Search

Present-day software development faces three major challenges: complexity, time consumption, and high costs. Developing large software systems often requires battalions of teams and considerable time for meetings, which end without any action, resulting in unproductive cycles, delayed progress, and increased cost. What if, instead of large meetings with no immediate results, the software product is completed by the end of the meeting? In response, we present a vision for a system called TimeLess, designed to reshape the software development process by enabling immediate action during meetings. The goal is to shift meetings from planning discussions to productive, action-oriented sessions. This approach minimizes the time and effort required for development, allowing teams to focus on critical decision-making while AI agents execute development tasks based on the meeting discussions. We will employ multiple AI agents that work collaboratively to capture human discussions and execute development tasks in real time. This represents a step toward next-generation software development environments, where human expertise drives strategy and AI accelerates task execution., Comment: 5 pages, 4 figure, and 1 table

A sensitivity increase of two orders of magnitude in proton (1H) and carbon (13C) spins via dynamic nuclear polarization (DNP) has been accomplished recently using a compact benchtop DNP polarizer operating at 1 T and 77 K. However the DNP mechanisms at play at such low magnetic field and high operating temperature are still not elucidated. A deeper understanding of the dominant polarization transfer mechanisms between electrons and 1H and 13C spins at these unconventional benchtop conditions is therefore required if one wants to devise strategies to boost sensitivity further. In this study, we found that DNP is generally dominated by solid effect for narrow electron paramagnetic resonance (EPR) line radicals (15 mM trityl OX063) and cross effect for broad EPR line radicals (50 mM TEMPOL). For both radicals, the dominant DNP mechanisms were investigated varying the microwave frequency and measuring the 1H and 13C DNP enhancement factors to obtain 1H and 13C DNP spectra. The impact of varying the microwave power on the 1H DNP buildup times and the 1H nuclear spin relaxation times were important as well to distinguish between solid effect and cross effect DNP. Finally, time-resolved electron saturation simulations under continuous microwave irradiation could replicate the experimental 1H and 13C DNP spectra at 1 T and 77 K for both radicals considering their electron relaxation properties. Only for trityl OX063, the 13C DNP spectra showed additional DNP maxima compared to the simulations. This has been attributed to methyl rotor induced 1H-13C heteronuclear cross relaxation in [1-13C] acetate present at 1 T and 77 K., Comment: Main text, 15 pages, 6 figures. Supporting information, 7 pages, 5 figures

In this report, we introduce INTELLECT-1, the first 10 billion parameter language model collaboratively trained across the globe, demonstrating that large-scale model training is no longer confined to large corporations but can be achieved through a distributed, community-driven approach. INTELLECT-1 was trained on 1 trillion tokens using up to 14 concurrent nodes distributed across 3 continents, with contributions from 30 independent compute providers dynamically joining and leaving the training process, while maintaining 83-96% compute utilization and 36.2-41.4% model FLOPS utilization. We leverage PRIME, our scalable distributed training framework designed for fault-tolerant, high-performance training on unreliable, globally distributed nodes. Key innovations in PRIME include the ElasticDeviceMesh, which manages dynamic global process groups for fault-tolerant communication across the internet and local process groups for communication within a node, live checkpoint recovery kernels, and a hybrid DiLoCo-FSDP2 implementation. Using PRIME with DiLoCo and our custom int8 all-reduce, we achieve a 400x reduction in communication bandwidth compared to traditional data-parallel training settings while delivering comparable performance. These results demonstrate the feasibility and promise of training frontier foundation models in a decentralized network of global GPU resources., Comment: 19 pages, 6 figures

In this work we develop theoretical techniques for analysing the performance of the quantum approximate optimization algorithm (QAOA) when applied to random boolean constraint satisfaction problems (CSPs), and use these techniques to compare the complexity of a variety of CSPs, such as $k$-SAT, 1-in-$k$ SAT, and NAE-SAT. Our techniques allow us to compute the success probability of QAOA with one layer and given parameters, when applied to randomly generated instances of CSPs with $k$ binary variables per constraint, in time polynomial in $n$ and $k$. We apply this algorithm to all boolean CSPs with $k=3$ and a large number of CSPs with $k=4$, $k=5$, and compare the resulting complexity with the complexity of solving the corresponding CSP using the standard solver MapleSAT, determined experimentally. We find that random $k$-SAT seems to be the most promising of these CSPs for demonstrating a quantum-classical separation using QAOA., Comment: 33+7 pages, 13 figures

