Search

In real-world scenarios, it is desirable for embodied agents to have the ability to leverage human language to gain explicit or implicit knowledge for learning tasks. Despite recent progress, most previous approaches adopt simple low-level instructions as language inputs, which may not reflect natural human communication. It's not clear how to incorporate rich language use to facilitate task learning. To address this question, this paper studies different types of language inputs in facilitating reinforcement learning (RL) embodied agents. More specifically, we examine how different levels of language informativeness (i.e., feedback on past behaviors and future guidance) and diversity (i.e., variation of language expressions) impact agent learning and inference. Our empirical results based on four RL benchmarks demonstrate that agents trained with diverse and informative language feedback can achieve enhanced generalization and fast adaptation to new tasks. These findings highlight the pivotal role of language use in teaching embodied agents new tasks in an open world. Project website: https://github.com/sled-group/Teachable_RL, Comment: EMNLP 2024 Main. Project website: https://github.com/sled-group/Teachable_RL

Deneb, the prototype alpha Cygni variable, is a blue-white supergiant that shows irregular variability with quasi-period around 12 days in brightness and radial velocity. Abt et al. (2023) found that larger amplitude 12-day variations appear to resume abruptly and at an arbitrary phase and damp out after several cycles, with an interval of around 70 days between these resumptions. Here we make use of an 8.6-year photometric data set for Deneb from the Solar Mass Ejection Imager (SMEI) to better characterize this behavior. We find that the interval between pulsation resumptions is not exact, with the most common intervals between 100 and 120 days. Sometimes one or more intervals are skipped. We also examine AAVSO and Transiting Exoplanet Survey Satellite (TESS) light curves for alpha Cyg variables Rigel, Saiph, and Alnilam in Orion, Aludra in Canis Major, and 6 Cas to compare with the behavior of alpha Cyg. Except for 6 Cas, the time series are too short, or the observations too infrequent to draw any conclusions about similarities between the behavior of these stars and alpha Cyg. We also summarize results of evolution and pulsation modeling for Deneb and alpha Cyg variables from the literature. The alpha Cyg variables may not be a homogenous group with a common mechanism for their variability. It has not been determined whether they are on the first crossing of the Hertzsprung-Russell diagram toward the red supergiant phase or are on their second crossing after having been red supergiants. Future plans include examining BRITE Constellation data for Deneb, processing SMEI data for other bright alpha Cyg variables, and comparing 6 Cas light curves from AAVSO and TESS data taken concurrently., Comment: 9 pages, 18 figures, submitted to Proceedings for the 43rd Annual Conference of the Society for Astronomical Sciences SAS-2024 Symposium on Telescope Science, eds. John C. Martin, Robert K. Buchheim, Robert M. Gill, Wayne Green, and John Menke, https://socastrosci.org/wp-content/uploads/2024/08/2024-Proceedings_Ver1.3d.pdf

Deneb (alpha Cygni) is a bright (V magnitude 1.25) blue-white supergiant (spectral type A2 Ia) which shows variability in both radial velocity and photometric measurements. H. Abt reviewed radial velocity measurements by Paddock (1935) using the Lick observatory 36-inch telescope spectrograph during 1927-1935. Abt noticed resumptions of pulsations with a dominant quasi-period of around 12 days that occur at intervals of around 70 days, and damp out after a few cycles. These resumptions appear to happen at arbitrary phase. Perhaps another event like this was captured in a shorter series of radial velocity measurements by Abt in 1956. We examined subsequent radial velocity and photometric data available in the literature, along with photometric measurements by the TESS spacecraft and V-magnitude observations by AAVSO observers. We find some evidence for periodic resumptions of larger-amplitude pulsations in these data. However, longer contiguous data sets combined with more frequent sampling are needed to confirm this periodicity in resumption of pulsations, and to help answer many more questions about Deneb and the alpha Cygni variables., Comment: 10 pages, 10 figures. Submitted to Proceedings of the 112th Annual Meeting of The American Association of Variable Star Observers, ed. Joyce A. Guzik, November 3-5, 2023, Somerville, MA, ISBN 978-1-939538-68-0, available at https://www.aavso.org/112

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

Many online platforms maintain user profiles for personalization. Unfortunately, these profiles are typically not interpretable or easily modifiable by the user. To remedy this shortcoming, we explore natural language-based user profiles, as they promise enhanced transparency and scrutability of recommender systems. While existing work has shown that language-based profiles from standard LLMs can be effective, such generalist LLMs are unlikely to be optimal for this task. In this paper, we introduce LangPTune, the first end-to-end learning method for training LLMs to produce language-based user profiles that optimize recommendation effectiveness. Through comprehensive evaluations of LangPTune across various training configurations and benchmarks, we demonstrate that our approach significantly outperforms existing profile-based methods. In addition, it approaches performance levels comparable to state-of-the-art, less transparent recommender systems, providing a robust and interpretable alternative to conventional systems. Finally, we validate the relative interpretability of these language-based user profiles through user studies involving crowdworkers and GPT-4-based evaluations. Implementation of LangPTune can be found at https://github.com/ZhaolinGao/LangPTune.

