Search

We present a new evolution of the Very Little Eel-Inspired roBot, the VLEIBot++, a 900-mg swimmer driven by two 10-mg bare high-work density (HWD) actuators, whose functionality is based on the use of shape-memory alloy (SMA) wires. An actuator of this type consumes an average power of about 40 mW during in-air operation. We integrated onboard power and computation into the VLEIBot++ using a custom-built printed circuit board (PCB) and an 11-mAh 3.7-V 507-mg single-cell lithium-ion (Li-Ion) battery, which in conjunction enable autonomous swimming for about 20 min on a single charge. This robot can swim at speeds of up to 18.7 mm/s (0.46 Bl/s) and is the first subgram microswimmer with onboard power, actuation, and computation developed to date. Unfortunately, the approach employed to actuate VLEIBot++ prototypes is infeasible for underwater applications because a typical 10-mg bare SMA-based microactuator requires an average power on the order of 800 mW when operating underwater. To address this issue, we introduce a new 13-mg power-efficient high-performance SMA-based microactuator that can function with similar power requirements (approx. 80 mW on average) and actuation performance (approx. 3 mm at low frequencies) in air and water. This design is based on the use of a sealed flexible air-capsule that encloses the SMA wires that drive the microactuator with the purpose of passively controlling the heat-transfer rate of the thermal system. Furthermore, this new power-efficient encapsulated actuator requires low voltages of excitation (3 to 4 V) and simple power electronics to function. The breakthroughs presented in this paper represent a path towards the creation of insect-scale autonomous underwater vehicles (AUVs)., Comment: To be presented at the 40th International Symposium on Robotics Research (ISRR 2024) in Long Beach, CA. on December 12th

Diffusion models have significant impact on wide range of generative tasks, especially on image inpainting and restoration. Although the improvements on aiming for decreasing number of function evaluations (NFE), the iterative results are still computationally expensive. Consistency models are as a new family of generative models, enable single-step sampling of high quality data without the need for adversarial training. In this paper, we introduce the beta noise distribution, which provides flexibility in adjusting noise levels. This is combined with a sinusoidal curriculum that enhances the learning of the trajectory between the noise distribution and the posterior distribution of interest, allowing High Noise Improved Consistency Training (HN-iCT) to be trained in a supervised fashion. Additionally, High Noise Improved Consistency Training with Image Condition (HN-iCT-CN) architecture is introduced, enables to take Low Dose images as a condition for extracting significant features by Weighted Attention Gates (WAG).Our results indicate that unconditional image generation using HN-iCT significantly outperforms basic CT and iCT training techniques with NFE=1 on the CIFAR10 and CelebA datasets. Moreover, our image-conditioned model demonstrates exceptional performance in enhancing low-dose (LD) CT scans.

Ultraviolet flare emission can drive photochemistry in exoplanet atmospheres and even serve as the primary source of uncertainty in atmospheric retrievals. Additionally, flare energy budgets are not well-understood due to a paucity of simultaneous observations. We present new near-UV (NUV) and optical observations of flares from three M dwarfs obtained at 20 s cadence with Swift and TESS, along with a re-analysis of flares from two M dwarfs in order to explore the energy budget and timing of flares at NUV--optical wavelengths. We find a 9000 K blackbody underestimates the NUV flux by $\geq$2$\times$ for 54$\pm$14% of flares and 14.8$\times$ for one flare. We report time lags between the bands of 0.5--6.6 min and develop a method to predict the qualitative flare shape and time lag to 36$\pm$30% accuracy. The scatter present in optical-NUV relations is reduced by a factor of 2.0$\pm$0.6 when comparing the total NUV energy with the TESS energy during the FWHM duration due to the exclusion of the $T_\mathrm{eff}\approx$5000 K tail. We show the NUV light curve can be used to remove flares from the optical light curve and consistently detect planets with 20% smaller transits than is possible without flare detrending. Finally, we demonstrate a 10$\times$ increase in the literature number of multi-wavelength flares with the Early eVolution Explorer (EVE), an astrophysics Small Explorer concept to observe young clusters with simultaneous NUV and optical bands in order to detect young planets, assess their photochemical radiation environments, and observe accretion., Comment: 27 pages, 13 figures, 4 tables, accepted to The Astronomical Journal

Coronary artery disease (CAD), one of the leading causes of mortality worldwide, necessitates effective risk assessment strategies, with coronary artery calcium (CAC) scoring via computed tomography (CT) being a key method for prevention. Traditional methods, primarily based on UNET architectures implemented on pre-built models, face challenges like the scarcity of annotated CT scans containing CAC and imbalanced datasets, leading to reduced performance in segmentation and scoring tasks. In this study, we address these limitations by incorporating the self-supervised learning (SSL) technique of DINO (self-distillation with no labels), which trains without requiring CAC-specific annotations, enhancing its robustness in generating distinct features. The DINO-LG model, which leverages label guidance to focus on calcified areas, achieves significant improvements, with a sensitivity of 89% and specificity of 90% for detecting CAC-containing CT slices, compared to the standard DINO model's sensitivity of 79% and specificity of 77%. Additionally, false-negative and false-positive rates are reduced by 49% and 59%, respectively, instilling greater confidence in clinicians when ruling out calcification in low-risk patients and minimizing unnecessary imaging reviews by radiologists. Further, CAC scoring and segmentation tasks are conducted using a basic UNET architecture, applied specifically to CT slices identified by the DINO-LG model as containing calcified areas. This targeted approach enhances CAC scoring accuracy by feeding the UNET model with relevant slices, significantly improving diagnostic precision, reducing both false positives and false negatives, and ultimately lowering overall healthcare costs by minimizing unnecessary tests and treatments, presenting a valuable advancement in CAD risk assessment., Comment: Developed by Center for Applied Artificial Intelligence (CAAI), University of Kentucky

This paper examines potential biases and inconsistencies in emotional evocation of images produced by generative artificial intelligence (AI) models and their potential bias toward negative emotions. In particular, we assess this bias by comparing the emotions evoked by an AI-produced image to the emotions evoked by prompts used to create those images. As a first step, the study evaluates three approaches for identifying emotions in images -- traditional supervised learning, zero-shot learning with vision-language models, and cross-modal auto-captioning -- using EmoSet, a large dataset of image-emotion annotations that categorizes images across eight emotional types. Results show fine-tuned models, particularly Google's Vision Transformer (ViT), significantly outperform zero-shot and caption-based methods in recognizing emotions in images. For a cross-modality comparison, we then analyze the differences between emotions in text prompts -- via existing text-based emotion-recognition models -- and the emotions evoked in the resulting images. Findings indicate that AI-generated images frequently lean toward negative emotional content, regardless of the original prompt. This emotional skew in generative models could amplify negative affective content in digital spaces, perpetuating its prevalence and impact. The study advocates for a multidisciplinary approach to better align AI emotion recognition with psychological insights and address potential biases in generative AI outputs across digital media.

