Search

This white paper is the output of a multidisciplinary workshop in Nairobi (Nov 2023). Led by a cross-organisational team including Microsoft Research, NEPAD, Lelapa AI, and University of Oxford. The workshop brought together diverse thought-leaders from various sectors and backgrounds to discuss the implications of Generative AI for the future of work in Africa. Discussions centred around four key themes: Macroeconomic Impacts; Jobs, Skills and Labour Markets; Workers' Perspectives and Africa-Centris AI Platforms. The white paper provides an overview of the current state and trends of generative AI and its applications in different domains, as well as the challenges and risks associated with its adoption and regulation. It represents a diverse set of perspectives to create a set of insights and recommendations which aim to encourage debate and collaborative action towards creating a dignified future of work for everyone across Africa.

This article delves into a numerical exploration of two-dimensional, incompressible, laminar flow within a confined diverging jet. The study aims to understand how variations in the inlet opening fraction and Reynolds number affect the heat transfer and flow patterns. The research employs the finite volume method with a collocated mesh to solve the governing equations. Across a broad spectrum of inlet opening fractions (0.2, 0.4, and 0.6) and Reynolds numbers (ranging from 500 to 900), the findings reveal that increasing the inlet opening fraction of the jet in the diverging channel can lead to a remarkable (53%) improvement in heat transfer while simultaneously reducing pressure loss by 90%. This outcome holds the potential to conserve energy by requiring less pumping power. Notably, this investigation is pioneering and offers novel and valuable insights into enhancing heat transfer through the utilization of a diverging channel.

This article addresses the challenge of UAV survey coverage path planning for areas that are complex concave polygons, containing exclusion zones or obstacles. While standard drone path planners typically generate coverage paths for simple convex polygons, this study proposes a method to manage more intricate regions, including boundary splits, merges, and interior holes. To achieve this, polygonal decomposition techniques are used to partition the target area into convex sub-regions. The sub-polygons are then merged using a depth-first search algorithm, followed by the generation of continuous Boustrophedon paths based on connected components. Polygonal offset by the straight skeleton method was used to ensure a constant safe distance from the exclusion zones. This approach allows UAV path planning in environments with complex geometric constraints.

Software workplaces are increasingly recognized as key spaces for professional development, where developers encounter various challenges in their roles, which they often discuss in online forums. This paper analyzes 47,368 posts on the Workplace StackExchange site, aggregating developer insights and applying topic modeling techniques. Through manual analysis, we identified 46 distinct topics grouped into seven categories: Employee Wellness, Communication, Career Movement \& Hiring, Conflicts \& Mistakes, Corporate Policies, Management/Supervisor Responsibilities, and Learning \& Technical Skills. Our findings show that approximately 30\% of discussions involve workplace conflicts, marking this as the most prominent topic. Additionally, we found that workplace culture, harassment, and other corporate policy-related issues represent significant areas of difficulty commonly discussed among developers.

We present the first broadband near- to mid-infrared (3-12 microns) transmission spectrum of the highly-irradiated (T_eq = 981 K) M dwarf rocky planet L 168-9 b (TOI-134 b) observed with the NIRSpec and MIRI instruments aboard JWST. We measure the near-infrared transit depths to a combined median precision of 20 ppm across the three visits in 54 spectroscopic channels with uniform widths of 60 pixels (~0.2 microns wide; R~100), and the mid-infrared transit depths to 61 ppm median precision in 48 wavelength bins (~0.15 microns wide; R~50). We compare the transmission spectrum of L 168-9 b to a grid of 1D thermochemical equilibrium forward models, and rule out atmospheric metallicities of less than 100x solar (mean molecular weights <4 g mol$^{-1}$) to 3-sigma confidence assuming high surface pressure (>1 bar), cloudless atmospheres. Based on photoevaporation models for L 168-9 b with initial atmospheric mass fractions ranging from 2-100%, we find that this planet could not have retained a primordial H/He atmosphere beyond the first 200 Myr of its lifetime. Follow-up MIRI eclipse observations at 15 microns could make it possible to confidently identify a CO2-dominated atmosphere on this planet if one exists., Comment: 19 pages, 11 figures, accepted for publication in The Astronomical Journal

The population of the urban areas is increasing daily, and this migration is causing serious environmental pollution. A larger population is creating pressure on the municipality's waste management and the city corporations of developing countries such as Bangladesh, further threatening human health. New generation technologies, such as the Internet of Things (IoT)-based waste management systems, can help improve this serious issue. IoT-enabled smart dustbins and mobile applications-based interactive management can effectively solve this problem. In this article, we combine these two technologies to offer an acceptable solution to this problem. The proposed waste management model enables smart dustbins to communicate with waste collectors or waste control centers whenever it is necessary. Additionally, city dwellers can use mobile applications to report their observations in their neighborhoods. As a result, both sensors and humans are involved directly in the development loop. We have conducted a detailed survey to study the acceptance of such a system in the community and received some encouraging results.

Artificial intelligence may assist healthcare systems in meeting increasing demand for pathology services while maintaining diagnostic quality and reducing turnaround time and costs. We aimed to investigate the performance of an institutionally developed system for prostate cancer detection, grading, and workflow optimization and to contrast this with commercial alternatives. From August 2021 to March 2023, we scanned 21,396 slides from 1,147 patients with positive biopsies. We developed models for cancer detection, grading, and screening of equivocal cases for IHC ordering. We compared a task-specific model trained using the PANDA dataset of prostate cancer biopsies with one built using features extracted by the general-purpose histology foundation model, UNI and compare their performance in an unfiltered prospectively collected dataset that reflects our patient population (1737 slides,95 patients). We evaluated the contributions of a bespoke model designed to improve sensitivity in detecting small cancer foci and scoring of broader patterns observed at lower resolution. We found high concordance between the developed systems and pathologist reference in detection (AUC 98.5, sensitivity 95.0, and specificity 97.8), ISUP grading (quadratic Cohen's kappa 0.869), grade group 3 or higher (AUC 97.5, sensitivity 94.9, specificity 96.6) and comparable to published data from commercial systems. Screening could reduce IHC ordering for equivocal cases by 44.5% with an overall error rate of 1.8% (1.4% false positive, 0.4% false negative rates). Institutions like academic medical centers that have high scanning volumes and report abstraction capabilities can develop accurate computational pathology models for internal use. These models have the potential to aid in quality control role and to improve workflow in the pathology lab to help meet future challenges in prostate cancer diagnosis.