Retrieval-augmented generation (RAG) enhances the question-answering (QA) abilities of large language models (LLMs) by integrating external knowledge. However, adapting general-purpose RAG systems to specialized fields such as science and medicine poses unique challenges due to distribution shifts and limited access to domain-specific data. To tackle this, we propose SimRAG, a self-training approach that equips the LLM with joint capabilities of question answering and question generation for domain adaptation. Our method first fine-tunes the LLM on instruction-following, question-answering, and search-related data. Then, it prompts the same LLM to generate diverse domain-relevant questions from unlabeled corpora, with an additional filtering strategy to retain high-quality synthetic examples. By leveraging these synthetic examples, the LLM can improve their performance on domain-specific RAG tasks. Experiments on 11 datasets, spanning two backbone sizes and three domains, demonstrate that SimRAG outperforms baselines by 1.2\%--8.6\%., Comment: Work in Progress

Spatial expressions in situated communication can be ambiguous, as their meanings vary depending on the frames of reference (FoR) adopted by speakers and listeners. While spatial language understanding and reasoning by vision-language models (VLMs) have gained increasing attention, potential ambiguities in these models are still under-explored. To address this issue, we present the COnsistent Multilingual Frame Of Reference Test (COMFORT), an evaluation protocol to systematically assess the spatial reasoning capabilities of VLMs. We evaluate nine state-of-the-art VLMs using COMFORT. Despite showing some alignment with English conventions in resolving ambiguities, our experiments reveal significant shortcomings of VLMs: notably, the models (1) exhibit poor robustness and consistency, (2) lack the flexibility to accommodate multiple FoRs, and (3) fail to adhere to language-specific or culture-specific conventions in cross-lingual tests, as English tends to dominate other languages. With a growing effort to align vision-language models with human cognitive intuitions, we call for more attention to the ambiguous nature and cross-cultural diversity of spatial reasoning., Comment: Accepted to Pluralistic Alignment @ NeurIPS 2024 | Project page: https://spatial-comfort.github.io/

In order to better understand manifold neural networks (MNNs), we introduce Manifold Filter-Combine Networks (MFCNs). The filter-combine framework parallels the popular aggregate-combine paradigm for graph neural networks (GNNs) and naturally suggests many interesting families of MNNs which can be interpreted as the manifold analog of various popular GNNs. We then propose a method for implementing MFCNs on high-dimensional point clouds that relies on approximating the manifold by a sparse graph. We prove that our method is consistent in the sense that it converges to a continuum limit as the number of data points tends to infinity., Comment: Accepted to NeurIPS Workshop on Symmetry and Geometry in Neural Representations (Extended Abstract Track)

The total variation filtering technique emerges as a highly effective strategy for restoring signals with discontinuities in various parts of their structure. This study presents and implements a one-dimensional signal filtering algorithm based on total variation. The aim is to demonstrate the effectiveness of this algorithm through a series of synthetic filtering tests. The results presented in this paper were significant in demonstrating the proposed algorithm's effectiveness. Through a series of rigorously conducted experiments, the algorithm's ability to solve complex noise removal problems in various scenarios was evidenced., Comment: 9 pages,5 figures

Large vision and language assistants have enabled new capabilities for interpreting natural images. These approaches have recently been adapted to earth observation data, but they are only able to handle single image inputs, limiting their use for many real-world tasks. In this work, we develop a new vision and language assistant called TEOChat that can engage in conversations about temporal sequences of earth observation data. To train TEOChat, we curate an instruction-following dataset composed of many single image and temporal tasks including building change and damage assessment, semantic change detection, and temporal scene classification. We show that TEOChat can perform a wide variety of spatial and temporal reasoning tasks, substantially outperforming previous vision and language assistants, and even achieving comparable or better performance than specialist models trained to perform these specific tasks. Furthermore, TEOChat achieves impressive zero-shot performance on a change detection and change question answering dataset, outperforms GPT-4o and Gemini 1.5 Pro on multiple temporal tasks, and exhibits stronger single image capabilities than a comparable single EO image instruction-following model. We publicly release our data, models, and code at https://github.com/ermongroup/TEOChat .