Frontier AI safety policies highlight automation of AI research and development (R&D) by AI agents as an important capability to anticipate. However, there exist few evaluations for AI R&D capabilities, and none that are highly realistic and have a direct comparison to human performance. We introduce RE-Bench (Research Engineering Benchmark, v1), which consists of 7 challenging, open-ended ML research engineering environments and data from 71 8-hour attempts by 61 distinct human experts. We confirm that our experts make progress in the environments given 8 hours, with 82% of expert attempts achieving a non-zero score and 24% matching or exceeding our strong reference solutions. We compare humans to several public frontier models through best-of-k with varying time budgets and agent designs, and find that the best AI agents achieve a score 4x higher than human experts when both are given a total time budget of 2 hours per environment. However, humans currently display better returns to increasing time budgets, narrowly exceeding the top AI agent scores given an 8-hour budget, and achieving 2x the score of the top AI agent when both are given 32 total hours (across different attempts). Qualitatively, we find that modern AI agents possess significant expertise in many ML topics -- e.g. an agent wrote a faster custom Triton kernel than any of our human experts' -- and can generate and test solutions over ten times faster than humans, at much lower cost. We open-source the evaluation environments, human expert data, analysis code and agent trajectories to facilitate future research.

An important aspect of deploying face recognition (FR) algorithms in real-world applications is their ability to learn new face identities from a continuous data stream. However, the online training of existing deep neural network-based FR algorithms, which are pre-trained offline on large-scale stationary datasets, encounter two major challenges: (I) catastrophic forgetting of previously learned identities, and (II) the need to store past data for complete retraining from scratch, leading to significant storage constraints and privacy concerns. In this paper, we introduce CLFace, a continual learning framework designed to preserve and incrementally extend the learned knowledge. CLFace eliminates the classification layer, resulting in a resource-efficient FR model that remains fixed throughout lifelong learning and provides label-free supervision to a student model, making it suitable for open-set face recognition during incremental steps. We introduce an objective function that employs feature-level distillation to reduce drift between feature maps of the student and teacher models across multiple stages. Additionally, it incorporates a geometry-preserving distillation scheme to maintain the orientation of the teacher model's feature embedding. Furthermore, a contrastive knowledge distillation is incorporated to continually enhance the discriminative power of the feature representation by matching similarities between new identities. Experiments on several benchmark FR datasets demonstrate that CLFace outperforms baseline approaches and state-of-the-art methods on unseen identities using both in-domain and out-of-domain datasets., Comment: Accepted for publication in the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2025)

Accurate calculations of starspot spectra are essential for multiple applications in astronomy. The current standard is to represent starspot spectra by spectra of stars that are cooler than the quiet star regions. This implies approximating a starspot as a non-magnetic 1D structure in radiative-convective equilibrium, parametrizing convective energy transport by mixing length theory. It is the inhibition of convection by the starspot magnetic field that is emulated by using a lower spot temperature relative to the quiet stellar regions. Here, we take a different approach avoiding the approximate treatment of convection and instead self-consistently accounting for the interaction between matter, radiation, and the magnetic field. We simulate spots on G2V, K0V, M0V stars with the 3D radiative magnetohydrodynamics code MURaM and calculate spectra ($R \approx 500$ from 250~nm to 6000~nm) using ray-by-ray radiative transfer with the MPS-ATLAS code. We find that the 1D models fail to return accurate umbral and penumbral spectra on K0V and M0V stars where convective and radiative transfer of energy is simultaneously important over a broad range of atmospheric heights rendering mixing length theory inaccurate. However, 1D models work well for G2V stars, where both radiation and convection significantly contribute to energy transfer only in a narrow region near the stellar surface. Quantitatively, the 1D approximation leads to errors longward of 500 nm of about 50\% for both umbral and penumbral flux contrast relative to quiet star regions on M0V stars, and less than 2\% (for umbrae) and 10\% (for penumbrae) for G2V stars., Comment: Accepted for publication in the Astrophysical Journal Letters

Large Language Models (LLMs) have become increasingly popular for generating RTL code. However, producing error-free RTL code in a zero-shot setting remains highly challenging for even state-of-the-art LLMs, often leading to issues that require manual, iterative refinement. This additional debugging process can dramatically increase the verification workload, underscoring the need for robust, automated correction mechanisms to ensure code correctness from the start. In this work, we introduce AIvril2, a self-verifying, LLM-agnostic agentic framework aimed at enhancing RTL code generation through iterative corrections of both syntax and functional errors. Our approach leverages a collaborative multi-agent system that incorporates feedback from error logs generated by EDA tools to automatically identify and resolve design flaws. Experimental results, conducted on the VerilogEval-Human benchmark suite, demonstrate that our framework significantly improves code quality, achieving nearly a 3.4$\times$ enhancement over prior methods. In the best-case scenario, functional pass rates of 77% for Verilog and 66% for VHDL were obtained, thus substantially improving the reliability of LLM-driven RTL code generation.

