos with matter using the NOvA Experiment, and examines their effects on the determination of the standard oscillation parameters. Data from $\nu_{\mu}(\bar{\nu}_{\mu})\rightarrow\nu_{\mu}(\bar{\nu}_{\mu})$ and $\nu_{\mu}(\bar{\nu}_{\mu})\rightarrow\nu_{e}(\bar{\nu}_{e})$ oscillation channels are used to measure the effect of the NSI parameters $\varepsilon_{e\mu}$ and $\varepsilon_{e\tau}$. With 90% C.L. the magnitudes of the NSI couplings are constrained to be $|\varepsilon_{e\mu}| \, \lesssim 0.3$ and $|\varepsilon_{e\tau}| \, \lesssim 0.4$. A degeneracy at $|\varepsilon_{e\tau}| \, \approx 1.8$ is reported, and we observe that the presence of NSI limits sensitivity to the standard CP phase $\delta_{\tiny\text{CP}}$.

The submillimeter and millimeter ((sub-)mm) sky contains a vast wealth of information that is both complementary and inaccessible to other wavelengths. Over half the light we receive is observable at (sub-)mm wavelengths, yet we have mapped only a small portion of the sky at sufficient spatial resolution and sensitivity to detect and resolve distant galaxies or star forming cores within their their large-scale environment. For decades the astronomical community has highlighted the need for a large, high-throughput (sub-)mm ($\lambda\sim0.35-10$ mm) single dish; The Atacama Large Aperture Submillimeter Telescope (AtLAST), with its 50-m aperture and $2^\circ$ maximal field of view, aims to be such a facility. We present here the full design concept for AtLAST, developed through an EU-funded project. Our design approach begins with a long lineage of (sub-)mm telescopes, relies on calculations and simulations to realize the optics, and uses finite element analysis to optimize the mechanical structure and subsystems. The demanding design requirements for AtLAST, set by transformative science goals, combine novel concepts with lessons learned from past experience. The result is an innovative rocking chair design with six instrument bays, two of which are mounted on Nasmyth platforms. AtLAST will be capable of $3^\circ\,\rm s^{-1}$ scanning and $1^\circ\,\rm s^{-2}$ acceleration, and aims to feature a surface accuracy of $\leq 20~\mu$m half wavefront error under static deformations, allowing observations up to $\approx 950$~GHz. Crucially, AtLAST will be a sustainable, visionary facility that will allow upgrades for decades to come. While some aspects require further testing, prototyping, and field demonstrations, we estimate that the design will be construction-ready this decade., Comment: This version addresses the first round A&A referee reports

NOvA is a long-baseline neutrino oscillation experiment that measures oscillations in charged-current $\nu_{\mu} \rightarrow \nu_{\mu}$ (disappearance) and $\nu_{\mu} \rightarrow \nu_{e}$ (appearance) channels, and their antineutrino counterparts, using neutrinos of energies around 2 GeV over a distance of 810 km. In this work we reanalyze the dataset first examined in our previous paper [Phys. Rev. D 106, 032004 (2022)] using an alternative statistical approach based on Bayesian Markov Chain Monte Carlo. We measure oscillation parameters consistent with the previous results. We also extend our inferences to include the first NOvA measurements of the reactor mixing angle $\theta_{13}$ and the Jarlskog invariant. We use these results to quantify the strength of our inferences about CP violation, as well as to examine the effects of constraints from short-baseline measurements of $\theta_{13}$ using antineutrinos from nuclear reactors when making NOvA measurements of $\theta_{23}$. Our long-baseline measurement of $\theta_{13}$ is also shown to be consistent with the reactor measurements, supporting the general applicability and robustness of the PMNS framework for neutrino oscillations., Comment: 20 pages, 17 figures; version accepted by Phys. Rev. D. Data associated with this paper is available at https://doi.org/10.15484/2349444

The Simons Observatory (SO) is a cosmic microwave background instrumentation suite being deployed in the Atacama Desert in northern Chile. The telescopes within SO use three types of dichroic transition-edge sensor (TES) detector arrays, with the 90 and 150 GHz Mid-Frequency (MF) arrays containing 65% of the approximately 68,000 detectors in the first phase of SO. All of the 26 required MF detector arrays have now been fabricated, packaged into detector modules, and tested in laboratory cryostats. Across all modules, we find an average operable detector yield of 84% and median saturation powers of (2.8, 8.0) pW with interquartile ranges of (1, 2) pW at (90, 150) GHz, respectively, falling within their targeted ranges. We measure TES normal resistances and superconducting transition temperatures on each detector wafer to be uniform within 3%, with overall central values of 7.5 mohm and 165 mK, respectively. Results on time constants, optical efficiency, and noise performance are also presented and are consistent with achieving instrument sensitivity forecasts., Comment: 8 pages, 3 figures. Proceedings of the 20th International Conference on Low Temperature Detectors (LTD20). Accepted to JLTP

Transition-edge sensor (TES) bolometers are broadly used for background-limited astrophysical measurements from the far-infrared to mm-waves. Many planned future instruments require increasingly large detector arrays, but their scalability is limited by their cryogenic readout electronics. Microwave SQUID multiplexing offers a highly capable scaling solution through the use of inherently broadband circuitry, enabling readout of hundreds to thousands of channels per microwave line. As with any multiplexing technique, the channelization mechanism gives rise to electrical crosstalk which must be understood and controlled so as to not degrade the instrument sensitivity. Here, we explore implications relevant for TES bolometer array applications, focusing in particular on upcoming mm-wave observatories such as the Simons Observatory and AliCPT. We model the relative contributions of the various underlying crosstalk mechanisms, evaluate the difference between fixed-tone and tone-tracking readout systems, and discuss ways in which crosstalk nonlinearity will complicate on-sky measurements., Comment: Accepted for publication in the Journal of Low Temperature Physics