Autonomous systems for software engineering are now capable of fixing bugs and developing features. These systems are commonly evaluated on SWE-bench (Jimenez et al., 2024a), which assesses their ability to solve software issues from GitHub repositories. However, SWE-bench uses only Python repositories, with problem statements presented predominantly as text and lacking visual elements such as images. This limited coverage motivates our inquiry into how existing systems might perform on unrepresented software engineering domains (e.g., front-end, game development, DevOps), which use different programming languages and paradigms. Therefore, we propose SWE-bench Multimodal (SWE-bench M), to evaluate systems on their ability to fix bugs in visual, user-facing JavaScript software. SWE-bench M features 617 task instances collected from 17 JavaScript libraries used for web interface design, diagramming, data visualization, syntax highlighting, and interactive mapping. Each SWE-bench M task instance contains at least one image in its problem statement or unit tests. Our analysis finds that top-performing SWE-bench systems struggle with SWE-bench M, revealing limitations in visual problem-solving and cross-language generalization. Lastly, we show that SWE-agent's flexible language-agnostic features enable it to substantially outperform alternatives on SWE-bench M, resolving 12% of task instances compared to 6% for the next best system.

We present a novel approach to reconstruct gas and dark matter projected density maps of galaxy clusters using score-based generative modeling. Our diffusion model takes in mock SZ and X-ray images as conditional observations, and generates realizations of corresponding gas and dark matter maps by sampling from a learned data posterior. We train and validate the performance of our model by using mock data from a hydrodynamical cosmological simulation. The model accurately reconstructs both the mean and spread of the radial density profiles in the spatial domain to within 5\%, indicating that the model is able to distinguish between clusters of different sizes. In the spectral domain, the model achieves close-to-unity values for the bias and cross-correlation coefficients, indicating that the model can accurately probe cluster structures on both large and small scales. Our experiments demonstrate the ability of score models to learn a strong, nonlinear, and unbiased mapping between input observables and fundamental density distributions of galaxy clusters. These diffusion models can be further fine-tuned and generalized to not only take in additional observables as inputs, but also real observations and predict unknown density distributions of galaxy clusters., Comment: 15 pages, 9 figures, submitted to The Open Journal of Astrophysics

Empowering people who are blind or visually impaired (BVI) to enhance their orientation and mobility skills is critical to equalizing their access to social and economic opportunities. To manage this crucial challenge, we employed a novel design process based on a large-scale, longitudinal, community-based structure. Across three annual programs we engaged with the BVI community in online and in-person modes. In total, our team included 67 total BVI participatory design participants online, 11 BVI co-designers in-person, and 4 BVI program coordinators. Through this design process we built a mobile application that enables users to generate, share, and navigate maps of indoor and outdoor environments without the need to instrument each environment with beacons or fiducial markers. We evaluated this app at a healthcare facility, and participants in the evaluation rated the app highly with respect to its design, features, and potential for positive impact on quality of life.

This paper aims to answer one central question: to what extent can open-source generative text models be used in a workflow to approximate thematic analysis in social science research? To answer this question, we present the Generative AI-enabled Theme Organization and Structuring (GATOS) workflow, which uses open-source machine learning techniques, natural language processing tools, and generative text models to facilitate thematic analysis. To establish validity of the method, we present three case studies applying the GATOS workflow, leveraging these models and techniques to inductively create codebooks similar to traditional procedures using thematic analysis. Specifically, we investigate the extent to which a workflow comprising open-source models and tools can inductively produce codebooks that approach the known space of themes and sub-themes. To address the challenge of gleaning insights from these texts, we combine open-source generative text models, retrieval-augmented generation, and prompt engineering to identify codes and themes in large volumes of text, i.e., generate a qualitative codebook. The process mimics an inductive coding process that researchers might use in traditional thematic analysis by reading text one unit of analysis at a time, considering existing codes already in the codebook, and then deciding whether or not to generate a new code based on whether the extant codebook provides adequate thematic coverage. We demonstrate this workflow using three synthetic datasets from hypothetical organizational research settings: a study of teammate feedback in teamwork settings, a study of organizational cultures of ethical behavior, and a study of employee perspectives about returning to their offices after the pandemic. We show that the GATOS workflow is able to identify themes in the text that were used to generate the original synthetic datasets.