Although recent diffusion-based single-step super-resolution methods achieve better performance as compared to SinSR, they are computationally complex. To improve the performance of SinSR, we investigate preserving the high-frequency detail features during super-resolution (SR) because the downgraded images lack detailed information. For this purpose, we introduce a high-frequency perceptual loss by utilizing an invertible neural network (INN) pretrained on the ImageNet dataset. Different feature maps of pretrained INN produce different high-frequency aspects of an image. During the training phase, we impose to preserve the high-frequency features of super-resolved and ground truth (GT) images that improve the SR image quality during inference. Furthermore, we also utilize the Jenson-Shannon divergence between GT and SR images in the pretrained DINO-v2 embedding space to match their distribution. By introducing the $\textbf{h}igh$- $\textbf{f}requency$ preserving loss and distribution matching constraint in the single-step $\textbf{diff}usion-based$ SR ($\textbf{HF-Diff}$), we achieve a state-of-the-art CLIPIQA score in the benchmark RealSR, RealSet65, DIV2K-Val, and ImageNet datasets. Furthermore, the experimental results in several datasets demonstrate that our high-frequency perceptual loss yields better SR image quality than LPIPS and VGG-based perceptual losses. Our code will be released at https://github.com/shoaib-sami/HF-Diff., Comment: 8 pages

This study proposes a novel approach for real-time facial expression recognition utilizing short-range Frequency-Modulated Continuous-Wave (FMCW) radar equipped with one transmit (Tx), and three receive (Rx) antennas. The system leverages four distinct modalities simultaneously: Range-Doppler images (RDIs), micro range-Doppler Images (micro-RDIs), range azimuth images (RAIs), and range elevation images (REIs). Our innovative architecture integrates feature extractor blocks, intermediate feature extractor blocks, and a ResNet block to accurately classify facial expressions into smile, anger, neutral, and no-face classes. Our model achieves an average classification accuracy of 98.91% on the dataset collected using a 60 GHz short-range FMCW radar. The proposed solution operates in real-time in a person-independent manner, which shows the potential use of low-cost FMCW radars for effective facial expression recognition in various applications., Comment: Accepted at IEEE SENSORS 2024

Event cameras capture sparse, asynchronous brightness changes which offer high temporal resolution, high dynamic range, low power consumption, and sparse data output. These advantages make them ideal for Space Situational Awareness, particularly in detecting resident space objects moving within a telescope's field of view. However, the output from event cameras often includes substantial background activity noise, which is known to be more prevalent in low-light conditions. This noise can overwhelm the sparse events generated by satellite signals, making detection and tracking more challenging. Existing noise-filtering algorithms struggle in these scenarios because they are typically designed for denser scenes, where losing some signal is acceptable. This limitation hinders the application of event cameras in complex, real-world environments where signals are extremely sparse. In this paper, we propose new event-driven noise-filtering algorithms specifically designed for very sparse scenes. We categorise the algorithms into logical-based and learning-based approaches and benchmark their performance against 11 state-of-the-art noise-filtering algorithms, evaluating how effectively they remove noise and hot pixels while preserving the signal. Their performance was quantified by measuring signal retention and noise removal accuracy, with results reported using ROC curves across the parameter space. Additionally, we introduce a new high-resolution satellite dataset with ground truth from a real-world platform under various noise conditions, which we have made publicly available. Code, dataset, and trained weights are available at \url{https://github.com/samiarja/dvs_sparse_filter}., Comment: 17 pages, 8 figures, 1 table

As Large-Scale Cloud Systems (LCS) become increasingly complex, effective anomaly detection is critical for ensuring system reliability and performance. However, there is a shortage of large-scale, real-world datasets available for benchmarking anomaly detection methods. To address this gap, we introduce a new high-dimensional dataset from IBM Cloud, collected over 4.5 months from the IBM Cloud Console. This dataset comprises 39,365 rows and 117,448 columns of telemetry data. Additionally, we demonstrate the application of machine learning models for anomaly detection and discuss the key challenges faced in this process. This study and the accompanying dataset provide a resource for researchers and practitioners in cloud system monitoring. It facilitates more efficient testing of anomaly detection methods in real-world data, helping to advance the development of robust solutions to maintain the health and performance of large-scale cloud infrastructures.

The difficulty of exploring and training online on real production systems limits the scope of real-time online data/feedback-driven decision making. The most feasible approach is to adopt offline reinforcement learning from limited trajectory samples. However, after deployment, such policies fail due to exogenous factors that temporarily or permanently disturb/alter the transition distribution of the assumed decision process structure induced by offline samples. This results in critical policy failures and generalization errors in sensitive domains like Real-Time Communication (RTC). We solve this crucial problem of identifying robust actions in presence of domain shifts due to unseen exogenous stochastic factors in the wild. As it is impossible to learn generalized offline policies within the support of offline data that are robust to these unseen exogenous disturbances, we propose a novel post-deployment shaping of policies (Streetwise), conditioned on real-time characterization of out-of-distribution sub-spaces. This leads to robust actions in bandwidth estimation (BWE) of network bottlenecks in RTC and in standard benchmarks. Our extensive experimental results on BWE and other standard offline RL benchmark environments demonstrate a significant improvement ($\approx$ 18% on some scenarios) in final returns wrt. end-user metrics over state-of-the-art baselines.