Identifying stars belonging to different classes is vital in order to build up statistical samples of different phases and pathways of stellar evolution. In the era of surveys covering billions of stars, an automated method of identifying these classes becomes necessary. Many classes of stars are identified based on their emitted spectra. In this paper, we use a combination of the multi-class multi-label Machine Learning (ML) method XGBoost and the PySSED spectral-energy-distribution fitting algorithm to classify stars into nine different classes, based on their photometric data. The classifier is trained on subsets of the SIMBAD database. Particular challenges are the very high sparsity (large fraction of missing values) of the underlying data as well as the high class imbalance. We discuss the different variables available, such as photometric measurements on the one hand, and indirect predictors such as Galactic position on the other hand. We show the difference in performance when excluding certain variables, and discuss in which contexts which of the variables should be used. Finally, we show that increasing the number of samples of a particular type of star significantly increases the performance of the model for that particular type, while having little to no impact on other types. The accuracy of the main classifier is ~0.7 with a macro F1 score of 0.61. While the current accuracy of the classifier is not high enough to be reliably used in stellar classification, this work is an initial proof of feasibility for using ML to classify stars based on photometry.

Prime-Cam, one of the primary instruments for the Fred Young Submillimeter Telescope (FYST) developed by the CCAT Collaboration, will house up to seven instrument modules, with the first operating at 280 GHz. Each module will include three arrays of superconducting microwave kinetic inductance detectors (KIDs). The first KID array fabricated for the 280 GHz module uses titanium-nitride (TiN) as the superconducting material and has 3,456 individual detectors, while the other two arrays use aluminum. This paper presents the design and laboratory characterization of the 280 GHz TiN array, which is cooled below its critical temperature to ~0.1 K and read out over six RF feedlines. LED mapping, a technique for matching the measured resonant frequency of a detector to its physical position, was performed on the array so that the results can be used to lithographically trim the KID capacitors and increase the yield of the array by reducing frequency collisions. We present the methods and results of LED mapping the 280 GHz TiN KID array before deployment on FYST., Comment: 4 pages, 4 figures, submitted to IEEE Transactions on Applied Superconductivity (IEEE TAS)

We present efforts in the fields of machine learning and time series forecasting to accurately predict counts of future opioid overdose incidents recorded by Emergency Medical Services (EMS) in the state of Kentucky. Forecasts are useful to state government agencies to properly prepare and distribute resources related to opioid overdoses effectively. Our approach uses county and district level aggregations of EMS opioid overdose encounters and forecasts future counts for each month. A variety of additional covariates were tested to determine their impact on the model's performance. Models with different levels of complexity were evaluated to optimize training time and accuracy. Our results show that when special precautions are taken to address data sparsity, useful predictions can be generated with limited error by utilizing yearly trends and covariance with additional data sources.

As gravitational-wave observations expand in scope and detection rate, the data analysis infrastructure must be modernized to accommodate rising computational demands and ensure sustainability. We present a scalable gravitational-wave search pipeline which modernizes the GstLAL pipeline by adapting the core filtering engine to the PyTorch framework, enabling flexible execution on both Central Processing Units (CPUs) and Graphics Processing Units (GPUs). Offline search results on the same 8.8 day stretch of public gravitational-wave data indicate that the GstLAL and the PyTorch adaptation demonstrate comparable search performance, even with float16 precision. Lastly, computational benchmarking results show that the GPU float16 configuration of the PyTorch adaptation executed on an A100 GPU can achieve a speedup factor of up to 169 times compared to GstLAL's performance on a single CPU core., Comment: 8 pages, 4 figures

Large Language Models (LLMs) have demonstrated remarkable capabilities in generating coherent text but remain limited by the static nature of their training data. Retrieval Augmented Generation (RAG) addresses this issue by combining LLMs with up-to-date information retrieval, but also expand the attack surface of the system. This paper investigates prompt injection attacks on RAG, focusing on malicious objectives beyond misinformation, such as inserting harmful links, promoting unauthorized services, and initiating denial-of-service behaviors. We build upon existing corpus poisoning techniques and propose a novel backdoor attack aimed at the fine-tuning process of the dense retriever component. Our experiments reveal that corpus poisoning can achieve significant attack success rates through the injection of a small number of compromised documents into the retriever corpus. In contrast, backdoor attacks demonstrate even higher success rates but necessitate a more complex setup, as the victim must fine-tune the retriever using the attacker poisoned dataset., Comment: 12 pages, 5 figures

We conducted systematic radiative transfer (RT) modeling of the Mg II doublet line profiles for 33 low-redshift Lyman continuum (LyC) leakers, and Ly$\alpha$ modeling for a subset of six objects, using a multiphase, clumpy circumgalactic medium (CGM) model. Our RT models successfully reproduced the Mg II line profiles for all 33 galaxies, revealing a necessary condition for strong LyC leakage: high maximum clump outflow velocity ($v_{\rm MgII,\,max} \gtrsim 390\,\rm km\,s^{-1}$) and low total Mg II column density ($N_{\rm MgII,\,tot} \lesssim 10^{14.3}\,\rm cm^{-2}$). We found that the clump outflow velocity and total Mg II column density have the most significant impact on Mg II spectra and emphasized the need for full RT modeling to accurately extract the CGM gas properties. In addition, using archival HST COS/G160M data, we modeled Ly$\alpha$ profiles for six objects and found that their spectral properties do not fully align with the conventional LyC leakage criteria, yet no clear correlation was identified between the modeled parameters and observed LyC escape fractions. We inferred LyC escape fractions based on HI properties from Ly$\alpha$ RT modeling and found that LyC leakage is primarily governed by the number of optically thick HI clumps per sightline ($f_{\rm cl}$). Intriguingly, two galaxies with relatively low observed LyC leakage exhibited the highest RT-inferred LyC escape fractions due to their lowest $f_{\rm cl}$ values, driven by the strong blue peaks of their Ly$\alpha$ emission. Future high-resolution, spatially resolved observations are crucial for resolving this puzzle. Overall, our results support a "picket fence" geometry over a "density-bounded" scenario for the CGM, where a combination of high Mg II outflow velocities and low Mg II column densities may be correlated with the presence of more low-density HI channels that facilitate LyC escape., Comment: 22 pages, 16 figures, comments are welcome