Detections of fast X-ray transients (FXTs) have been accrued over the last few decades. However, their origin has remained mysterious. There is now rapid progress thanks to timely discoveries and localisations with the Einstein Probe mission. Early results indicate that FXTs may frequently, but not always, be associated with gamma-ray bursts (GRBs). Here, we report on the multi-wavelength counterpart of FXT EP240414a, which has no reported gamma-ray counterpart. The transient is located 25.7~kpc in projection from a massive galaxy at $z=0.40$. We perform comprehensive photometric and spectroscopic follow-up. The optical light curve shows at least three distinct emission episodes with timescales of $\sim 1, 4$ and 15 days and peak absolute magnitudes of $M_R \sim -20$, $-21$, and $-19.5$, respectively. The optical spectrum at early times is extremely blue, inconsistent with afterglow emission. It may arise from the interaction of both jet and supernova shock waves with the stellar envelope and a dense circumstellar medium, as has been suggested for some Fast Blue Optical Transients (LFBOTs). At late times, the spectrum evolves to a broad-lined~Type~Ic supernova, similar to those seen in collapsar long-GRBs. This implies that the progenitor of EP240414a is a massive star creating a jet-forming supernova inside a dense envelope, resulting in an X-ray outburst with a luminosity of $\sim 10^{48}$ erg s$^{-1}$, and the complex observed optical/IR light curves. If correct, this argues for a causal link between the progenitors of long-GRBs, FXTs and LFBOTs., Comment: 36 pages, 13 figures, submitted to ApJ

Developing robust and correctable visuomotor policies for robotic manipulation is challenging due to the lack of self-recovery mechanisms from failures and the limitations of simple language instructions in guiding robot actions. To address these issues, we propose a scalable data generation pipeline that automatically augments expert demonstrations with failure recovery trajectories and fine-grained language annotations for training. We then introduce Rich languAge-guided failure reCovERy (RACER), a supervisor-actor framework, which combines failure recovery data with rich language descriptions to enhance robot control. RACER features a vision-language model (VLM) that acts as an online supervisor, providing detailed language guidance for error correction and task execution, and a language-conditioned visuomotor policy as an actor to predict the next actions. Our experimental results show that RACER outperforms the state-of-the-art Robotic View Transformer (RVT) on RLbench across various evaluation settings, including standard long-horizon tasks, dynamic goal-change tasks and zero-shot unseen tasks, achieving superior performance in both simulated and real world environments. Videos and code are available at: https://rich-language-failure-recovery.github.io., Comment: Project Website: https://rich-language-failure-recovery.github.io

Human-robot physical interaction (pHRI) is a rapidly evolving research field with significant implications for physical therapy, search and rescue, and telemedicine. However, a major challenge lies in accurately understanding human constraints and safety in human-robot physical experiments without an IRB and physical human experiments. Concerns regarding human studies include safety concerns, repeatability, and scalability of the number and diversity of participants. This paper examines whether a physical approximation can serve as a stand-in for human subjects to enhance robot autonomy for physical assistance. This paper introduces the SHULDRD (Shoulder Haptic Universal Limb Dynamic Repositioning Device), an economical and anatomically similar device designed for real-time testing and deployment of pHRI planning tasks onto robots in the real world. SHULDRD replicates human shoulder motion, providing crucial force feedback and safety data. The device's open-source CAD and software facilitate easy construction and use, ensuring broad accessibility for researchers. By providing a flexible platform able to emulate infinite human subjects, ensure repeatable trials, and provide quantitative metrics to assess the effectiveness of the robotic intervention, SHULDRD aims to improve the safety and efficacy of human-robot physical interactions., Comment: Submitted to ICRA '25

Heavy precipitation from tropical cyclones (TCs) may result in disasters, such as floods and landslides, leading to substantial economic damage and loss of life. Prediction of TC precipitation based on ensemble post-processing procedures using machine learning (ML) approaches has received considerable attention for its flexibility in modeling and its computational power in managing complex models. However, when applying ML techniques to TC precipitation for a specific area, the available observation data are typically insufficient for comprehensive training, validation, and testing of the ML model, primarily due to the rapid movement of TCs. We propose to use the convolutional neural network (CNN) as a deep ML model to leverage the spatial information of precipitation. The proposed model has three distinct features that differentiate it from traditional CNNs applied in meteorology. First, it utilizes data augmentation to alleviate challenges posed by the small sample size. Second, it contains geographical and dynamic variables to account for area-specific features and the relative distance between the study area and the moving TC. Third, it applies unequal weights to accommodate the temporal structure in the training data when calculating the objective function. The proposed CNN-all model is then illustrated with the TC Soudelor's impact on Taiwan. Soudelor was the strongest TC of the 2015 Pacific typhoon season. The results show that the inclusion of augmented data and dynamic variables improves the prediction of heavy precipitation. The proposed CNN-all outperforms traditional CNN models, based on the continuous probability skill score (CRPSS), probability plots, and reliability diagram. The proposed model has the potential to be utilized in a wide range of meteorological studies., Comment: 21 pages, 6 figures, 2 tables, 2 supplementary figures