This work focuses on the efficiency of the knowledge distillation approach in generating a lightweight yet powerful BERT based model for natural language processing applications. After the model creation, we applied the resulting model, LastBERT, to a real-world task classifying severity levels of Attention Deficit Hyperactivity Disorder (ADHD)-related concerns from social media text data. Referring to LastBERT, a customized student BERT model, we significantly lowered model parameters from 110 million BERT base to 29 million, resulting in a model approximately 73.64% smaller. On the GLUE benchmark, comprising paraphrase identification, sentiment analysis, and text classification, the student model maintained strong performance across many tasks despite this reduction. The model was also used on a real-world ADHD dataset with an accuracy and F1 score of 85%. When compared to DistilBERT (66M) and ClinicalBERT (110M), LastBERT demonstrated comparable performance, with DistilBERT slightly outperforming it at 87%, and ClinicalBERT achieving 86% across the same metrics. These findings highlight the LastBERT model's capacity to classify degrees of ADHD severity properly, so it offers a useful tool for mental health professionals to assess and comprehend material produced by users on social networking platforms. The study emphasizes the possibilities of knowledge distillation to produce effective models fit for use in resource-limited conditions, hence advancing NLP and mental health diagnosis. Furthermore underlined by the considerable decrease in model size without appreciable performance loss is the lower computational resources needed for training and deployment, hence facilitating greater applicability. Especially using readily available computational tools like Google Colab. This study shows the accessibility and usefulness of advanced NLP methods in pragmatic world applications., Comment: 20 figures, 31 pages, review 1 from plos one journal

Organizational success in todays competitive employment market depends on choosing the right staff. This work evaluates software engineer profiles using an automated staff selection method based on advanced natural language processing (NLP) techniques. A fresh dataset was generated by collecting LinkedIn profiles with important attributes like education, experience, skills, and self-introduction. Expert feedback helped transformer models including RoBERTa, DistilBERT, and a customized BERT variation, LastBERT, to be adjusted. The models were meant to forecast if a candidate's profile fit the selection criteria, therefore allowing automated ranking and assessment. With 85% accuracy and an F1 score of 0.85, RoBERTa performed the best; DistilBERT provided comparable results at less computing expense. Though light, LastBERT proved to be less effective, with 75% accuracy. The reusable models provide a scalable answer for further categorization challenges. This work presents a fresh dataset and technique as well as shows how transformer models could improve recruiting procedures. Expanding the dataset, enhancing model interpretability, and implementing the system in actual environments will be part of future activities., Comment: 6 pages, 12 figures, conference paper

Vector Symbolic Architectures (VSAs) are one approach to developing Neuro-symbolic AI, where two vectors in $\mathbb{R}^d$ are `bound' together to produce a new vector in the same space. VSAs support the commutativity and associativity of this binding operation, along with an inverse operation, allowing one to construct symbolic-style manipulations over real-valued vectors. Most VSAs were developed before deep learning and automatic differentiation became popular and instead focused on efficacy in hand-designed systems. In this work, we introduce the Hadamard-derived linear Binding (HLB), which is designed to have favorable computational efficiency, and efficacy in classic VSA tasks, and perform well in differentiable systems. Code is available at https://github.com/FutureComputing4AI/Hadamard-derived-Linear-Binding, Comment: To appear in the 38th Conference on Neural Information Processing Systems (NeurIPS 2024)

Large Language Models (LLMs) have shown great potential in automating code generation; however, their ability to generate accurate circuit-level SPICE code remains limited due to a lack of hardware-specific knowledge. In this paper, we analyze and identify the typical limitations of existing LLMs in SPICE code generation. To address these limitations, we present SPICEPilot a novel Python-based dataset generated using PySpice, along with its accompanying framework. This marks a significant step forward in automating SPICE code generation across various circuit configurations. Our framework automates the creation of SPICE simulation scripts, introduces standardized benchmarking metrics to evaluate LLM's ability for circuit generation, and outlines a roadmap for integrating LLMs into the hardware design process. SPICEPilot is open-sourced under the permissive MIT license at https://github.com/ACADLab/SPICEPilot.git., Comment: 6 pages, 2 figures, 5 tables

The integration of the Internet of Things (IoT) and modern Artificial Intelligence (AI) has given rise to a new paradigm known as the Artificial Intelligence of Things (AIoT). In this survey, we provide a systematic and comprehensive review of AIoT research. We examine AIoT literature related to sensing, computing, and networking & communication, which form the three key components of AIoT. In addition to advancements in these areas, we review domain-specific AIoT systems that are designed for various important application domains. We have also created an accompanying GitHub repository, where we compile the papers included in this survey: https://github.com/AIoT-MLSys-Lab/AIoT-Survey. This repository will be actively maintained and updated with new research as it becomes available. As both IoT and AI become increasingly critical to our society, we believe AIoT is emerging as an essential research field at the intersection of IoT and modern AI. We hope this survey will serve as a valuable resource for those engaged in AIoT research and act as a catalyst for future explorations to bridge gaps and drive advancements in this exciting field., Comment: Accepted in ACM Transactions on Sensor Networks (TOSN)

Knee osteoarthritis(KO) is a degenerative joint disease that can cause severe pain and impairment. With increased prevalence, precise diagnosis by medical imaging analytics is crucial for appropriate illness management. This research investigates a comparative analysis between traditional machine learning techniques and new deep learning models for diagnosing KO severity from X-ray pictures. This study does not introduce new architectural innovations but rather illuminates the robust applicability and comparative effectiveness of pre-existing ViT models in a medical imaging context, specifically for KO severity diagnosis. The insights garnered from this comparative analysis advocate for the integration of advanced ViT models in clinical diagnostic workflows, potentially revolutionizing the precision and reliability of KO assessments. This study does not introduce new architectural innovations but rather illuminates the robust applicability and comparative effectiveness of pre-existing ViT models in a medical imaging context, specifically for KO severity diagnosis. The insights garnered from this comparative analysis advocate for the integration of advanced ViT models in clinical diagnostic workflows, potentially revolutionizing the precision & reliability of KO assessments. The study utilizes an osteoarthritis dataset from the Osteoarthritis Initiative (OAI) comprising images with 5 severity categories and uneven class distribution. While classic machine learning models like GaussianNB and KNN struggle in feature extraction, Convolutional Neural Networks such as Inception-V3, VGG-19 achieve better accuracy between 55-65% by learning hierarchical visual patterns. However, Vision Transformer architectures like Da-VIT, GCViT and MaxViT emerge as indisputable champions, displaying 66.14% accuracy, 0.703 precision, 0.614 recall, AUC exceeding 0.835 thanks to self-attention processes.