Binary Code Similarity Analysis (BCSA) has a wide spectrum of applications, including plagiarism detection, vulnerability discovery, and malware analysis, thus drawing significant attention from the security community. However, conventional techniques often face challenges in balancing both accuracy and scalability simultaneously. To overcome these existing problems, a surge of deep learning-based work has been recently proposed. Unfortunately, many researchers still find it extremely difficult to conduct relevant studies or extend existing approaches. First, prior work typically relies on proprietary benchmark without making the entire dataset publicly accessible. Consequently, a large-scale, well-labeled dataset for binary code similarity analysis remains precious and scarce. Moreover, previous work has primarily focused on comparing at the function level, rather than exploring other finer granularities. Therefore, we argue that the lack of a fine-grained dataset for BCSA leaves a critical gap in current research. To address these challenges, we construct a benchmark dataset for fine-grained binary code similarity analysis called BinSimDB, which contains equivalent pairs of smaller binary code snippets, such as basic blocks. Specifically, we propose BMerge and BPair algorithms to bridge the discrepancies between two binary code snippets caused by different optimization levels or platforms. Furthermore, we empirically study the properties of our dataset and evaluate its effectiveness for the BCSA research. The experimental results demonstrate that BinSimDB significantly improves the performance of binary code similarity comparison., Comment: Published by International Conference on Security and Privacy in Communication Networks (SecureComm 2024)

We experimentally studied the formation of CH(A) radicals in butane seeded plasma generated with chirp-controlled ultrashort laser pulses (\(\sim 760 \, \mu \text{J}/\text{pulse}\), 890 nm, 1 kHz, 8 fs). The focused beam with high peak intensity (\(\sim 10^{14} - 10^{16} \, \text{W/cm}^2\)) caused Coulomb explosion (CE). The time dependent emission spectra were observed with the Fourier-transform Visible spectroscopy (FTVis) step-scan method. The average signal intensity decreased with the chirp in the Ar\(^+\) > C\(_2\) > H-\(\alpha\) \(\sim\) CH(A) order with a plateau for CH(A) in the \(-200\) to \(-100 \, \text{fs}^2\) range. The short rise time of the CH(A) emission signal, the monoexponential emission decay, and the nearly constant rotational and vibrational temperatures of the CH(A) radicals (\(\sim 3000 \, \text{K}\) and \(\sim 3800 \, \text{K}\)) all support their formation as a primary product. Our TDDFT calculations predict that CH and many other fragments can be formed beyond CE at \(\sim 7 \times 10^{14} \, \text{W/cm}^2\) intensity. The average charge of CH (+0.6) and its relative abundance (0.5\%) support the formation of detectable CH(A) within 120 fs.

Optically pumped magnetometers (OPMs) are compact and lightweight sensors that can measure magnetic fields generated by current flow in neuronal assemblies in the brain. Such sensors enable construction of magnetoencephalography (MEG) instrumentation, with significant advantages over conventional MEG devices including adaptability to head size, enhanced movement tolerance, lower complexity and improved data quality. However, realising the potential of OPMs depends on our ability to perform system calibration, which means finding sensor locations, orientations, and the relationship between the sensor output and magnetic field (termed sensor gain). Such calibration is complex in OPMMEG since, for example, OPM placement can change from subject to subject (unlike in conventional MEG where sensor locations or orientations are fixed). Here, we present two methods for calibration, both based on generating well-characterised magnetic fields across a sensor array. Our first device (the HALO) is a head mounted system that generates dipole like fields from a set of coils. Our second (the matrix coil (MC)) generates fields using coils embedded in the walls of a magnetically shielded room. Our results show that both methods offer an accurate means to calibrate an OPM array (e.g. sensor locations within 2 mm of the ground truth) and that the calibrations produced by the two methods agree strongly with each other. When applied to data from human MEG experiments, both methods offer improved signal to noise ratio after beamforming suggesting that they give calibration parameters closer to the ground truth than factory settings and presumed physical sensor coordinates and orientations. Both techniques are practical and easy to integrate into real world MEG applications. This advances the field significantly closer to the routine use of OPMs for MEG recording., Comment: 36 pages, 10 figures

One of the most challenging problems in the computational study of localization in quantum manybody systems is to capture the effects of rare events, which requires sampling over exponentially many disorder realizations. We implement an efficient procedure on a quantum processor, leveraging quantum parallelism, to efficiently sample over all disorder realizations. We observe localization without disorder in quantum many-body dynamics in one and two dimensions: perturbations do not diffuse even though both the generator of evolution and the initial states are fully translationally invariant. The disorder strength as well as its density can be readily tuned using the initial state. Furthermore, we demonstrate the versatility of our platform by measuring Renyi entropies. Our method could also be extended to higher moments of the physical observables and disorder learning.

We introduce a new capability of the Neil Gehrels Swift Observatory, dubbed `continuous commanding,' achieving 10 seconds latency response time on-orbit to unscheduled Target of Opportunity requests. This allows Swift to respond to early warning gravitational-wave detections, rapidly slewing the Burst Alert Telescope (BAT) across the sky to place the GW origin in the BAT field of view at merger time. This will dramatically increase the GW/GRB co-detection rate, and enable prompt arcminute localization of a neutron star merger. We simulate the full Swift response to a GW early warning alert, including input sky maps produced at different warning times, a complete model of Swift's attitude control system, and a full accounting of the latency between the GW detectors and the spacecraft. 60 s of early warning doubles the rate of prompt GRB detections with arcminute position, and 140 s guarantees observation anywhere on the unocculted sky, even with localization areas >> 1000 deg$^2$. While 140 s is beyond current gravitational wave detector sensitivities, 30-70 s is achievable today. We show that the detection yield is now limited by the latency of LIGO/Virgo cyber-infrastructure, and motivate focus on its reduction. Continuous commanding is now a general capability of Swift, significantly increasing its versatility in response to the growing demands of time-domain astrophysics. We demonstrate this potential on an externally triggered Fast Radio Burst, slewing 81 degrees across the sky, and collecting X-ray and UV photons from the source position < 150 s after the trigger was received from the Canadian Hydrogen Intensity Mapping Experiment (CHIME), thereby setting the earliest and deepest such constraints on high energy activity from non-repeating FRBs. The Swift Team invites proposals for novel scientific applications of ultra-low latency UV, X-ray, and gamma-ray observations., Comment: Update to version accepted for publication in ApJL. Comments, questions, and new ideas for applications of ultra-low latency UV, X-ray, and gamma-ray observations are welcome!