Large and Small Language Models (LMs) are typically pretrained using extensive volumes of text, which are sourced from publicly accessible platforms such as Wikipedia, Book Corpus, or through web scraping. These models, due to their exposure to a wide range of language data, exhibit impressive generalization capabilities and can perform a multitude of tasks simultaneously. However, they often fall short when it comes to domain-specific tasks due to their broad training data. This paper introduces SecEncoder, a specialized small language model that is pretrained using security logs. SecEncoder is designed to address the domain-specific limitations of general LMs by focusing on the unique language and patterns found in security logs. Experimental results indicate that SecEncoder outperforms other LMs, such as BERTlarge, DeBERTa-v3-large and OpenAI's Embedding (textembedding-ada-002) models, which are pretrained mainly on natural language, across various tasks. Furthermore, although SecEncoder is primarily pretrained on log data, it outperforms models pretrained on natural language for a range of tasks beyond log analysis, such as incident prioritization and threat intelligence document retrieval. This suggests that domain specific pretraining with logs can significantly enhance the performance of LMs in security. These findings pave the way for future research into security-specific LMs and their potential applications.

Supporting the health and well-being of dynamic populations around the world requires governmental agencies, organizations and researchers to understand and reason over complex relationships between human behavior and local contexts in order to identify high-risk groups and strategically allocate limited resources. Traditional approaches to these classes of problems often entail developing manually curated, task-specific features and models to represent human behavior and the natural and built environment, which can be challenging to adapt to new, or even, related tasks. To address this, we introduce a Population Dynamics Foundation Model (PDFM) that aims to capture the relationships between diverse data modalities and is applicable to a broad range of geospatial tasks. We first construct a geo-indexed dataset for postal codes and counties across the United States, capturing rich aggregated information on human behavior from maps, busyness, and aggregated search trends, and environmental factors such as weather and air quality. We then model this data and the complex relationships between locations using a graph neural network, producing embeddings that can be adapted to a wide range of downstream tasks using relatively simple models. We evaluate the effectiveness of our approach by benchmarking it on 27 downstream tasks spanning three distinct domains: health indicators, socioeconomic factors, and environmental measurements. The approach achieves state-of-the-art performance on all 27 geospatial interpolation tasks, and on 25 out of the 27 extrapolation and super-resolution tasks. We combined the PDFM with a state-of-the-art forecasting foundation model, TimesFM, to predict unemployment and poverty, achieving performance that surpasses fully supervised forecasting. The full set of embeddings and sample code are publicly available for researchers., Comment: 28 pages, 16 figures, preprint; v2: updated github url

Graph convolutional networks (GCNs) are an effective skeleton-based human action recognition (HAR) technique. GCNs enable the specification of CNNs to a non-Euclidean frame that is more flexible. The previous GCN-based models still have a lot of issues: (I) The graph structure is the same for all model layers and input data., Comment: This paper accepted in Computers in Human Behavior Journal

The telecommunications industry's rapid evolution demands intelligent systems capable of managing complex networks and adapting to emerging technologies. While large language models (LLMs) show promise in addressing these challenges, their deployment in telecom environments faces significant constraints due to edge device limitations and inconsistent documentation. To bridge this gap, we present TeleOracle, a telecom-specialized retrieval-augmented generation (RAG) system built on the Phi-2 small language model (SLM). To improve context retrieval, TeleOracle employs a two-stage retriever that incorporates semantic chunking and hybrid keyword and semantic search. Additionally, we expand the context window during inference to enhance the model's performance on open-ended queries. We also employ low-rank adaption for efficient fine-tuning. A thorough analysis of the model's performance indicates that our RAG framework is effective in aligning Phi-2 to the telecom domain in a downstream question and answer (QnA) task, achieving a 30% improvement in accuracy over the base Phi-2 model, reaching an overall accuracy of 81.20%. Notably, we show that our model not only performs on par with the much larger LLMs but also achieves a higher faithfulness score, indicating higher adherence to the retrieved context.