Scientific workflows have become highly heterogenous, leveraging distributed facilities such as High Performance Computing (HPC), Artificial Intelligence (AI), Machine Learning (ML), scientific instruments (data-driven pipelines) and edge computing. As a result, Identity and Access Management (IAM) and Cybersecurity challenges across the diverse hardware and software stacks are growing. Nevertheless, scientific productivity relies on lowering access barriers via seamless, single sign-on (SSO) and federated login while ensuring access controls and compliance. We present an implementation of a federated IAM solution, which is coupled with multiple layers of security controls, multi-factor authentication, cloud-native protocols, and time-limited role-based access controls (RBAC) that has been co-designed and deployed for the Isambard-AI and HPC supercomputing Digital Research Infrastructures (DRIs) in the UK. Isambard DRIs as a national research resource are expected to comply with regulatory frameworks. Implementation details for monitoring, alerting and controls are outlined in the paper alongside selected user stories for demonstrating IAM workflows for different roles., Comment: 9 pages, 2 figures

Ultraviolet wavelengths offer unique insights into aerosols in exoplanetary atmospheres. However, only a handful of exoplanets have been observed in the ultraviolet to date. Here, we present the ultraviolet-visible transmission spectrum of the inflated hot Jupiter WASP-127b. We observed one transit of WASP-127b with WFC3/UVIS G280 as part of the Hubble Ultraviolet-optical Survey of Transiting Legacy Exoplanets (HUSTLE), obtaining a transmission spectrum from 200-800 nm. Our reductions yielded a broad-band transit depth precision of 91 ppm and a median precision of 240 ppm across 59 spectral channels. Our observations are suggestive of a high-altitude cloud layer with forward modeling showing they are composed of sub-micron particles and retrievals indicating a high opacity patchy cloud. While our UVIS/G280 data only offers weak evidence for Na, adding archival HST WFC3/IR and STIS observations raises the overall Na detection significance to 4.1-sigma. Our work demonstrates the capabilities of HST WFC3/UVIS G280 observations to probe the aerosols and atmospheric composition of transiting hot Jupiters with comparable precision to HST STIS., Comment: 25 pages, 8 figures, 6 tables

This study applies Bayesian models to predict hotel booking cancellations, a key challenge affecting resource allocation, revenue, and customer satisfaction in the hospitality industry. Using a Kaggle dataset with 36,285 observations and 17 features, Bayesian Logistic Regression and Beta-Binomial models were implemented. The logistic model, applied to 12 features and 5,000 randomly selected observations, outperformed the Beta-Binomial model in predictive accuracy. Key predictors included the number of adults, children, stay duration, lead time, car parking space, room type, and special requests. Model evaluation using Leave-One-Out Cross-Validation (LOO-CV) confirmed strong alignment between observed and predicted outcomes, demonstrating the model's robustness. Special requests and parking availability were found to be the strongest predictors of cancellation. This Bayesian approach provides a valuable tool for improving booking management and operational efficiency in the hotel industry.

More than 3 kV breakdown voltage was demonstrated in monolithic bidirectional GaN HEMTs for the first time having potential applications in 1200V or 1700V-class novel power converters. The on resistance of the fabricated transistors was ~20 ohm.mm or ~11 mili ohm.cm^2. Breakdown voltage was optimized by utilizing two field plates in either side of the transistor and optimizing their geometry. Shorter first field plate lengths (less than 2 micron) resulted in higher breakdown voltage and the possible reason for this was discussed. The transistors had a steep subthreshold swing of 92 mV / dec. The on/off ratio was greater than 10^5 and it was limited by the tool capacity. The fabricated 3 kV transistor was benchmarked against the state-of-the-art monolithic bidirectional GaN HEMTs in the performance matrices of breakdown voltage and on resistance, that showed crucial progress., Comment: 4 pages, 7 figures

Large Language Models (LLMs) have demonstrated remarkable success as general-purpose task solvers across various fields, including NLP, healthcare, finance, and law. However, their capabilities remain limited when addressing domain-specific problems, particularly in downstream NLP tasks. Research has shown that models fine-tuned on instruction-based downstream NLP datasets outperform those that are not fine-tuned. While most efforts in this area have primarily focused on resource-rich languages like English and broad domains, little attention has been given to multilingual settings and specific domains. To address this gap, this study focuses on developing a specialized LLM, LlamaLens, for analyzing news and social media content in a multilingual context. To the best of our knowledge, this is the first attempt to tackle both domain specificity and multilinguality, with a particular focus on news and social media. Our experimental setup includes 19 tasks, represented by 52 datasets covering Arabic, English, and Hindi. We demonstrate that LlamaLens outperforms the current state-of-the-art (SOTA) on 16 testing sets, and achieves comparable performance on 10 sets. We make the models and resources publicly available for the research community.(https://huggingface.co/QCRI), Comment: LLMs, Multilingual, Language Diversity, Large Language Models, Social Media, News Media, Specialized LLMs, Fact-checking, Media Analysis

The financial sector's dependence on digital infrastructure increases its vulnerability to cybersecurity threats, requiring strong IT security protocols with other entities. This collaboration, however, is often identified as the most vulnerable link in the chain of cybersecurity. Adopting both symbolic and substantive measures lessens the impact of IT security spending on decreasing the frequency of data security breaches in the long run. The Protection Motivation Theory clarifies actions triggered by data sharing with other organizations, and the Institutional theory aids in comprehending the intricate relationship between transparency and organizational conduct. We investigate how things like regulatory pressure, teamwork among institutions, and people's motivations to protect themselves influence cybersecurity. By using simple theories to understand these factors, this research aims to provide insights that can help financial institutions make better decisions to protect. We have also included the discussion, conclusion, and future directions in regard to collaboration in financial sector cybersecurity for exploring impact of resource sharing., Comment: 8 pages, 2 figures