Fragmentation dynamics in the Coulomb explosion of hydrocarbons, specifically methane, ethane, propane, and butane, are investigated using time dependent density functional theory (TDDFT) simulations. The goal of this work is to elucidate the distribution of fragments generated under laser-driven Coulomb explosion conditions. Detailed analysis reveals the types of fragments formed, their respective charge states, and the optimal laser intensities required for achieving various fragmentations. Our results indicate distinct fragmentation patterns for each hydrocarbon, correlating with the molecular structure and ionization potential. Additionally, we identify the laser parameters that maximize fragmentation efficiency, providing valuable insights for experimental setups. This research advances our understanding of Coulomb explosion mechanisms and offers a foundation for further studies in controlled molecular fragmentation.

Previous work has shown that item response theory may be used to rank incorrect response options to multiple-choice items on commonly used assessments. This work has shown that, when the correct response to each item is specified, a nominal response model (NRM) may be used to rank the incorrect options. We seek to determine the robustness of these results by applying the NRM to all response choices, without specifying the correct response. We apply these analyses to multiple data sets (each with more than 9,000 response sets), including pre-instruction and post-instruction responses. We find that the rankings generated without specifying the correct response are consistent with the previously published rankings for one data set; however, we find noticeable differences between rankings generated from different data sets. We provide evidence that discrepancies may result from differences in response rates for less commonly chosen responses.

Making successful use of cloud computing for deploying scalable web applications requires nuanced approaches to both system design and deployment methodology, involving reasoning about the elasticity, cost, and security models of cloud services. Students commonly interact with cloud abstractions early in their technical careers, including during internships and academic research. Building cloud-native applications without a firm understanding of the fundamentals of cloud engineering can leave students susceptible to cost and security pitfalls common to cloud platforms. Yet, cloud computing is not commonly taught at the undergraduate level, because the technology and practices behind modern cloud deployment, such as containerization and infrastructure-as-code (IaC), have only recently matured into a set of general principles independent from specific providers' offerings. To address this gap, we designed an undergraduate-level course around these principles that framed cloud infrastructure deployment as a software engineering practice in support of scalable web applications, emphasizing the value of both cloud deployment and application design skills in building robust cloud-native systems. Our course featured a number of hands-on assignments that gave students experience with modern, best-practice concepts and tools such as IaC, containerization, observability, serverless computing, and continuous integration and deployment. We describe the design of the course, our experience teaching its initial offering in Winter 2024, and provide our reflections on what worked well and potential areas for improvement. Our course material is publicly available at https://infracourse.cloud., Comment: Accepted to SIGCSE TS 2025

We characterize the $L^2(\mathbb{R}^n)$-compact operators within a wide class of pseudodifferential operators. This result follows from a general theory for localized operators on Hilbert spaces equipped with a Parseval frame, which we develop and apply to a more general collection of pseudodifferential operators., Comment: 11 pages

Galactic outflows provide important feedback effects to regulate the evolution of the host galaxies. Two primary diagnostics of galactic outflows are broad and/or blueshifted emission and absorption lines. Even though well-established methods exist to analyze these outflow signatures, connections between them are rarely studied and largely unknown. In this paper, we present the first detailed comparisons of the outflow properties measured independently from the two outflow diagnostics for a sample of 33 low-redshift star-forming galaxies. Their UV absorption lines are detected by the Hubble Space Telescope/Cosmic Origin Spectrograph, and optical emission lines are observed by the Keck/Echellette Spectrograph and Imager. We find that several outflow properties derived from emission and absorption lines are tightly correlated. These include outflow maximum velocity, line width, and sizes. Specifically, in a given galaxy, outflows seen in emission lines have smaller maximum velocities, narrower line widths, and smaller sizes than those measured from the absorption lines. These findings can be interpreted by the fact that emission line luminosity is weighted by density squared, while absorption line depth is weighted by density. We then test both spherical and bi-conical outflow models, and find the same outflow velocity and density distributions can explain the observed outflow features in emission and absorption lines for individual galaxies. These results provide novel calibration between galactic outflow properties measured from the two diagnostics and provide valuable insights for future models of galactic outflows by potentially doubling the number of observational constraints., Comment: 3 tables, 17 figures, 30 pages, submitted to ApJ

In this paper, we develop an embodied AI system for human-in-the-loop navigation with a wheeled mobile robot. We propose a direct yet effective method of monitoring the robot's current plan to detect changes in the environment that impact the intended trajectory of the robot significantly and then query a human for feedback. We also develop a means to parse human feedback expressed in natural language into local navigation waypoints and integrate it into a global planning system, by leveraging a map of semantic features and an aligned obstacle map. Extensive testing in simulation and physical hardware experiments with a resource-constrained wheeled robot tasked to navigate in a real-world environment validate the efficacy and robustness of our method. This work can support applications like precision agriculture and construction, where persistent monitoring of the environment provides a human with information about the environment state.

This study reports a comprehensive environmental scan of the generative AI (GenAI) infrastructure in the national network for clinical and translational science across 36 institutions supported by the Clinical and Translational Science Award (CTSA) Program led by the National Center for Advancing Translational Sciences (NCATS) of the National Institutes of Health (NIH) at the United States. With the rapid advancement of GenAI technologies, including large language models (LLMs), healthcare institutions face unprecedented opportunities and challenges. This research explores the current status of GenAI integration, focusing on stakeholder roles, governance structures, and ethical considerations by administering a survey among leaders of health institutions (i.e., representing academic medical centers and health systems) to assess the institutional readiness and approach towards GenAI adoption. Key findings indicate a diverse range of institutional strategies, with most organizations in the experimental phase of GenAI deployment. The study highlights significant variations in governance models, with a strong preference for centralized decision-making but notable gaps in workforce training and ethical oversight. Moreover, the results underscore the need for a more coordinated approach to GenAI governance, emphasizing collaboration among senior leaders, clinicians, information technology staff, and researchers. Our analysis also reveals concerns regarding GenAI bias, data security, and stakeholder trust, which must be addressed to ensure the ethical and effective implementation of GenAI technologies. This study offers valuable insights into the challenges and opportunities of GenAI integration in healthcare, providing a roadmap for institutions aiming to leverage GenAI for improved quality of care and operational efficiency.