This study investigates the 5$f$-electron-conduction electron hybridization process in the heavy fermion superconductor UPd$_2$Al$_3$ using a combination of angle-resolved photoemission spectroscopy (ARPES) and time-resolved photoemission spectroscopy (tr-PES). ARPES measurements reveal the formation of a hybridization gap at a temperature of approximately 75 K, which becomes more pronounced as the temperature decreases. Notably, the persistence of a flat U 5$f$ band at temperatures well above the hybridization onset challenges conventional understanding. Our findings demonstrate a non-monotonic temperature dependence of the quasiparticle relaxation time, with an anomalous decrease at 20 K, suggesting complex electronic and magnetic interactions. These findings provide detailed insights into the 5$f$-electron hybridization process in UPd$_2$Al$_3$, with significant implications for the understanding of heavy fermion superconductivity and the role of 5$f$-electron hybridization in uranium-based materials., Comment: 5 pages, 4 figures

Globular clusters offer a powerful way to test the properties of stellar populations and the late stages of low-mass stellar evolution. In this paper we study oscillating giant stars and overtone RR Lyrae-type pulsators in the nearest globular cluster, M4, with the help of high-precision, continuous light curves collected by the Kepler space telescope in the K2 mission. We determine the frequency composition of five RRc stars and model their physical parameters with a grid of linear pulsation models. We are able, for the first time, to compare seismic masses of RR Lyrae stars directly to the masses of the very similar red horizontal branch stars in the same stellar population, independently determined from asteroseismic scaling relations. We find a close match, with an average seismic mass of $0.651\pm0.028\,M_\odot$ for RR Lyrae stars and $0.657\pm0.034\,M_\odot$ for red horizontal-branch stars. While the validity of our RR Lyrae masses still relies on the similarity of neighboring horizontal branch subgroups, this result strongly indicates that RRc stars may indeed exhibit high-degree, $l = 8$ and 9 non-radial modes, and modeling these modes can provide realistic mass estimates. We also determine the He content of the cluster to be $Y = 0.266\pm 0.008$, and compare the seismic masses for our sample of RR Lyrae to theoretical mass relations and highlight the limitations of these relations., Comment: 11 pages, 10 figures, submitted to A&A for review

Transit is essential for urban transportation and achieving net-zero targets. In urban areas like the Chicago Metropolitan Region, transit enhances mobility and connects people, fostering a dynamic economy. To quantify the mobility and selected economic impacts of transit, we use a novel agent-based simulation model POLARIS to compare baseline service against a scenario in which transit is completely removed. The transit-removal scenario assumes higher car ownership and results in higher traffic congestion, numerous activity cancellations, and economic decline. In this scenario, average travel times increase by 14.2% regionally and 34.7% within the City of Chicago. The resulting congestion causes significant activity cancellations despite increased car ownership: 11.8% of non-work and 2.8% of work/school activities regionally, totaling an 8.6% overall cancellation rate. In the city, non-work cancellations would reach 26.9%, and work/school cancellations 7.3%, leading to a 19.9% overall cancellation rate. The impact varies between groups. Women and lower-income individuals are more likely to cancel activities than men and higher-income groups. Women account for 53.7% of non-work and 53.0% of total cancellations. The lowest 40% income group experiences 50.2% of non-work and 48.0% of overall cancellations. Combined, activity cancellations, travel time losses, and increased car ownership cost the region $35.4 billion. With annual public transit funding at $2.7 billion, the ratio is 13 to 1, underscoring transit's critical role in mobility, equity, and economic health.

There is growing interest in artificial intelligence (AI) tools, especially high-profile tools like ChatGPT, and these tools now appear to be part of the education experience for many high school students. To investigate students' use of AI tools for school assignments, their impressions of how using the tools might affect them cognitively and academically, and their thoughts on using AI tools to write their college admissions essays, ACT developed a survey in June 2023 which was administered to a large nationwide sample of students in Grades 10 through 12. In this study, almost half of the participating students reported that they had used AI tools, and the most common tool they used was ChatGPT. Among students who did not use AI tools, the top reason for not using them was having no interest in them. About two thirds of students also reported that they did not trust the information provided by AI tools, and a little over half indicated that they did not know enough about AI tools to use them. Students with higher academic performance were significantly more likely to use AI tools than were students with lower academic performance. Findings show that these tools appear to have much potential to enhance student learning, but there are concerns about appropriate use and potential negative outcomes.