The synergy between real and reciprocal space topology is anticipated to yield a diverse array of topological properties in quantum materials. We address this pursuit by achieving topologically safeguarded magnetic order in novel Weyl metallic Heusler alloy, Mn$_{2}$Pd$_{0.5}$Ir$_{0.5}$Sn. The system possesses non-centrosymmetric D$_{2d}$ crystal symmetry with notable spin-orbit coupling effects. Our first principles calculations confirm the topological non-trivial nature of band structure, including 42 pairs of Weyl nodes at/near the Fermi level, offering deeper insights into the observed anomalous Hall effect mediated by intrinsic Berry curvature. A unique canted magnetic ordering facilitates such rich topological features, manifesting through an exceptionally large topological Hall effect at low fields. The latter is sustained even at room temperature and compared with other known topological magnetic materials. Detailed micromagnetic simulations demonstrate the possible existence of an antiskyrmion lattice. Our results underscore the $D_{2d}$ Heusler magnets as a possible platform to explore the intricate interplay of non-trivial topology across real and reciprocal spaces to leverage a plethora of emergent properties for spintronic applications.

Visual Question Answer (VQA) poses the problem of answering a natural language question about a visual context. Bangla, despite being a widely spoken language, is considered low-resource in the realm of VQA due to the lack of a proper benchmark dataset. The absence of such datasets challenges models that are known to be performant in other languages. Furthermore, existing Bangla VQA datasets offer little cultural relevance and are largely adapted from their foreign counterparts. To address these challenges, we introduce a large-scale Bangla VQA dataset titled ChitroJera, totaling over 15k samples where diverse and locally relevant data sources are used. We assess the performance of text encoders, image encoders, multimodal models, and our novel dual-encoder models. The experiments reveal that the pre-trained dual-encoders outperform other models of its scale. We also evaluate the performance of large language models (LLMs) using prompt-based techniques, with LLMs achieving the best performance. Given the underdeveloped state of existing datasets, we envision ChitroJera expanding the scope of Vision-Language tasks in Bangla.

While content moderation in online platforms marginalizes users in the Global South at large, users of certain identities are further marginalized. Such users often come from Indigenous ethnic minority groups or identify as women. Through a qualitative study based on 18 semi-structured interviews, this paper explores how such users' experiences with hate speech online in Bangladesh are shaped by their intersectional identities. Through a civics-oriented approach, we examined the spectrum of their legal status, membership, rights, and participation as users of online platforms. Drawing analogies with the concept of citizenship, we develop the concept of usership that offers a user-centered metaphor in studying moderation and platform governance., Comment: accepted to ICTD 24

Cricket, a popular bat-and-ball game in South Asia, is played between two 11-player teams. The Pakistan Super League (PSL) is a commercial T20 domestic league comprised of six franchise-owned teams, where player selection is competitive. In this study, an existing role-based ranking structure is assessed that evaluates player performance in the context of team belongingness to generate optimal Pakistan cricket teams for international tournaments. The underlying assumption is that since cricket is fundamentally a team sport, the performance of players compared to their peers plays a crucial role in their selection. To accomplish this, a network is generated using ball-by-ball data from previous PSL matches (2016-2022), and social network analysis (SNA) techniques such as centrality and clustering coefficient measures, are employed to quantify the level of belongingness among Pakistani cricket players within the PSL network. Characteristic network models, such as the Erd\"os-R\'enyi, Watts-Strogatz, and Barab\'asi-Albert models are utilized to gain insights into the small-world properties of the network. By ranking players using centrality and clustering coefficient metrics, four teams are formulated, and these teams are subsequently compared to the official squad selected by the Pakistan Cricket Board (PCB) for the recent ICC Men's T20 World Cup in 2022. This evaluation sheds light on the allegations of nepotism and favoritism in team formations that have been attributed to the PCB over the years. Based on our findings, out of the 18 players in the World Cup squad, 11 were included in the teams we formed. While most of the 7 players who were not included in our teams were still selected for the ICC Men's T20 World Cup 2022, they ranked highly in our rankings, suggesting their potential and competence., Comment: 8 ages, 10 figures, 2 tables

Effective real-time monitoring technique is crucial for detecting material degradation and maintaining the structural integrity of nuclear systems to ensure both safety and operational efficiency. Traditional physical sensor systems face limitations such as installation challenges, high costs, and difficulties in measuring critical parameters in hard-to-reach or harsh environments, often resulting in incomplete data coverage. Machine learning-driven virtual sensors offer a promising solution by enhancing physical sensor capabilities to monitor critical degradation indicators like pressure, velocity, and turbulence. However, conventional machine learning models struggle with real-time monitoring due to the high-dimensional nature of reactor data and the need for frequent retraining. This paper explores the use of Deep Operator Networks (DeepONet) within a digital twin (DT) framework to predict key thermal-hydraulic parameters in the hot leg of an AP-1000 Pressurized Water Reactor (PWR). In this study, DeepONet is trained with different operational conditions, which relaxes the requirement of continuous retraining, making it suitable for online and real-time prediction components for DT. Our results show that DeepONet achieves accurate predictions with low mean squared error and relative L2 error and can make predictions on unknown data 160,000 times faster than traditional finite element (FE) simulations. This speed and accuracy make DeepONet a powerful tool for tracking conditions that contribute to material degradation in real-time, enhancing reactor safety and longevity.

The proliferation of transliterated texts in digital spaces has emphasized the need for detecting and classifying hate speech in languages beyond English, particularly in low-resource languages. As online discourse can perpetuate discrimination based on target groups, e.g. gender, religion, and origin, multi-label classification of hateful content can help in comprehending hate motivation and enhance content moderation. While previous efforts have focused on monolingual or binary hate classification tasks, no work has yet addressed the challenge of multi-label hate speech classification in transliterated Bangla. We introduce BanTH, the first multi-label transliterated Bangla hate speech dataset comprising 37.3k samples. The samples are sourced from YouTube comments, where each instance is labeled with one or more target groups, reflecting the regional demographic. We establish novel transformer encoder-based baselines by further pre-training on transliterated Bangla corpus. We also propose a novel translation-based LLM prompting strategy for transliterated text. Experiments reveal that our further pre-trained encoders are achieving state-of-the-art performance on the BanTH dataset, while our translation-based prompting outperforms other strategies in the zero-shot setting. The introduction of BanTH not only fills a critical gap in hate speech research for Bangla but also sets the stage for future exploration into code-mixed and multi-label classification challenges in underrepresented languages.