There is a great need for data in computing education research. Data is needed to understand how students behave, to train models of student behavior to optimally support students, and to develop and validate new assessment tools and learning analytics techniques. However, relatively few computing education datasets are shared openly, often due to privacy regulations and issues in making sure the data is anonymous. Large language models (LLMs) offer a promising approach to create large-scale, privacy-preserving synthetic data, which can be used to explore various aspects of student learning, develop and test educational technologies, and support research in areas where collecting real student data may be challenging or impractical. This work explores generating synthetic buggy code submissions for introductory programming exercises using GPT-4o. We compare the distribution of test case failures between synthetic and real student data from two courses to analyze the accuracy of the synthetic data in mimicking real student data. Our findings suggest that LLMs can be used to generate synthetic incorrect submissions that are not significantly different from real student data with regard to test case failure distributions. Our research contributes to the development of reliable synthetic datasets for computing education research and teaching, potentially accelerating progress in the field while preserving student privacy., Comment: 7 pages, 3 tables, 1 figure

Large experimental Collaborations at the LHC (ALICE, ATLAS, CMS, and LHCb) bring together over 13,000 people from hundreds of institutes over the world. There are many (working) cultures inside these international Collaborations. It is important to acknowledge that cultural differences exist and manifest in our workspaces in various ways such as communication style, attitude and expectations, and ways to provide feedback on work, just to name a few examples. Various studies over the world have raised concerns on mental wellbeing of academics, especially PhD students. A survey on mental wellbeing was conducted among all major LHC Collaborations in 2023. The survey attracted a total of 404 replies ranging from students to experienced researchers. Although the number of responses is not small, it corresponds to only a few per cent of the whole LHC community, and hence we cannot be sure of how representative the sample is. However, this limited sample contained worrisome findings on wellbeing among LHC researchers. While moderate stress can enhance working performance and help in keeping deadlines, stress should remain in manageable level and one should take care of adequate recovery. However, there are some stress factors related to environment and ways we work that cause unnecessary load. In these proceedings, we will raise a few results on the LHC mental wellbeing survey and discuss trainings on inclusivity and cultural diversity in ALICE. These trainings aim to create a welcoming and positive work environment by strengthening the sense of community and commitment to the Collaboration. These, in turn, can have a positive impact both on wellbeing and productivity., Comment: 6 pages, ICHEP 2024

This work takes a pedagogical lens to explore the implications of generative AI (GenAI) models and tools, such as ChatGPT and GitHub Copilot, in a semester-long 2nd-year undergraduate Software Engineering Team Project. Qualitative findings from survey (39 students) and interviews (eight students) provide insights into the students' views on the impact of GenAI use on their coding experience, learning, and self-efficacy. Our results address a particular gap in understanding the role and implications of GenAI on teamwork, team-efficacy, and team dynamics. The analysis of the learning aspects is distinguished by the application of learning and pedagogy informed lenses to discuss the data. We propose a preliminary design space for GenAI-based programming learning tools highlighting the importance of considering the roles that GenAI can play during the learning process, the varying support-ability patterns that can be applied to each role, and the importance of supporting transparency in GenAI for team members and students in addition to educators.

The effectiveness of feedback in enhancing learning outcomes is well documented within Educational Data Mining (EDM). Various prior research has explored methodologies to enhance the effectiveness of feedback. Recent developments in Large Language Models (LLMs) have extended their utility in enhancing automated feedback systems. This study aims to explore the potential of LLMs in facilitating automated feedback in math education. We examine the effectiveness of LLMs in evaluating student responses by comparing 3 different models: Llama, SBERT-Canberra, and GPT4 model. The evaluation requires the model to provide both a quantitative score and qualitative feedback on the student's responses to open-ended math problems. We employ Mistral, a version of Llama catered to math, and fine-tune this model for evaluating student responses by leveraging a dataset of student responses and teacher-written feedback for middle-school math problems. A similar approach was taken for training the SBERT model as well, while the GPT4 model used a zero-shot learning approach. We evaluate the model's performance in scoring accuracy and the quality of feedback by utilizing judgments from 2 teachers. The teachers utilized a shared rubric in assessing the accuracy and relevance of the generated feedback. We conduct both quantitative and qualitative analyses of the model performance. By offering a detailed comparison of these methods, this study aims to further the ongoing development of automated feedback systems and outlines potential future directions for leveraging generative LLMs to create more personalized learning experiences., Comment: 12 pages including references, 4 figures, 9 tables

This paper is about generating motion plans for high degree-of-freedom systems that account for collisions along the entire body. A particular class of mathematical programs with complementarity constraints become useful in this regard. Optimization-based planners can tackle confined-space trajectory planning while being cognizant of robot constraints. However, introducing obstacles in this setting transforms the formulation into a non-convex problem (oftentimes with ill-posed bilinear constraints), which is non-trivial in a real-time setting. To this end, we present the FLIQC (Fast LInear Quadratic Complementarity based) motion planner. Our planner employs a novel motion model that captures the entire rigid robot as well as the obstacle geometry and ensures non-penetration between the surfaces due to the imposed constraint. We perform thorough comparative studies with the state-of-the-art, which demonstrate improved performance. Extensive simulation and hardware experiments validate our claim of generating continuous and reactive motion plans at 1 kHz for modern collaborative robots with constant minimal parameters.