This review paper describes the energy-upgraded CEBAF accelerator. This superconducting linac has achieved 12 GeV beam energy by adding 11 new high-performance cryomodules containing eighty-eight superconducting cavities that have operated CW at an average accelerating gradient of 20 MV/m. After reviewing the attributes and performance of the previous 6 GeV CEBAF accelerator, we discuss the upgraded CEBAF accelerator system in detail with particular attention paid to the new beam acceleration systems. In addition to doubling the acceleration in each linac, the upgrade included improving the beam recirculation magnets, adding more helium cooling capacity to allow the newly installed modules to run cold, adding a new experimental hall, and improving numerous other accelerator components. We review several of the techniques deployed to operate and analyze the accelerator performance, and document system operating experience and performance. In the final portion of the document, we present much of the current planning regarding projects to improve accelerator performance and enhance operating margins, and our plans for ensuring CEBAF operates reliably into the future. For the benefit of potential users of CEBAF, the performance and quality measures for beam delivered to each of the experimental halls is summarized in the appendix., Comment: 66 pages, 73 figures, 21 tables

Recent technological advancements have enhanced our ability to collect and analyze rich multimodal data (e.g., speech, video, and eye gaze) to better inform learning and training experiences. While previous reviews have focused on parts of the multimodal pipeline (e.g., conceptual models and data fusion), a comprehensive literature review on the methods informing multimodal learning and training environments has not been conducted. This literature review provides an in-depth analysis of research methods in these environments, proposing a taxonomy and framework that encapsulates recent methodological advances in this field and characterizes the multimodal domain in terms of five modality groups: Natural Language, Video, Sensors, Human-Centered, and Environment Logs. We introduce a novel data fusion category -- mid fusion -- and a graph-based technique for refining literature reviews, termed citation graph pruning. Our analysis reveals that leveraging multiple modalities offers a more holistic understanding of the behaviors and outcomes of learners and trainees. Even when multimodality does not enhance predictive accuracy, it often uncovers patterns that contextualize and elucidate unimodal data, revealing subtleties that a single modality may miss. However, there remains a need for further research to bridge the divide between multimodal learning and training studies and foundational AI research., Comment: Submitted to ACM Computing Surveys. Currently under review

The increasing popularity of Artificial Intelligence in recent years has led to a surge in interest in image classification, especially in the agricultural sector. With the help of Computer Vision, Machine Learning, and Deep Learning, the sector has undergone a significant transformation, leading to the development of new techniques for crop classification in the field. Despite the extensive research on various image classification techniques, most have limitations such as low accuracy, limited use of data, and a lack of reporting model size and prediction. The most significant limitation of all is the need for model explainability. This research evaluates four different approaches for crop classification, namely traditional ML with handcrafted feature extraction methods like SIFT, ORB, and Color Histogram; Custom Designed CNN and established DL architecture like AlexNet; transfer learning on five models pre-trained using ImageNet such as EfficientNetV2, ResNet152V2, Xception, Inception-ResNetV2, MobileNetV3; and cutting-edge foundation models like YOLOv8 and DINOv2, a self-supervised Vision Transformer Model. All models performed well, but Xception outperformed all of them in terms of generalization, achieving 98% accuracy on the test data, with a model size of 80.03 MB and a prediction time of 0.0633 seconds. A key aspect of this research was the application of Explainable AI to provide the explainability of all the models. This journal presents the explainability of Xception model with LIME, SHAP, and GradCAM, ensuring transparency and trustworthiness in the models' predictions. This study highlights the importance of selecting the right model according to task-specific needs. It also underscores the important role of explainability in deploying AI in agriculture, providing insightful information to help enhance AI-driven crop management strategies.

We predict the existence of $\alpha$ Ori B, a low-mass companion orbiting Betelgeuse. This is motivated by the presence of a 2170-day Long Secondary Period (LSP) in Betelgeuse's lightcurve, a periodicity $\approx5$ times longer than the star's 416 day fundamental radial pulsation mode. While binarity is currently the leading hypothesis for LSPs in general, the LSP and the radial velocity variation observed in Betelgeuse, taken together, necessitate a revision of the prevailing physical picture. The lightcurve-RV phase difference requires a companion to be behind Betelgeuse at the LSP luminosity minimum, 180 degrees out of phase with the system orientation associated with occultation. We demonstrate the consistency of this model with available observational constraints and identify tensions in all other proposed LSP hypotheses. Within this framework, we calculate a mass for $\alpha$ Ori B of $M\sin i =1.17\pm0.7\,M_\odot$ and an orbital separation of $1850\pm70\,R_\odot$, or $2.43^{+0.21}_{-0.32}$ times the radius of Betelgeuse. We then describe the features of the companion as constrained by the fundamental parameters of Betelgeuse and its orbital system, and discuss what would be required to confirm the companion's existence observationally., Comment: 18 pages, 4 figures, 3 tables, accepted to the Astrophysical Journal