We introduce a new low-power 13-mg microactuator driven by shape-memory alloy (SMA) wires for underwater operation. The development of this device was motivated by the recent creation of microswimmers such as the FRISHBot, WaterStrider, VLEIBot, VLEIBot+, and VLEIBot++. The first four of these robots, ranging from 30 to 90 mg, function tethered to an electrical power supply while the last platform is an 810-mg fully autonomous system. These five robots are driven by dry SMA-based microactuators first developed for microrobotic crawlers such as the SMALLBug and SMARTI. As shown in this abstract, dry SMA-based actuators do not operate efficiently under water due to high heat-transfer rates in this medium; for example, the actuators that drive the VLEIBot++ require about 40 mW of average power at 1 Hz in dry air while requiring about 900 mW of average power at 1 Hz in water. In contrast, the microactuator presented in this abstract consumes about 150 mW of average power at 1 Hz in both dry air and water; additionally, it can be excited directly using an onboard battery through simple power electronics implemented on a custom-built printed circuit board (PCB). This technological breakthrough was enabled by the integration of a soft structure that encapsulates the SMA wires that drive the actuator in order to passively control the rates of heat transfer. The results presented here represent preliminary, yet compelling, experimental evidence that the proposed actuation approach will enable the development of fully autonomous and controllable submersible microswimmers. To accomplish this objective, we will evolve the current version of the VLEIBot++ and introduce new bioinspired underwater propulsion mechanisms., Comment: Presented at 40th Anniversary of the IEEE International Conference on Robotics and Automation (ICRA@40)

We refine ideas from Knill 1996, Jones 2016, Chamberland 2020, Gidney 2023+2024, Bombin 2024, and Hirano 2024 to efficiently prepare good $|T\rangle$ states. We call our construction "magic state cultivation" because it gradually grows the size and reliability of one state. Cultivation fits inside a surface code patch and uses roughly the same number of physical gates as a lattice surgery CNOT gate of equivalent reliability. We estimate the infidelity of cultivation (from injection to idling at distance 15) using a mix of state vector simulation, stabilizer simulation, error enumeration, and Monte Carlo sampling. Compared to prior work, cultivation uses an order of magnitude fewer qubit-rounds to reach logical error rates as low as $2 \cdot 10^{-9}$ when subjected to $10^{-3}$ uniform depolarizing circuit noise. Halving the circuit noise to $5 \cdot 10^{-4}$ improves the achievable logical error rate to $4 \cdot 10^{-11}$. Cultivation's efficiency and strong response to improvements in physical noise suggest that further magic state distillation may never be needed in practice.

Lattice gauge theories (LGTs) can be employed to understand a wide range of phenomena, from elementary particle scattering in high-energy physics to effective descriptions of many-body interactions in materials. Studying dynamical properties of emergent phases can be challenging as it requires solving many-body problems that are generally beyond perturbative limits. We investigate the dynamics of local excitations in a $\mathbb{Z}_2$ LGT using a two-dimensional lattice of superconducting qubits. We first construct a simple variational circuit which prepares low-energy states that have a large overlap with the ground state; then we create particles with local gates and simulate their quantum dynamics via a discretized time evolution. As the effective magnetic field is increased, our measurements show signatures of transitioning from deconfined to confined dynamics. For confined excitations, the magnetic field induces a tension in the string connecting them. Our method allows us to experimentally image string dynamics in a (2+1)D LGT from which we uncover two distinct regimes inside the confining phase: for weak confinement the string fluctuates strongly in the transverse direction, while for strong confinement transverse fluctuations are effectively frozen. In addition, we demonstrate a resonance condition at which dynamical string breaking is facilitated. Our LGT implementation on a quantum processor presents a novel set of techniques for investigating emergent particle and string dynamics.

Water is often overused in irrigation, making efficient management of it crucial. Precision Agriculture emphasizes tools like stem water potential (SWP) analysis for better plant status determination. However, such tools often require labor-intensive in-situ sampling. Automation and machine learning can streamline this process and enhance outcomes. This work focused on automating stem detection and xylem wetness classification using the Scholander Pressure Chamber, a widely used but demanding method for SWP measurement. The aim was to refine stem detection and develop computer-vision-based methods to better classify water emergence at the xylem. To this end, we collected and manually annotated video data, applying vision- and learning-based methods for detection and classification. Additionally, we explored data augmentation and fine-tuned parameters to identify the most effective models. The identified best-performing models for stem detection and xylem wetness classification were evaluated end-to-end over 20 SWP measurements. Learning-based stem detection via YOLOv8n combined with ResNet50-based classification achieved a Top-1 accuracy of 80.98%, making it the best-performing approach for xylem wetness classification.

We numerically study two-dimensional active nematics with periodic activity patterning. For stripes of activity, we observe a transition from two-dimensional to one-dimensional active turbulence as the maximum active force and distance between activity stripes increases, followed by a transition to stable vortices ordered antiferromagnetically along the stripes and ferromagnetically transverse to the stripes. If the stripes are replaced by columns of activity circles, the vortex ordering is lost. A triangular lattice of activity circles can induce the formation of two-dimensional active turbulence. Our results provide a mechanism for inducing non-equilibrium phase transitions in active nematics using activity inhomogeneity, which can be exploited to create activity patterned ordered phases., Comment: 6 + 9 pages, 5 + 3 figures

Administrative documentation is a major driver of rising healthcare costs and is linked to adverse outcomes, including physician burnout and diminished quality of care. This paper introduces a secure system that applies recent advancements in speech-to-text transcription and speaker-labeling (diarization) to patient-provider conversations. This system is optimized to produce accurate transcriptions and highlight potential errors to promote rapid human verification, further reducing the necessary manual effort. Applied to over 40 hours of simulated conversations, this system offers a promising foundation for automating clinical transcriptions., Comment: 7 pages, 6 figures

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!