Recent advancements in robotics have transformed industries such as manufacturing, logistics, surgery, and planetary exploration. A key challenge is developing efficient motion planning algorithms that allow robots to navigate complex environments while avoiding collisions and optimizing metrics like path length, sweep area, execution time, and energy consumption. Among the available algorithms, sampling-based methods have gained the most traction in both research and industry due to their ability to handle complex environments, explore free space, and offer probabilistic completeness along with other formal guarantees. Despite their widespread application, significant challenges still remain. To advance future planning algorithms, it is essential to review the current state-of-the-art solutions and their limitations. In this context, this work aims to shed light on these challenges and assess the development and applicability of sampling-based methods. Furthermore, we aim to provide an in-depth analysis of the design and evaluation of ten of the most popular planners across various scenarios. Our findings highlight the strides made in sampling-based methods while underscoring persistent challenges. This work offers an overview of the important ongoing research in robotic motion planning., Comment: 20 pages, 11 figures

There has been a long history of women innovators producing outstanding contributions to society and public benefit yet having their work passed over or sidelined or attributed to male colleagues. This phenomenon has been coined the Matilda Effect. The amendments to the record of human achievements are now taking place, with an increasing pace in recent times due to greater social enlightenment and awareness and the interest in social justice. However, there remains a gap that must be addressed. In this article, we demonstrate the disparity in scientific recognition for a handful of case studies through search data collected via Google Trends and plotted as time-series figures and choropleth maps. Search trends reflect a noticeable divergence between recognition of female and male innovators. However, we note that in more well-known cases of the Matilda Effect, such as the historical account of Rosalind Franklin vs. James Watson and Francis Crick, the differences become less pronounced, emphasizing the importance of publicizing recognition. In response to this revelation, this article presents the stories of several women innovators and their great achievements. We identify the truth behind several discoveries and inventions, while revealing the full nature of this historical problem of social exclusion.

Stellar limb darkening must be properly accounted for to accurately determine the radii of exoplanets at various wavelengths. The standard approach to address limb darkening involves either using laws with coefficients from modelled stellar spectra or determining the coefficients empirically during light curve fitting of the data. Here, we test how accurately three common laws -- quadratic, power, and a three-parameter law -- can reproduce stellar limb darkening at different wavelengths and across a broad range of stars. We show that using a quadratic limb darkening law, which is most frequently employed by the community, leads to wavelength-dependent offsets in retrieved transmission spectra. For planets with high impact parameters ($b$ larger than about 0.5) the amplitude of these offsets can reach 1\% of the transit depth which is some cases is comparable to and can even exceed the expected signals from the planetary atmosphere. Furthermore, the quadratic law causes an offset in the value of the impact parameter when it is determined by fitting the broadband transit light curves. In contrast, using the Kipping--Sing three-parameter law leads to robust retrievals. We advocate the use of this law in retrievals, especially for transits with large impact parameters., Comment: submitted to ApJ

Advanced automated AI techniques allow us to classify protein sequences and discern their biological families and functions. Conventional approaches for classifying these protein families often focus on extracting N-Gram features from the sequences while overlooking crucial motif information and the interplay between motifs and neighboring amino acids. Recently, convolutional neural networks have been applied to amino acid and motif data, even with a limited dataset of well-characterized proteins, resulting in improved performance. This study presents a model for classifying protein families using the fusion of 1D-CNN, BiLSTM, and an attention mechanism, which combines spatial feature extraction, long-term dependencies, and context-aware representations. The proposed model (ProFamNet) achieved superior model efficiency with 450,953 parameters and a compact size of 1.72 MB, outperforming the state-of-the-art model with 4,578,911 parameters and a size of 17.47 MB. Further, we achieved a higher F1 score (98.30% vs. 97.67%) with more instances (271,160 vs. 55,077) in fewer training epochs (25 vs. 30)., Comment: It is the authors original Work

Chest X-rays (X-ray images) have been proven to be effective for the diagnosis of chest diseases, including Pneumonia, Lung Opacity, and COVID-19. However, relying on traditional medical methods for diagnosis from X-ray images is prone to delays and inaccuracies because the medical personnel who evaluate the X-ray images may have preconceived biases. For this reason, researchers have proposed the use of deep learning-based techniques to facilitate the diagnosis process. The preeminent method is the use of sophisticated Convolutional Neural Networks (CNNs). In this paper, we propose a novel detection model named \textbf{AttCDCNet} for the task of X-ray image diagnosis, enhancing the popular DenseNet121 model by adding an attention block to help the model focus on the most relevant regions, using focal loss as a loss function to overcome the imbalance of the dataset problem, and utilizing depth-wise convolution to reduce the parameters to make the model lighter. Through extensive experimental evaluations, the proposed model demonstrates exceptional performance, showing better results than the original DenseNet121. The proposed model achieved an accuracy, precision and recall of 94.94%, 95.14% and 94.53%, respectively, on the COVID-19 Radiography Dataset.