We describe a method for depositing nanoscale liquid Au droplets, initially levitated in an ion trap in high vacuum, onto a remote substrate. A levitated Au nanosphere is melted, expelled from the trap, and maintained in the molten state with a laser directed along the droplet trajectory until it reaches the substrate and rapidly solidifies. Also during transit, the charged droplets are focused to a small region of the substrate with an electrostatic lens. After deposition, the substrate can be removed from the vacuum chamber and imaged and analyzed by techniques such as electron microscopy and energy dispersive spectroscopy (EDS). Over 90% of launched particles are deposited on the substrate, and when the lens is focused, particles land in a region of diameter 120 $\mu$m after traversing a distance of 236 mm. Our technique is of value for analysis of materials prepared or processed while levitated that can be melted. Also, Au droplets may be useful as tracers for future experiments involving smaller projectiles or oriented solids.

Convolutional neural networks (CNNs) are a popular choice of model for tasks in computer vision. When CNNs are made with many layers, resulting in a deep neural network, skip connections may be added to create an easier gradient optimization problem while retaining model expressiveness. In this paper, we show that arbitrarily complex, trained, linear CNNs with skip connections can be simplified into a single-layer model, resulting in greatly reduced computational requirements during prediction time. We also present a method for training nonlinear models with skip connections that are gradually removed throughout training, giving the benefits of skip connections without requiring computational overhead during during prediction time. These results are demonstrated with practical examples on Residual Networks (ResNet) architecture., Comment: Accepted for publication in the proceedings of the 27th International Workshop on Computer Algebra in Scientific Computing (CASC 2024)

Circularly polarized femtosecond UV laser pulse can remove a {\pi} electron from benzene in such a way that the leftover hole circulates around the cation as a persistent ring current. We investigate the time dependent strength of the current as the molecule relaxes from the D6h symmetry of the neutral to the D2h symmetry of the cation due to the Jahn-Teller effect. We explore the effect of spontaneous symmetry breaking on persistent ring currents for benzene cation, because it is one of the most comprehensively studied examples of the Jahn-Teller effect.

Very metal-poor stars ([Fe/H]<-2) are important laboratories for testing stellar models and reconstructing the formation history of our galaxy. Asteroseismology is a powerful tool to probe stellar interiors and measure ages, but few asteroseismic detections are known in very metal-poor stars and none have allowed detailed modeling of oscillation frequencies. We report the discovery of a low-luminosity Kepler red giant (KIC 8144907) with high S/N oscillations, [Fe/H]=-2.66+/-0.08 and [alpha/Fe]=0.38+/-0.06, making it by far the most metal-poor star to date for which detailed asteroseismic modeling is possible. By combining the oscillation spectrum from Kepler with high-resolution spectroscopy we measure an asteroseismic mass and age of 0.79+/-0.02(ran)+/-0.01(sys) Msun and 12.0+/-0.6(ran)+/-0.4(sys) Gyr, with remarkable agreement across different codes and input physics, demonstrating that stellar models and asteroseismology are reliable for very metal-poor stars when individual frequencies are used. The results also provide a direct age anchor for the early formation of the Milky Way, implying that substantial star formation did not commence until redshift z~3 (if the star formed in-situ) or that the Milky Way has undergone merger events for at least ~12 Gyr (if the star was accreted by a dwarf satellite merger such as Gaia Enceladus)., Comment: 10 pages, 4 figures, accepted for publication in ApJ

The rapid advancement of quantum computing poses a significant threat to many current security algorithms used for secure communication, digital authentication, and information encryption. A sufficiently powerful quantum computer could potentially exploit vulnerabilities in these algorithms, rendering data in transit insecure. This threat is expected to materialize within the next 20 years. Immediate transition to quantum-resilient cryptographic schemes is crucial, primarily to mitigate store-now-decrypt-later attacks and to ensure the security of products with decade-long operational lives. This transition requires a systematic approach to identifying and upgrading vulnerable cryptographic implementations. This work developed a quantum assessment tool for organizations, providing tailored recommendations for transitioning their security protocols into a post-quantum world. The work included a systematic evaluation of the proposed solution using qualitative feedback from network administrators and cybersecurity experts. This feedback was used to refine the accuracy and usability of the assessment process. The results demonstrate its effectiveness and usefulness in helping organizations prepare for quantum computing threats. The assessment tool is publicly available at (https://quantum-watch.soton.ac.uk).