One of the fundamental questions of cosmology is the origin and mechanism(s) responsible for the reionization of the Universe beyond $z\sim6$. To address this question, many studies over the past decade have focused on local ($z\sim0.3$) galaxies which leak ionizing radiation (Lyman continuum or LyC). However, line-of-sight effects and data quality have prohibited deeper insight into the nature of LyC escape. To circumvent these limitations, we analyze stacks of a consolidated sample of {\it HST}/COS observations of the LyC in 89 galaxies at $z\sim0.3$. From fitting of the continuum, we obtain information about the underlying stellar populations and neutral ISM geometry. We find that most LyC non-detections are not leaking appreciable LyC ($f_{esc}^{\rm LyC}<1$\%) but also that exceptional cases point to spatial variations in the LyC escape fraction $f_{esc}^{\rm LyC}$. Stellar populations younger than 3 Myr lead to an increase in ionizing feedback, which in turn increases the isotropy of LyC escape. Moreover, mechanical feedback from supernovae in 8-10 Myr stellar populations is important for anisotropic gas distributions needed for LyC escape. While mechanical feedback is necessary for any LyC escape, high $f_{esc}^{\rm LyC}$ ($>5$\%) also requires a confluence of young stars and ionizing feedback. A two-stage burst of star formation could facilitate this optimal LyC escape scenario., Comment: Submitted to AAS publications, 42 pages, 25 figures

orbitize! is a package for Bayesian modeling of the orbital parameters of resolved binary objects from time series measurements. It was developed with the needs of the high-contrast imaging community in mind, and has since also become widely used in the binary star community. A generic orbitize! use case involves translating relative astrometric time series, optionally combined with radial velocity or astrometric time series, into a set of derived orbital posteriors. This paper is published alongside the release of orbitize! version 3.0, which has seen significant enhancements in functionality and accessibility since the release of version 1.0 (Blunt et al., 2020)., Comment: Published in JOSS

This paper proposes a novel spiking artificial neuron design based on a combined spin valve/magnetic tunnel junction (SV/MTJ). Traditional hardware used in artificial intelligence and machine learning faces significant challenges related to high power consumption and scalability. To address these challenges, spintronic neurons, which can mimic biologically inspired neural behaviors, offer a promising solution. We present a model of an SV/MTJ-based neuron which uses technologies that have been successfully integrated with CMOS in commercially available applications. The operational dynamics of the neuron are derived analytically through the Landau-Lifshitz-Gilbert-Slonczewski (LLGS) equation, demonstrating its ability to replicate key spiking characteristics of biological neurons, such as response latency and refractive behavior. Simulation results indicate that the proposed neuron design can operate on a timescale of about 1 ns, without any bias current, and with power consumption as low as 50 uW., Comment: 7 pages, 5 figures, letter

We prove weighted weak-type $(r,r)$ estimates for operators satisfying $(r,s)$ limited-range sparse domination of $\ell^q$-type. Our results contain improvements for operators satisfying limited-range and square function sparse domination. In the case of operators $T$ satisfying standard sparse form domination such as Calder\'on-Zygmund operators, we provide a new and simple proof of the sharp bound $$ \|T\|_{L^1_w(\mathbf{R}^d)\rightarrow L^{1,\infty}_w(\mathbf{R}^d)} \lesssim [w]_1(1+\log [w]_{\text{FW}}). $$, Comment: 13 pages

Crown-like structures (CLS) in breast adipose tissue are formed as a result of macrophages clustering around necrotic adipocytes in specific patterns. As a histologic marker of local inflammation, CLS could have potential diagnostic utility as a biomarker for breast cancer risk. However, given the scale of whole slide images and the rarity of CLS (a few cells in an entire tissue sample), microscope-based manual identification is a challenge for the pathologist. In this report, we describe an artificial intelligence pipeline to solve this needle-in-a-haystack problem. We developed a zero-cost, zero-footprint web platform to enable remote operation on digital whole slide imaging data directly in the web browser, supporting collaborative annotation of the data by multiple experts. The annotated images then allow for incremental training and fine tuning of deep neural networks via active learning. The platform is reusable and requires no backend or installations, thus ensuring the data remains secure and private under the governance of the end user. Using this platform, we iteratively trained a CLS identification model, evaluating the performance after each round and adding examples to the training data to overcome failure cases. The resulting model, with an AUC of 0.90, shows promise as a first-pass screening tool to detect CLS in breast adipose tissue, considerably reducing the workload of the pathologist. Platform available at: https://episphere.github.io/path

The Epoch of Reionization Spectrometer (EoR-Spec) is an upcoming Line Intensity Mapping (LIM) instrument designed to study the evolution of the early universe (z = 3.5 to 8) by probing the redshifted [CII] 158 $\mu$m fine-structure line from aggregates of galaxies. The [CII] emission is an excellent tracer of star formation since it is the dominant cooling line from neutral gas heated by OB star light and thus can be used to probe the reionization of the early Universe due to star formation. EoR-Spec will be deployed on Prime-Cam, a modular direct-detection receiver for the 6-meter Fred Young Submillimeter Telescope (FYST), currently under construction by CPI Vertex Antennentechnik GmbH and to be installed near the summit of Cerro Chajnantor in the Atacama Desert. This instrument features an image plane populated with more than 6500 Microwave Kinetic Inductance Detectors (MKIDs) that are illuminated by a 4-lens optical design with a cryogenic, scanning Fabry-Perot Interferometer (FPI) at the pupil of the optical system. The FPI is designed to provide a spectral resolving power of $R\sim100$ over the full spectral range of 210--420 GHz. EoR-Spec will tomographically survey the E-COSMOS and E-CDFS fields with a depth of about 4000 hours over a 5 year period. Here we give an update on EoR-Spec's final mechanical/optical design and the current status of fabrication, characterization and testing towards first light in 2026., Comment: Presented at SPIE Millimeter, Submillimeter, and Far-Infrared Detectors and Instrumentation for Astronomy XII

A set of vertices of a graph is said to be in general position if no three vertices from the set lie on a common geodesic. Recently Klav\v{z}ar, Rall and Yero generalized this notion by defining a set of vertices to be in general $d$-position if no three vertices from the set lie on a common geodesic of length at most $d$. We generalize this notion further by defining a set of vertices to be in $k$-general $d$-position if no $k$ vertices of the set lie on a common geodesic of length at most $d$. The $k$-general $d$-position number of a graph is the largest cardinality of a $k$-general $d$-position set. We provide upper and lower bounds on the $k$-general $d$-position number of graphs in terms of the $k$-general $d$-position number of certain kinds of subgraphs. We compute the $k$-general $d$-position number of finite paths and cycles. Along the way we establish that the maximally even subsets of cycles, which were introduced in Clough and Douthett's work on music theory, provide the largest possible $k$-general $d$-position sets in $n$-cycles. We generalize Klav\v{z}ar and Manuel's notion of monotone-geodesic labeling to that of $k$-monotone-geodesic labeling in order to calculate the $k$-general $d$-position number of the infinite two-dimensional grid. We also prove a formula for the $k$-general $d$-position number of certain thin finite grids, providing a partial answer to a question asked by Klav\v{z}ar, Rall and Yero.