Millimeter wave (mmWave)-enabled unmanned aerial vehicle (UAV) swarm networks (UAVSNs) can utilize a large spectrum of resources to provide low latency and high data transmission rate. Additionally, owing to the short wavelength, UAVs equipped with large antenna arrays can form secure narrow directive beam to establish communication with less interference. However, due to the high UAV mobility, limited beam coverage, beam misalignment, and high path loss, it is very challenging to adopt the mmWave communication in UAVSNs. In this article, we present a comprehensive survey on neighbor discovery and beam alignment techniques for directional communication in mmWave-enabled UAVSNs. The existing techniques are reviewed and compared with each other. We also discuss key open issues and challenges with potential research direction.

Isambard-AI is a new, leadership-class supercomputer, designed to support AI-related research. Based on the HPE Cray EX4000 system, and housed in a new, energy efficient Modular Data Centre in Bristol, UK, Isambard-AI employs 5,448 NVIDIA Grace-Hopper GPUs to deliver over 21 ExaFLOP/s of 8-bit floating point performance for LLM training, and over 250 PetaFLOP/s of 64-bit performance, for under 5MW. Isambard-AI integrates two, all-flash storage systems: a 20 PiByte Cray ClusterStor and a 3.5 PiByte VAST solution. Combined these give Isambard-AI flexibility for training, inference and secure data accesses and sharing. But it is the software stack where Isambard-AI will be most different from traditional HPC systems. Isambard-AI is designed to support users who may have been using GPUs in the cloud, and so access will more typically be via Jupyter notebooks, MLOps, or other web-based, interactive interfaces, rather than the approach used on traditional supercomputers of sshing into a system before submitting jobs to a batch scheduler. Its stack is designed to be quickly and regularly upgraded to keep pace with the rapid evolution of AI software, with full support for containers. Phase 1 of Isambard-AI is due online in May/June 2024, with the full system expected in production by the end of the year., Comment: 11 pages, 11 figures

Ternary content addressable memories (TCAMs) are useful for certain computing tasks since they allow us to compare a search query with a whole dataset stored in the memory array. They can also unlock unique advantages for cryogenic applications like quantum computing, high-performance computing, and space exploration by improving speed and energy efficiency through parallel searching. This paper explores the design and implementation of a cryogenic ternary content addressable memory based on ferroelectric superconducting quantum interference devices (FeSQUIDs). The use of FeSQUID for designing the TCAM provides several unique advantages. First, we can get binary decisions (zero or non-zero voltage) for matching and mismatching conditions without using any peripheral circuitry. Moreover, the proposed TCAM needs ultra-low energy (1.36 aJ and 26.5 aJ average energy consumption for 1-bit binary and ternary search, respectively), thanks to the use of energy-efficient SQUIDs. Finally, we show the efficiency of FeSQUID through the brain-inspired application of Hyperdimensional Computing (HDC). Here, the FeSQUID-based TCAM implements the associative memory to support the highly parallel search needed in the inference step. We estimate an energy consumption of 89.4 fJ per vector comparison using a vector size of 10,000 bits. We also compare the FeSQUID-based TCAM array with the 5nm FinFET-based cryogenic SRAM-based TCAM array and observe that the proposed FeSQUID-based TCAM array consumes over one order of magnitude lower energy while performing the same task., Comment: 6 figures

We measure the growth of cosmic density fluctuations on large scales and across the redshift range $0.3

Low-light image enhancement (LLIE) is essential for numerous computer vision tasks, including object detection, tracking, segmentation, and scene understanding. Despite substantial research on improving low-quality images captured in underexposed conditions, clear vision remains critical for autonomous vehicles, which often struggle with low-light scenarios, signifying the need for continuous research. However, paired datasets for LLIE are scarce, particularly for street scenes, limiting the development of robust LLIE methods. Despite using advanced transformers and/or diffusion-based models, current LLIE methods struggle in real-world low-light conditions and lack training on street-scene datasets, limiting their effectiveness for autonomous vehicles. To bridge these gaps, we introduce a new dataset LoLI-Street (Low-Light Images of Streets) with 33k paired low-light and well-exposed images from street scenes in developed cities, covering 19k object classes for object detection. LoLI-Street dataset also features 1,000 real low-light test images for testing LLIE models under real-life conditions. Furthermore, we propose a transformer and diffusion-based LLIE model named "TriFuse". Leveraging the LoLI-Street dataset, we train and evaluate our TriFuse and SOTA models to benchmark on our dataset. Comparing various models, our dataset's generalization feasibility is evident in testing across different mainstream datasets by significantly enhancing images and object detection for practical applications in autonomous driving and surveillance systems. The complete code and dataset is available on https://github.com/tanvirnwu/TriFuse., Comment: Accepted by the Asian Conference on Computer Vision (ACCV 2024)

Dynamic quantum circuits (DQCs) incorporate mid-circuit measurements and gates conditioned on these measurement outcomes. DQCs can prepare certain long-range entangled states in constant depth, making them a promising route to preparing complex quantum states on devices with a limited coherence time. Almost all constructions of DQCs for state preparation have been formulated analytically, relying on special structure in the target states. In this work, we approach the problem of state preparation variationally, developing scalable tensor network algorithms which find high-fidelity DQC preparations for generic states. We apply our algorithms to critical states, random matrix product states, and subset states. We consider both DQCs with a fixed number of ancillae and those with an extensive number of ancillae. Even in the few ancillae regime, the DQCs discovered by our algorithms consistently prepare states with lower infidelity than a static quantum circuit of the same depth. Notably, we observe constant fidelity gains across system sizes and circuit depths. For DQCs with an extensive number of ancillae, we introduce scalable methods for decoding measurement outcomes, including a neural network decoder and a real-time decoding protocol. Our work demonstrates the power of an algorithmic approach to generating DQC circuits, broadening their scope of applications to new areas of quantum computing.