Asteroseismic modelling is a powerful way to derive stellar properties. However, the derived quantities are limited by built-in assumptions used in stellar models. This work presents a detailed characterisation of stellar model uncertainties in asteroseismic red giants, focusing on the mixing-length parameter $\alpha_{\rm MLT}$, the initial helium fraction $Y_{\rm init}$, the solar abundance scale, and the overshoot parameters. First, we estimate error floors due to model uncertainties to be $\approx$0.4\% in mass, $\approx$0.2\% in radius, and $\approx$17\% in age, primarily due to the uncertain state of $\alpha_{\rm MLT}$ and $Y_{\rm init}$. The systematic uncertainties in age exceed typical statistical uncertainties, suggesting the importance of their evaluation in asteroseismic applications. Second, we demonstrate that the uncertainties from $\alpha_{\rm MLT}$ can be entirely mitigated by direct radius measurements or partially through $\nu_{\rm max}$. Utilizing radii from Kepler eclipsing binaries, we determined the $\alpha_{\rm MLT}$ values and calibrated the $\alpha_{\rm MLT}$--[M/H] relation. The correlation observed between the two variables is positive, consistent with previous studies using 1-D stellar models, but in contrast with outcomes from 3-D simulations. Third, we explore the implications of using asteroseismic modelling to test the $\nu_{\rm max}$ scaling relation. We found that a perceived dependency of $\nu_{\rm max}$ on [M/H] from individual frequency modelling can be largely removed by incorporating the calibrated $\alpha_{\rm MLT}$--[M/H] relation. Variations in $Y_{\rm init}$ can also affect $\nu_{\rm max}$ predictions. These findings suggest that $\nu_{\rm max}$ conveys information not fully captured by individual frequencies, and that it should be carefully considered as an important observable for asteroseismic modelling., Comment: 18 pages, 7 figures; submitted to ApJ

Social determinants of health (SDOH) $-$ the myriad of circumstances in which people live, grow, and age $-$ play an important role in health outcomes. However, existing outcome prediction models often only use proxies of SDOH as features. Recent open data initiatives present an opportunity to construct a more comprehensive view of SDOH, but manually integrating the most relevant data for individual patients becomes increasingly challenging as the volume and diversity of public SDOH data grows. Large language models (LLMs) have shown promise at automatically annotating structured data. Here, we conduct an end-to-end case study evaluating the feasibility of using LLMs to integrate SDOH data, and the utility of these SDOH features for clinical prediction. We first manually label 700+ variables from two publicly-accessible SDOH data sources to one of five semantic SDOH categories. Then, we benchmark performance of 9 open-source LLMs on this classification task. Finally, we train ML models to predict 30-day hospital readmission among 39k heart failure (HF) patients, and we compare the prediction performance of the categorized SDOH variables with standard clinical variables. Additionally, we investigate the impact of few-shot LLM prompting on LLM annotation performance, and perform a metadata ablation study on prompts to evaluate which information helps LLMs accurately annotate these variables. We find that some open-source LLMs can effectively, accurately annotate SDOH variables with zero-shot prompting without the need for fine-tuning. Crucially, when combined with standard clinical features, the LLM-annotated Neighborhood and Built Environment subset of the SDOH variables shows the best performance predicting 30-day readmission of HF patients., Comment: 36 pages including references and appendix. This is a work in progress

We discuss the interrelationship between the treatment of the elderly, the nature of production, and the transmission of culture. Respect for the elderly is endogenous. Parents cultivate an interest in consuming culture in their children; when they are older, children compensate their elders proportional to the degree to which their interests were previously cultivated. We show that this model is functionally equivalent to one in which cultural goods are transferred across generations. We focus on the relative well-being of the elderly and use the model to explain patterns in their relative well-being across societies. An important theme is that the cultivation of culture and norms for the respect and support of the elderly bear a nonlinear relationship with many economic variables, such as capital and or land intensity in production. We also discuss the interaction of property rights with production, assets such as productive resources, and relative treatment of the elderly. Insecurity of some types of property rights, such as rights over output, may benefit the elderly, while secure rights over productive resources may also benefit the elderly. We discuss how the elderly could be affected by demographic, technological and policy changes in both developing and developed economies.

Vision-and-Language Navigation (VLN) has gained increasing attention over recent years and many approaches have emerged to advance their development. The remarkable achievements of foundation models have shaped the challenges and proposed methods for VLN research. In this survey, we provide a top-down review that adopts a principled framework for embodied planning and reasoning, and emphasizes the current methods and future opportunities leveraging foundation models to address VLN challenges. We hope our in-depth discussions could provide valuable resources and insights: on one hand, to milestone the progress and explore opportunities and potential roles for foundation models in this field, and on the other, to organize different challenges and solutions in VLN to foundation model researchers., Comment: Authors contributed equally to this work, and supervisors contributed equal advising to this work