Feedback is widely recognized as an essential condition for Lyman continuum (LyC) escape in star-forming galaxies. However, the mechanisms by which galactic outflows clear neutral gas and dust remain unclear. In this paper, we model the Mg II 2796\r{A}, 2804\r{A} absorption + emission lines in 29 galaxies taken from the Low-z LyC Survey (LzLCS) to investigate the impact of (radiation + mechanical) feedback on LyC escape. Using constraints on Mg$^+$ and photoionization models, we map the outflows' neutral hydrogen content and predict $f_{esc}^{LyC}$ with a multiphase wind model. We measure mass, momentum, and energy loading factors for the neutral winds, which carry up to 10% of the momentum and 1% of the energy in SFR-based deposition rates. We use SED template fitting to determine the relative ages of stellar populations, allowing us to identify radiation feedback dominant systems. We then examine feedback related properties (stellar age, loading factors, etc.) under conditions that optimize feedback efficiency, specifically high star formation rate surface density and compact UV half-light radii. Our findings indicate that the strongest leakers are radiation feedback dominant, lack Mg II outflows, but have extended broad components in higher ionization lines like [O III] 5007\r{A}, as observed by Amor\'in et al. (2024). In contrast, galaxies experiencing supernovae feedback typically exhibit weaker $f_{esc}^{LyC}$ and show evidence of outflows in both Mg II and higher ionization lines. We attribute these findings to rapid or "catastrophic" cooling in the radiation-dominant systems, which, given the low metallicities in our sample, are likely experiencing delayed supernovae., Comment: 34 pages, 16 figures, 7 tables

Teleoperation of humanoid robots has long been a challenging domain, necessitating advances in both hardware and software to achieve seamless and intuitive control. This paper presents an integrated solution based on several elements: calibration-free motion capture and retargeting, low-latency fast whole-body kinematics streaming toolbox and high-bandwidth cycloidal actuators. Our motion retargeting approach stands out for its simplicity, requiring only 7 IMUs to generate full-body references for the robot. The kinematics streaming toolbox, ensures real-time, responsive control of the robot's movements, significantly reducing latency and enhancing operational efficiency. Additionally, the use of cycloidal actuators makes it possible to withstand high speeds and impacts with the environment. Together, these approaches contribute to a teleoperation framework that offers unprecedented performance. Experimental results on the humanoid robot Nadia demonstrate the effectiveness of the integrated system.

The EXperiment for Cryogenic Large-Aperture Intensity Mapping (EXCLAIM) is a balloon-borne telescope designed to survey star formation over cosmological time scales using intensity mapping in the 420 - 540 GHz frequency range. EXCLAIM uses a fully cryogenic telescope coupled to six on-chip spectrometers featuring kinetic inductance detectors (KIDs) to achieve high sensitivity, allowing for fast integration in dark atmospheric windows. The telescope receiver is cooled to $\approx$ 1.7 K by immersion in a superfluid helium bath and enclosed in a superfluid-tight shell with a meta-material anti-reflection coated silicon window. In addition to the optics and the spectrometer package, the receiver contains the magnetic shielding, the cryogenic segment of the spectrometer readout, and the sub-Kelvin cooling system. A three-stage continuous adiabatic demagnetization refrigerator (CADR) keeps the detectors at 100 mK while a $^4$He sorption cooler provides a 900 mK thermal intercept for mechanical suspensions and coaxial cables. We present the design of the EXCLAIM receiver and report on the flight-like testing of major receiver components, including the superfluid-tight receiver window and the sub-Kelvin coolers., Comment: 13 pages, 7 figures, 1 table

Prime-Cam, a first-generation science instrument for the Atacama-based Fred Young Submillimeter Telescope, is being built by the CCAT Collaboration to observe at millimeter and submillimeter wavelengths using kinetic inductance detectors (KIDs). Prime-Cam's 280 GHz instrument module will deploy with two aluminum-based KID arrays and one titanium nitride-based KID array, totaling approximately 10,000 detectors at the focal plane, all of which have been fabricated and are currently undergoing testing. One complication of fielding large arrays of KIDs under dynamic loading conditions is tuning the detector tone powers to maximize signal-to-noise while avoiding bifurcation due to the nonlinear kinetic inductance. For aluminum-based KIDs, this is further complicated by additional nonlinear effects which couple tone power to resonator quality factors and resonant frequencies. While both nonequilibrium quasiparticle dynamics and two-level system fluctuations have been shown to give rise to qualitatively similar distortions, modeling these effects alongside nonlinear kinetic inductance is inefficient when fitting thousands of resonators on-sky with existing models. For this reason, it is necessary to have a detailed understanding of the nonlinear effects across relevant detector loading conditions, including how they impact on on-sky noise and how to diagnose the detector's relative performance. We present a study of the competing nonlinearities seen in Prime-Cam's 280 GHz aluminum KIDs, with a particular emphasis on the resulting distortions to the resonator line shape and how these impact detector parameter estimation., Comment: 10 pages, 4 figures, Conference proceedings from SPIE Astronomical Telescopes + Instrumentation (AS24)