Inherited retinal diseases (IRDs) are a diverse group of genetic disorders that lead to progressive vision loss and are a major cause of blindness in working-age adults. The complexity and heterogeneity of IRDs pose significant challenges in diagnosis, prognosis, and management. Recent advancements in artificial intelligence (AI) offer promising solutions to these challenges. However, the rapid development of AI techniques and their varied applications have led to fragmented knowledge in this field. This review consolidates existing studies, identifies gaps, and provides an overview of AI's potential in diagnosing and managing IRDs. It aims to structure pathways for advancing clinical applications by exploring AI techniques like machine learning and deep learning, particularly in disease detection, progression prediction, and personalized treatment planning. Special focus is placed on the effectiveness of convolutional neural networks in these areas. Additionally, the integration of explainable AI is discussed, emphasizing its importance in clinical settings to improve transparency and trust in AI-based systems. The review addresses the need to bridge existing gaps in focused studies on AI's role in IRDs, offering a structured analysis of current AI techniques and outlining future research directions. It concludes with an overview of the challenges and opportunities in deploying AI for IRDs, highlighting the need for interdisciplinary collaboration and the continuous development of robust, interpretable AI models to advance clinical applications.

Producing optimized and accurate transmission spectra of exoplanets from telescope data has traditionally been a manual and labor-intensive procedure. Here we present the results of the first attempt to improve and standardize this procedure using artificial intelligence (AI) based processing of light curves and spectroscopic data from transiting exoplanets observed with the Hubble Space Telescope's (HST) Wide Field Camera 3 (WFC3) instrument. We implement an AI-based parameter optimizer that autonomously operates the Eureka pipeline to produce homogeneous transmission spectra of publicly available HST WFC3 datasets, spanning exoplanet types from hot Jupiters to sub-Neptunes. Surveying 43 exoplanets with temperatures between 280 and 2580 Kelvin, we confirm modeled relationships between the amplitude of the water band at 1.4um in hot Jupiters and their equilibrium temperatures. We also identify a similar, novel trend in Neptune/sub-Neptune atmospheres, but shifted to cooler temperatures. Excitingly, a planet mass versus equilibrium temperature diagram reveals a "Clear Sky Corridor," where planets between 700 and 1700 Kelvin (depending on the mass) show stronger 1.4um H2O band measurements. This novel trend points to metallicity as a potentially important driver of aerosol formation. As we unveil and include these new discoveries into our understanding of aerosol formation, we enter a thrilling future for the study of exoplanet atmospheres. With HST sculpting this foundational understanding for aerosol formation in various exoplanet types, ranging from Jupiters to sub-Neptunes, we present a compelling platform for the James Webb Space Telescope (JWST) to discover similar atmospheric trends for more planets across a broader wavelength range., Comment: Accepted to AJ. 14 pages, 5 figures, 3 tables

To address an ever-increasing demand for ubiquitous high-speed connectivity, mobile networks have intensified their deployment process. However, achieving this target has proven to be a challenge and has led to a surge in overall energy consumption. In recent years, non-terrestrial networks (NTNs) have been endorsed as a potential solution to these problems by complementing the coverage of the terrestrial network in areas with limited network deployment. To this end, this paper proposes an integrated terrestrial and non-terrestrial network (TN-NTN) that utilises the overall available communication resources to expand coverage and meet Quality of Service (QoS) requirements during high-traffic hours in any deployment scenario. Importantly, our framework allows to drastically reduce the terrestrial network energy consumption during low-traffic hours. Specifically, we introduce a novel radio resource management algorithm, BLASTER (Bandwidth SpLit, User ASsociation, and PowEr ContRol), which integrates bandwidth allocation, user equipment (UE) association, power control, and base station activation within the TN-NTN. This algorithm aims to optimize network resource allocation fairness and energy consumption dynamically, demonstrating new opportunities in deploying satellite networks in legacy cellular systems. Our study offers a comprehensive analysis of the integrated network model, emphasizing the effective balance between energy saving and QoS, and proposing practical solutions to meet the fluctuating traffic demands of cellular networks., Comment: Submitted to IEEE Transactions on Wireless Communications

Autonomous unmanned aerial vehicle (UAV) swarm networks (UAVSNs) can effectively execute surveillance, connectivity, and computing services to ground users (GUs). These missions require trajectory planning, UAV-GUs association, task offloading, next-hop selection, and resources such as transmit power, bandwidth, caching, and computing allocation to improve network performances. Owing to the highly dynamic topology, limited resources, and non-availability of global knowledge, optimizing network performance in UAVSNs is very intricate. Hence, it requires an adaptive joint optimization framework that can tackle both discrete and continuous decision variables to ensure optimal network performance under dynamic constraints. Multi-agent deep reinforcement learning-based adaptive actor-critic framework can efficiently address these problems. This paper investigates the recent evolutions of actor-critic frameworks to deal with joint optimization problems in UAVSNs. In addition, challenges and potential solutions are addressed as research directions.

Advancements in generative modeling are pushing the state-of-the-art in synthetic medical image generation. These synthetic images can serve as an effective data augmentation method to aid the development of more accurate machine learning models for medical image analysis. While the fidelity of these synthetic images has progressively increased, the diversity of these images is an understudied phenomenon. In this work, we propose the SDICE index, which is based on the characterization of similarity distributions induced by a contrastive encoder. Given a synthetic dataset and a reference dataset of real images, the SDICE index measures the distance between the similarity score distributions of original and synthetic images, where the similarity scores are estimated using a pre-trained contrastive encoder. This distance is then normalized using an exponential function to provide a consistent metric that can be easily compared across domains. Experiments conducted on the MIMIC-chest X-ray and ImageNet datasets demonstrate the effectiveness of SDICE index in assessing synthetic medical dataset diversity., Comment: Accepted at BMVC 2024 - PFATCV

This paper examines architectural challenges and opportunities arising from Nomadic Networks in the context of emerging 6G research. Nomadic networks are proposed as a solution to the limitations of stationary communication networks, providing enhanced connectivity for dynamic and mobile environments, such as large outdoor events, emergency situations and mobile industrial applications. The key requirements for nomadic networks are outlined, including functional split within the Radio Access Network and robust backhauling solutions. It also addresses the complexity of managing network components, ensuring interoperability with existing systems and maintaining stakeholder trust. A comprehensive architectural framework for Nomadic Networks in 6G is proposed. Different deployment scenarios for Nomadic Networks are investigated, including spawned, steered, and wandering Radio Access Networks as well as integrated, migrated and donated Core Networks. By introducing Nomadic-Network-as-a-Service and a related orchestration framework, the potential for flexible and scalable network management is emphasized. By addressing the architectural challenges, the paper provides a path for the successful implementation of Nomadic Networks towards more adaptable and flexible 6G networks that can meet the evolving needs of multiple sectors., Comment: 7 pages, 2 figures