Besides a dense coverage of their high latitudes by starspots, rapidly rotating cool stars also display low-latitude spots in Doppler images, although generally with a lower coverage. In contrast, flux emergence models of fast-rotating stars predict strong poleward deflection of radially rising magnetic flux as the Coriolis effect dominates over buoyancy, leaving a spot-free band around the equator. To resolve this discrepancy, we consider a flux tube near the base of the convection zone in a solar-type star rotating eight times faster than the Sun, assuming field intensification by weak-tube explosions. For the intensification to continue into to the buoyancy-dominated regime, the upper convection zone must have a significantly steeper temperature gradient than in the Sun, by a factor that is comparable with that found in 3D simulations of rotating convection. Within the hypothesis that stellar active regions stem from the base of the convection zone, flux emergence between 1-20 degree latitudes requires highly supercritical field strengths of up to 500 kG in rapidly rotating stars. These field strengths require explosions of 100-kG tubes within the convection zone, compatible with reasonable values of the superadiabatic temperature gradient associated with the more rapid rotation., Comment: 12 pages, 9 figures, published in ApJ

The recent COVID-19 pandemic has highlighted the need of studying extreme, life-threatening phenomena in advance. In this article, a zombie epidemic in Uusimaa region in Finland is modeled. A stochastic agent based simulation model is proposed and extensive simulations are conducted for this purpose. The model utilizes knowledge on defensive human behavior during crises. Studying the effects of a hypothetical zombie attack resembles examining the spread of deadly diseases and of rumors. A zombie attack is simulated in the most densely populated region in Finland. The region's exact population densities over its rasterized geographical map are utilized. Furthermore, the simulations are used to study the effect of implementing a (strict or partial) quarantine area in the epicenter. Computationally efficient Scala codes and video animations of the simulated epidemics are provided. The main findings emphasize the importance of implementing very strict measures, without delay, to stop the outbreak., Comment: 18 pages, 13 figures

By 2025, the internet of things (IoT) is projected to connect over 75 billion devices globally, fundamentally altering how we interact with our environments in both urban and rural settings. However, IoT device security remains challenging, particularly in the authentication process. Traditional cryptographic methods often struggle with the constraints of IoT devices, such as limited computational power and storage. This paper considers physical unclonable functions (PUFs) as robust security solutions, utilizing their inherent physical uniqueness to authenticate devices securely. However, traditional PUF systems are vulnerable to machine learning (ML) attacks and burdened by large datasets. Our proposed solution introduces a lightweight PUF mechanism, called LPUF-AuthNet, combining tandem neural networks (TNN) with a split learning (SL) paradigm. The proposed approach provides scalability, supports mutual authentication, and enhances security by resisting various types of attacks, paving the way for secure integration into future 6G technologies., Comment: Accepted to Proc. IEEE Globecom 2024

Recent advancements in Web agents have introduced novel architectures and benchmarks showcasing progress in autonomous web navigation and interaction. However, most existing benchmarks prioritize effectiveness and accuracy, overlooking factors like safety and trustworthiness which are essential for deploying web agents in enterprise settings. We present STWebAgentBench, a benchmark designed to evaluate web agents safety and trustworthiness across six critical dimensions, essential for reliability in enterprise applications. This benchmark is grounded in a detailed framework that defines safe and trustworthy (ST) agent behavior. Our work extends WebArena with safety templates and evaluation functions to assess safety policy compliance rigorously. We introduce the Completion Under Policy to measure task success while adhering to policies, alongside the Risk Ratio, which quantifies policy violations across dimensions, providing actionable insights to address safety gaps. Our evaluation reveals that current SOTA agents struggle with policy adherence and cannot yet be relied upon for critical business applications. We open-source this benchmark and invite the community to contribute, with the goal of fostering a new generation of safer, more trustworthy AI agents. All code, data, environment reproduction resources, and video demonstrations are available at https://sites.google.com/view/st-webagentbench/home.

A typical manipulation task consists of a manipulator equipped with a gripper to grasp and move an object with constraints on the motion of the hand-held object, which may be due to the nature of the task itself or from object-environment contacts. In this paper, we study the problem of computing joint torques and grasping forces for time-optimal motion of an object, while ensuring that the grasp is not lost and any constraints on the motion of the object, either due to dynamics, environment contact, or no-slip requirements, are also satisfied. We present a second-order cone program (SOCP) formulation of the time-optimal trajectory planning problem that considers nonlinear friction cone constraints at the hand-object and object-environment contacts. Since SOCPs are convex optimization problems that can be solved optimally in polynomial time using interior point methods, we can solve the trajectory optimization problem efficiently. We present simulation results on three examples, including a non-prehensile manipulation task, which shows the generality and effectiveness of our approach.