Quantum error correction provides a path to reach practical quantum computing by combining multiple physical qubits into a logical qubit, where the logical error rate is suppressed exponentially as more qubits are added. However, this exponential suppression only occurs if the physical error rate is below a critical threshold. In this work, we present two surface code memories operating below this threshold: a distance-7 code and a distance-5 code integrated with a real-time decoder. The logical error rate of our larger quantum memory is suppressed by a factor of $\Lambda$ = 2.14 $\pm$ 0.02 when increasing the code distance by two, culminating in a 101-qubit distance-7 code with 0.143% $\pm$ 0.003% error per cycle of error correction. This logical memory is also beyond break-even, exceeding its best physical qubit's lifetime by a factor of 2.4 $\pm$ 0.3. We maintain below-threshold performance when decoding in real time, achieving an average decoder latency of 63 $\mu$s at distance-5 up to a million cycles, with a cycle time of 1.1 $\mu$s. To probe the limits of our error-correction performance, we run repetition codes up to distance-29 and find that logical performance is limited by rare correlated error events occurring approximately once every hour, or 3 $\times$ 10$^9$ cycles. Our results present device performance that, if scaled, could realize the operational requirements of large scale fault-tolerant quantum algorithms., Comment: 10 pages, 4 figures, Supplementary Information

We present a method, called matching synthesis, for decoding quantum codes that produces an enhanced assignment of errors from an ensemble of decoders. We apply matching synthesis to develop a decoder named Libra, and show in simulations that Libra increases the error-suppression ratio $\Lambda$ by about $10\%$. Matching synthesis takes the solutions of an ensemble of approximate solvers for the minimum-weight hypergraph matching problem, and produces a new solution that combines the best local solutions, where locality depends on the hypergraph. We apply matching synthesis to an example problem of decoding surface codes with error correlations in the conventional circuit model, which induces a hypergraph with hyperedges that are local in space and time. We call the matching-synthesis decoder Libra, and in this example the ensemble consists of correlated minimum-weight matching using a different hypergraph with randomly perturbed error probabilities for each ensemble member. Furthermore, we extend matching synthesis to perform summation of probability for multiple low-weight solutions and at small computational overhead, approximating the probability of an equivalence class; in our surface code problem, this shows a modest additional benefit. We show that matching synthesis has favorable scaling properties where accuracy begins to saturate with an ensemble size of 60, and we remark on pathways to real-time decoding at near-optimal decoding accuracy if one has an accurate model for the distribution of errors.

The initial stellar carbon-to-oxygen (C/O) ratio can have a large impact on the resulting condensed species present in the protoplanetary disk and, hence, the composition of the bodies and planets that form. The observed C/O ratios of stars can vary from 0.1-2. We use a sequential dust condensation model to examine the impact of the C/O ratio on the composition of solids around a Solar-like star. We utilize this model in a focused examination of the impact of varying the initial stellar C/O ratio to isolate the effects of the C/O ratio in the context of Solar-like stars. We describe three different system types in our findings. The Solar system falls into the silicate-dominant, low C/O ratio systems which end at a stellar C/O ratio somewhere between 0.52 and 0.6. At C/O ratios between about 0.6 and 0.9, we have intermediate systems. Intermediate systems show a decrease in silicates while carbides begin to become significant. Carbide-dominant systems begin around a C/O ratio of 0.9. Carbide-dominant systems exhibit high carbide surface densities at inner radii with comparable levels of carbides and silicates at outer radii. Our models show that changes between C/O=0.8 and C/O=1 are more significant than previous studies, that carbon can exceed 80% of the condensed mass, and that carbon condensation can be significant at radii up to 6 AU., Comment: Submitted

Electromagnetic follow-up observations of gravitational wave events offer critical insights and provide significant scientific gain from this new class of astrophysical transients. Accurate identification of gravitational wave candidates and rapid release of sky localization information are crucial for the success of these electromagnetic follow-up observations. However, searches for gravitational wave candidates in real time suffer a non-negligible false alarm rate. By leveraging the sky localization information and other metadata associated with gravitational wave candidates, GWSkyNet, a machine learning classifier developed by Cabero et al. (2020), demonstrated promising accuracy for the identification of the origin of event candidates. We improve the performance of the classifier for LIGO-Virgo-KAGRA's fourth observing run by reviewing and updating the architecture and features used as inputs by the algorithm. We also retrain and fine-tune the classifier with data from the third observing run. To improve the prospect of electromagnetic follow-up observations, we incorporate GWSkyNet into LIGO-Virgo-KAGRA's low-latency infrastructure as an automatic pipeline for the evaluation of gravitational wave alerts in real time. We test the readiness of the algorithm on a LIGO-Virgo-KAGRA mock data challenge campaign. The results show that by thresholding on the GWSkyNet score, noise masquerading as astrophysical sources can be rejected efficiently and the majority of true astrophysical signals correctly identified.

We develop an approximate, analytical model for the velocity of defects in active nematics by combining recent results for the velocity of topological defects in nematic liquid crystals with the flow field generated from individual defects in active nematics. Importantly, our model takes into account the long-range interactions between defects that result from the flows they produce as well as the orientational coupling between defects inherent in nematics. Our work complements previous studies of active nematic defect motion by introducing a linear approximation that allows us to treat defect interactions as two-body interactions and incorporates the hydrodynamic screening length as a tuning parameter. We show that the model can analytically predict bound states between two $+1/2$ winding number defects, effective attraction between two $-1/2$ defects, and the scaling of a critical unbinding length between $\pm 1/2$ defects with activity. The model also gives predictions for the trajectories of defects, such as the scattering of $+1/2$ defects by $-1/2$ defects at a critical impact parameter that depends on activity. In the presence of circular confinement, the model predicts a braiding motion for three $+1/2$ defects that was recently seen in experiments, as well as stable and ergodic trajectories for four or more defects., Comment: 17 pages, 10 figures

Autonomous racing serves as a critical platform for evaluating automated driving systems and enhancing vehicle mobility intelligence. This work investigates offline reinforcement learning methods to train agents within the dynamic F1tenth racing environment. The study begins by exploring the challenges of online training in the Austria race track environment, where agents consistently fail to complete the laps. Consequently, this research pivots towards an offline strategy, leveraging `expert' demonstration dataset to facilitate agent training. A waypoint-based suboptimal controller is developed to gather data with successful lap episodes. This data is then employed to train offline learning-based algorithms, with a subsequent analysis of the agents' cross-track performance, evaluating their zero-shot transferability from seen to unseen scenarios and their capacity to adapt to changes in environment dynamics. Beyond mere algorithm benchmarking in autonomous racing scenarios, this study also introduces and describes the machinery of our return-conditioned decision tree-based policy, comparing its performance with methods that employ fully connected neural networks, Transformers, and Diffusion Policies and highlighting some insights into method selection for training autonomous agents in driving interactions.