Motivated by the challenges of analyzing high-dimensional ($p \gg n$) sequencing data from longitudinal microbiome studies, where samples are collected at multiple time points from each subject, we propose supervised functional tensor singular value decomposition (SupFTSVD), a novel dimensionality reduction method that leverages auxiliary information in the dimensionality reduction of high-dimensional functional tensors. Although multivariate functional principal component analysis is a natural choice for dimensionality reduction of multivariate functional data, it becomes computationally burdensome in high-dimensional settings. Low-rank tensor decomposition is a feasible alternative and has gained popularity in recent literature, but existing methods in this realm are often incapable of simultaneously utilizing the temporal structure of the data and subject-level auxiliary information. SupFTSVD overcomes these limitations by generating low-rank representations of high-dimensional functional tensors while incorporating subject-level auxiliary information and accounting for the functional nature of the data. Moreover, SupFTSVD produces low-dimensional representations of subjects, features, and time, as well as subject-specific trajectories, providing valuable insights into the biological significance of variations within the data. In simulation studies, we demonstrate that our method achieves notable improvement in tensor approximation accuracy and loading estimation by utilizing auxiliary information. Finally, we applied SupFTSVD to two longitudinal microbiome studies where biologically meaningful patterns in the data were revealed.

Since the publication of CLIP, the approach of using InfoNCE loss for contrastive pre-training has become widely popular for bridging two or more modalities. Despite its wide adoption, CLIP's original design choices of L2 normalization and cosine similarity logit have rarely been revisited. We have systematically experimented with alternative geometries and softmax logits for language-image pre-training and identified that variants with intuitive Euclidean geometry, Euclidean CLIP (EuCLIP), match or exceed the performance of CLIP and support hierarchical relationships at least as well as more complicated hyperbolic alternative., Comment: ECCV 2024 - Beyond Euclidean Workshop

Arabic, with its rich diversity of dialects, remains significantly underrepresented in Large Language Models, particularly in dialectal variations. We address this gap by introducing seven synthetic datasets in dialects alongside Modern Standard Arabic (MSA), created using Machine Translation (MT) combined with human post-editing. We present AraDiCE, a benchmark for Arabic Dialect and Cultural Evaluation. We evaluate LLMs on dialect comprehension and generation, focusing specifically on low-resource Arabic dialects. Additionally, we introduce the first-ever fine-grained benchmark designed to evaluate cultural awareness across the Gulf, Egypt, and Levant regions, providing a novel dimension to LLM evaluation. Our findings demonstrate that while Arabic-specific models like Jais and AceGPT outperform multilingual models on dialectal tasks, significant challenges persist in dialect identification, generation, and translation. This work contributes ~45K post-edited samples, a cultural benchmark, and highlights the importance of tailored training to improve LLM performance in capturing the nuances of diverse Arabic dialects and cultural contexts. We will release the dialectal translation models and benchmarks curated in this study., Comment: Benchmarking, Culturally Informed, Large Language Models, Arabic NLP, LLMs

In the medical domain, acquiring large datasets poses significant challenges due to privacy concerns. Nonetheless, the development of a robust deep-learning model for retinal disease diagnosis necessitates a substantial dataset for training. The capacity to generalize effectively on smaller datasets remains a persistent challenge. The scarcity of data presents a significant barrier to the practical implementation of scalable medical AI solutions. To address this issue, we've combined a wide range of data sources to improve performance and generalization to new data by giving it a deeper understanding of the data representation from multi-modal datasets and developed a self-supervised framework based on large language models (LLMs), SwinV2 to gain a deeper understanding of multi-modal dataset representations, enhancing the model's ability to extrapolate to new data for the detection of eye diseases using optical coherence tomography (OCT) images. We adopt a two-phase training methodology, self-supervised pre-training, and fine-tuning on a downstream supervised classifier. An ablation study conducted across three datasets employing various encoder backbones, without data fusion, with low data availability setting, and without self-supervised pre-training scenarios, highlights the robustness of our method. Our findings demonstrate consistent performance across these diverse conditions, showcasing superior generalization capabilities compared to the baseline model, ResNet-50., Comment: 25 pages, 9 tables, 10 figures

Coronary heart disease (CHD) is a severe cardiac disease, and hence, its early diagnosis is essential as it improves treatment results and saves money on medical care. The prevailing development of quantum computing and machine learning (ML) technologies may bring practical improvement to the performance of CHD diagnosis. Quantum machine learning (QML) is receiving tremendous interest in various disciplines due to its higher performance and capabilities. A quantum leap in the healthcare industry will increase processing power and optimise multiple models. Techniques for QML have the potential to forecast cardiac disease and help in early detection. To predict the risk of coronary heart disease, a hybrid approach utilizing an ensemble machine learning model based on QML classifiers is presented in this paper. Our approach, with its unique ability to address multidimensional healthcare data, reassures the method's robustness by fusing quantum and classical ML algorithms in a multi-step inferential framework. The marked rise in heart disease and death rates impacts worldwide human health and the global economy. Reducing cardiac morbidity and mortality requires early detection of heart disease. In this research, a hybrid approach utilizes techniques with quantum computing capabilities to tackle complex problems that are not amenable to conventional machine learning algorithms and to minimize computational expenses. The proposed method has been developed in the Raspberry Pi 5 Graphics Processing Unit (GPU) platform and tested on a broad dataset that integrates clinical and imaging data from patients suffering from CHD and healthy controls. Compared to classical machine learning models, the accuracy, sensitivity, F1 score, and specificity of the proposed hybrid QML model used with CHD are manifold higher., Comment: I found a mistake in methodology presentation. Also I have observed more precised results with new dataset. So my research guide ask me to modify the current version