Benchmark datasets for digital dermatology unwittingly contain inaccuracies that reduce trust in model performance estimates. We propose a resource-efficient data-cleaning protocol to identify issues that escaped previous curation. The protocol leverages an existing algorithmic cleaning strategy and is followed by a confirmation process terminated by an intuitive stopping criterion. Based on confirmation by multiple dermatologists, we remove irrelevant samples and near duplicates and estimate the percentage of label errors in six dermatology image datasets for model evaluation promoted by the International Skin Imaging Collaboration. Along with this paper, we publish revised file lists for each dataset which should be used for model evaluation. Our work paves the way for more trustworthy performance assessment in digital dermatology., Comment: Link to the revised file lists: https://github.com/Digital-Dermatology/SelfClean-Revised-Benchmarks

While hundreds of artificial intelligence (AI) algorithms are now approved or cleared by the US Food and Drugs Administration (FDA), many studies have shown inconsistent generalization or latent bias, particularly for underrepresented populations. Some have proposed that generative AI could reduce the need for real data, but its utility in model development remains unclear. Skin disease serves as a useful case study in synthetic image generation due to the diversity of disease appearance, particularly across the protected attribute of skin tone. Here we show that latent diffusion models can scalably generate images of skin disease and that augmenting model training with these data improves performance in data-limited settings. These performance gains saturate at synthetic-to-real image ratios above 10:1 and are substantially smaller than the gains obtained from adding real images. As part of our analysis, we generate and analyze a new dataset of 458,920 synthetic images produced using several generation strategies. Our results suggest that synthetic data could serve as a force-multiplier for model development, but the collection of diverse real-world data remains the most important step to improve medical AI algorithms.

Large language models underestimate the impact of negations on how much they change the meaning of a sentence. Therefore, learned evaluation metrics based on these models are insensitive to negations. In this paper, we propose NegBLEURT, a negation-aware version of the BLEURT evaluation metric. For that, we designed a rule-based sentence negation tool and used it to create the CANNOT negation evaluation dataset. Based on this dataset, we fine-tuned a sentence transformer and an evaluation metric to improve their negation sensitivity. Evaluating these models on existing benchmarks shows that our fine-tuned models outperform existing metrics on the negated sentences by far while preserving their base models' performances on other perturbations., Comment: Accepted to INLG 2023

The purpose of intrinsic decomposition is to separate an image into its albedo (reflective properties) and shading components (illumination properties). This is challenging because it's an ill-posed problem. Conventional approaches primarily concentrate on 2D imagery and fail to fully exploit the capabilities of 3D data representation. 3D point clouds offer a more comprehensive format for representing scenes, as they combine geometric and color information effectively. To this end, in this paper, we introduce Point Intrinsic Net (PoInt-Net), which leverages 3D point cloud data to concurrently estimate albedo and shading maps. The merits of PoInt-Net include the following aspects. First, the model is efficient, achieving consistent performance across point clouds of any size with training only required on small-scale point clouds. Second, it exhibits remarkable robustness; even when trained exclusively on datasets comprising individual objects, PoInt-Net demonstrates strong generalization to unseen objects and scenes. Third, it delivers superior accuracy over conventional 2D approaches, demonstrating enhanced performance across various metrics on different datasets. (Code Released), Comment: Code: https://github.com/xyxingx/PoInt-Net

Massive stars (~8-25Msun) stripped of their hydrogen-rich envelopes via binary interaction are thought to be the main progenitors for merging neutron stars and stripped-envelope supernovae. We recently presented the discovery of the first set of such stripped stars in a companion paper. Here, we fit the spectra of ten stars with new atmosphere models in order to constrain their stellar properties precisely. We find that the stellar properties align well with the theoretical expectations from binary evolution models for helium-core burning envelope-stripped stars. The fits confirm that the stars have high effective temperatures (Teff~50-100kK), high surface gravities (log g ~5), and hydrogen-poor/helium-rich surfaces (X(H, surf)~0-0.4) while showing for the first time a range of bolometric luminosities (10^3-10^5 Lsun), small radii (~0.5-1Rsun), and low Eddington factors (Gamma_e~0.006-0.4). Using these properties, we derive intermediate current masses (~1-8Msun), which suggest that their progenitors were massive stars (~5-25Msun) and that a subset will reach core-collapse, leaving behind neutron stars or black holes. Using the model fits, we also estimate the emission rates of ionizing photons for these stars, which agree well with previous model expectations. Further, by computing models for a range of mass-loss rates, we find that the stellar winds are weaker than predicted by any existing scheme (Mdot(wind)<~ 1e-9 Msun/yr). The properties of this first sample of intermediate mass helium stars suggest they both contain progenitors of type Ib and IIb supernovae, and provide important benchmarks for binary evolution and population synthesis models., Comment: Accepted for publication in ApJ. Spectral models available via zenodo as described in text. Please also see a companion manuscript detailing the discovery of these systems in today's arXiv posting