Public actors are often seen as slow, especially in renewing information systems, due to complex tendering and competition regulations, which delay decisions. This challenge is even greater in multi-company ecosystems. However, when faced with a common threat, the ecosystem needs to unite to face the challenge. This study explores how the Omaolo ecosystem in Finland evolved from traditional public-private cooperation to an alliance model during the COVID-19 pandemic from 2020 to 2022. It highlights how the crisis accelerated changes in operations and collaboration between public and private participants, identifying key shifts, benefits, and challenges. Key findings include the removal of traditional barriers and the creation of an alliance approach that sped up the development of Omaolo's symptom assessment tool. This improved collaboration, service scalability, and responsiveness to healthcare needs despite the initial regulatory and stakeholder alignment challenges. The study concludes that crises can drive agile responses in public ecosystems. The new collaboration model helped Omaolo to adapt quickly to changing service demands, managing healthcare patient loads more effectively. These findings highlight the value of flexible, collaborative strategies for responding to emergencies in public software ecosystems., Comment: 19 pages, 1 figure

We categorized clumps, embedded clusters, and open clusters and conducted a comparative analysis of their physical properties. Overall, the radii of open clusters are significantly larger than those of embedded clusters and clumps. The radii of embedded clusters are larger than those of clumps, which may be due to the expansion of embedded clusters. The open clusters have significantly higher masses than embedded clusters, by about one order of magnitude. Given the current mass distribution of clumps in the Milky Way, the evolutionary sequence from a single clump evolving into an embedded cluster and subsequently into an open cluster cannot account for the observed open clusters with old ages and high masses, which is also supported by N-body simulations of individual embedded clusters. To explain the mass and radius distributions of the observed open clusters, initial embedded clusters with masses higher than 3000 M$_{\odot}$ are necessary. However, the upper limit of the embedded cluster sample is less than 1000 M$_{\odot}$, and only a few ATLASGAL clumps have a mass higher than 3000 M$_{\odot}$. Thus, the currently observed clumps cannot be the "direct" precursors of the currently observed open clusters. If the Milky Way has a burst-like and time-dependent star formation history, the currently observed open clusters with old ages and high masses may come from massive clumps in the past. There is also a very real possibility that these open clusters originate from post-gas expulsion coalescence of multiple embedded clusters. We compared the separation of open clusters and the typical size of molecular clouds, and find that most molecular clouds may only form one open cluster, which supports the scenario of post-gas expulsion coalescence. Further study is necessary to distinguish between the different scenarios., Comment: Accepted for publication in A&A, 8 pages, 6 figures. arXiv admin note: text overlap with arXiv:2409.20271

As the demand for healthcare professionals increases, Latinos are increasing their representation in healthcare positions. Despite their increased representation, Latinos are more likely to be in healthcare support occupations rather than in healthcare management or professional occupation that offer higher salaries. "Excelencia" has analyzed national datasets to connect healthcare employers with institutions that are graduating and preparing Latinos for the workforce. This brief identifies the top institutions graduating Latinos (2019-2020) from certificates to the doctoral level in healthcare. These top 25 institutions are meeting the nation's economic needs by intentionally serving Latino students and producing Latino talent. This brief also highlights institutions and their intentional efforts to support post-completion success by providing educational credentials needed to support the current and future healthcare workforce.

Although more Latinos are entering the STEM workforce, they continue to be underrepresented in high-salary STEM occupations. To ensure America's future, institutions and STEM employers must both play an active role in preparing, selecting, and sourcing Latino talent for a global economy. Excelencia's research demonstrates that select institutions are investing in the recruitment, development, and completion of Latino talent and STEM employers have opportunities to strengthen their workforce. To bridge this gap, Excelencia has analyzed national datasets to connect STEM employers with institutions that are graduating and preparing Latinos for the workforce. This brief identifies the top institutions graduating Latinos (2019-20) in STEM from certificates to the doctorate level. These top 25 institutions are meeting the nation's economic needs by intentionally serving Latino students and producing Latino talent. This brief also highlights institutions and their intentional efforts to support post-completion success by providing educational credentials needed to support the current and future STEM workforce.

The nation is facing a shortage of educators prepared to meet the needs of an increasingly diverse student population. Latinos are the nation's fastest growing population but are underrepresented in higher-skilled occupations in the classroom and overrepresented in supportive roles with lower wages. This shows the need for employers and institutions to create pathways into the field of education as Latinos see the value to educate others and the community. To ensure America's future, institutions and employers in education must both play an active role in preparing, selecting, and sourcing Latino talent for a global economy. To bridge this gap, "Excelencia" has analyzed national datasets to connect employers in education with institutions that are graduating and preparing Latinos for the workforce. This brief identifies the top institutions graduating Latinos (2019-20) from certificates to the doctoral level in education. These top 25 institutions are meeting the nation's economic needs by intentionally serving Latino students and producing Latino talent. This brief also highlights institutions and their intentional efforts to support post-completion success by providing education credentials needed to support the current and future education workforce.