The large-scale deployment of Solidity smart contracts on the Ethereum mainnet has increasingly attracted financially-motivated attackers in recent years. A few now-infamous attacks in Ethereum's history includes DAO attack in 2016 (50 million dollars lost), Parity Wallet hack in 2017 (146 million dollars locked), Beautychain's token BEC in 2018 (900 million dollars market value fell to 0), and NFT gaming blockchain breach in 2022 ($600 million in Ether stolen). This paper presents a comprehensive investigation of the use of large language models (LLMs) and their capabilities in detecting OWASP Top Ten vulnerabilities in Solidity. We introduce a novel, class-balanced, structured, and labeled dataset named VulSmart, which we use to benchmark and compare the performance of open-source LLMs such as CodeLlama, Llama2, CodeT5 and Falcon, alongside closed-source models like GPT-3.5 Turbo and GPT-4o Mini. Our proposed SmartVD framework is rigorously tested against these models through extensive automated and manual evaluations, utilizing BLEU and ROUGE metrics to assess the effectiveness of vulnerability detection in smart contracts. We also explore three distinct prompting strategies-zero-shot, few-shot, and chain-of-thought-to evaluate the multi-class classification and generative capabilities of the SmartVD framework. Our findings reveal that SmartVD outperforms its open-source counterparts and even exceeds the performance of closed-source base models like GPT-3.5 and GPT-4 Mini. After fine-tuning, the closed-source models, GPT-3.5 Turbo and GPT-4o Mini, achieved remarkable performance with 99% accuracy in detecting vulnerabilities, 94% in identifying their types, and 98% in determining severity. Notably, SmartVD performs best with the `chain-of-thought' prompting technique, whereas the fine-tuned closed-source models excel with the `zero-shot' prompting approach.

With the increasing popularity of daily information sharing and acquisition on the Internet, this paper introduces an innovative approach for intent classification in Bangla language, focusing on social media posts where individuals share their thoughts and opinions. The proposed method leverages multimodal data with particular emphasis on authorship identification, aiming to understand the underlying purpose behind textual content, especially in the context of varied user-generated posts on social media. Current methods often face challenges in low-resource languages like Bangla, particularly when author traits intricately link with intent, as observed in social media posts. To address this, we present the Multimodal-based Author Bangla Intent Classification (MABIC) framework, utilizing text and images to gain deeper insights into the conveyed intentions. We have created a dataset named "Uddessho," comprising 3,048 instances sourced from social media. Our methodology comprises two approaches for classifying textual intent and multimodal author intent, incorporating early fusion and late fusion techniques. In our experiments, the unimodal approach achieved an accuracy of 64.53% in interpreting Bangla textual intent. In contrast, our multimodal approach significantly outperformed traditional unimodal methods, achieving an accuracy of 76.19%. This represents an improvement of 11.66%. To our best knowledge, this is the first research work on multimodal-based author intent classification for low-resource Bangla language social media posts., Comment: Accepted for publication in "18th International Conference on Information Technology and Applications (ICITA 2024)"

In the wake of a fabricated explosion image at the Pentagon, an ability to discern real images from fake counterparts has never been more critical. Our study introduces a novel multi-modal approach to detect AI-generated images amidst the proliferation of new generation methods such as Diffusion models. Our method, UGAD, encompasses three key detection steps: First, we transform the RGB images into YCbCr channels and apply an Integral Radial Operation to emphasize salient radial features. Secondly, the Spatial Fourier Extraction operation is used for a spatial shift, utilizing a pre-trained deep learning network for optimal feature extraction. Finally, the deep neural network classification stage processes the data through dense layers using softmax for classification. Our approach significantly enhances the accuracy of differentiating between real and AI-generated images, as evidenced by a 12.64% increase in accuracy and 28.43% increase in AUC compared to existing state-of-the-art methods.

We present a JWST NIRSpec transmission spectrum of the super-Earth exoplanet L 98-59 c. This small (R$_p=1.385\pm0.085$R$_\oplus$, M$_p=2.22\pm0.26$R$_\oplus$), warm (T$_\textrm{eq}=553$K) planet resides in a multi-planet system around a nearby, bright (J = 7.933) M3V star. We find that the transmission spectrum of L 98-59 c is featureless at the precision of our data. We achieve precisions of 22ppm in NIRSpec G395H's NRS1 detector and 36ppm in the NRS2 detector at a resolution R$\sim$200 (30 pixel wide bins). At this level of precision, we are able rule out primordial H$_2$-He atmospheres across a range of cloud pressure levels up to at least $\sim$0.1mbar. By comparison to atmospheric forward models, we also rule out atmospheric metallicities below $\sim$300$\times$ solar at 3$\sigma$ (or equivalently, atmospheric mean molecular weights below $\sim$10~g/mol). We also rule out pure methane atmospheres. The remaining scenarios that are compatible with our data include a planet with no atmosphere at all, or higher mean-molecular weight atmospheres, such as CO$_2$- or H$_2$O-rich atmospheres. This study adds to a growing body of evidence suggesting that planets $\lesssim1.5$R$_\oplus$ lack extended atmospheres., Comment: 21 Pages, 13 Figures; Accepted to AJ

The increasing reliance on machine learning (ML) in computer security, particularly for malware classification, has driven significant advancements. However, the replicability and reproducibility of these results are often overlooked, leading to challenges in verifying research findings. This paper highlights critical pitfalls that undermine the validity of ML research in Android malware detection, focusing on dataset and methodological issues. We comprehensively analyze Android malware detection using two datasets and assess offline and continual learning settings with six widely used ML models. Our study reveals that when properly tuned, simpler baseline methods can often outperform more complex models. To address reproducibility challenges, we propose solutions for improving datasets and methodological practices, enabling fairer model comparisons. Additionally, we open-source our code to facilitate malware analysis, making it extensible for new models and datasets. Our paper aims to support future research in Android malware detection and other security domains, enhancing the reliability and reproducibility of published results.

In the past decade, social media platforms have been used for information dissemination and consumption. While a major portion of the content is posted to promote citizen journalism and public awareness, some content is posted to mislead users. Among different content types such as text, images, and videos, memes (text overlaid on images) are particularly prevalent and can serve as powerful vehicles for propaganda, hate, and humor. In the current literature, there have been efforts to individually detect such content in memes. However, the study of their intersection is very limited. In this study, we explore the intersection between propaganda and hate in memes using a multi-agent LLM-based approach. We extend the propagandistic meme dataset with coarse and fine-grained hate labels. Our finding suggests that there is an association between propaganda and hate in memes. We provide detailed experimental results that can serve as a baseline for future studies. We will make the experimental resources publicly available to the community (https://github.com/firojalam/propaganda-and-hateful-memes)., Comment: propaganda, hate-speech, disinformation, misinformation, fake news, LLMs, GPT-4, multimodality, multimodal LLMs