Search

In this paper we propose a linear-time certifying algorithm for the single-source shortest-path problem capable of verifying graphs with positive, negative, and zero arc weights. Previously proposed linear-time approaches only work for graphs with positive arc weights.

Deep learning-based automated contouring and treatment planning has been proven to improve the efficiency and accuracy of radiotherapy. However, conventional radiotherapy treatment planning process has the automated contouring and treatment planning as separate tasks. Moreover in deep learning (DL), the contouring and dose prediction tasks for automated treatment planning are done independently. In this study, we applied the multi-task learning (MTL) approach in order to seamlessly integrate automated contouring and voxel-based dose prediction tasks, as MTL can leverage common information between the two tasks and be able able to increase the efficiency of the automated tasks. We developed our MTL framework using the two datasets: in-house prostate cancer dataset and the publicly available head and neck cancer dataset, OpenKBP. Compared to the sequential DL contouring and treatment planning tasks, our proposed method using MTL improved the mean absolute difference of dose volume histogram metrics of prostate and head and neck sites by 19.82% and 16.33%, respectively. Our MTL model for automated contouring and dose prediction tasks demonstrated enhanced dose prediction performance while maintaining or sometimes even improving the contouring accuracy. Compared to the baseline automated contouring model with the dice score coefficients of 0.818 for prostate and 0.674 for head and neck datasets, our MTL approach achieved average scores of 0.824 and 0.716 for these datasets, respectively. Our study highlights the potential of the proposed automated contouring and planning using MTL to support the development of efficient and accurate automated treatment planning for radiotherapy.

Misaligned circumbinary disks will produce dust traffic jams during alignment or anti-alignment to the binary orbital plane. We conduct a hydrodynamical simulation of an initially misaligned circumbinary disk undergoing polar alignment with multiple dust species. Due to differential precession between the gas and dust components, multiple dust traffic jams are produced within the disk during polar alignment. The radial locations of the dust traffic jams depend on the Stokes number of the grains, which depends on grain size. We compute the dust temperature structure using post-processing radiative transfer to produce continuum images at cm-wavelengths. Multiple distinct rings emerge in the continuum images, corresponding to the dust traffic jams. The angular resolution of upcoming observations from SKA and ngVLA will be sufficient to detect centimeter-sized grains in protoplanetary disks and resolve the widths of dust traffic jams. Therefore, dust traffic jams resulting from the differential precession of gas and dust in misaligned circumbinary disks will be a prime target for more extended wavelength observations., Comment: 12 pages, 5 figures, accepted for publication in ApJL

Models and observations have demonstrated that Twisted Flux Ropes (TFRs) play a significant role in the structure and eruptive dynamics of active regions. Their role in the dynamics of the quiet Sun atmosphere on has remained elusive, their fundamental relevance emerging mainly from theoretical models (Amari et al. 2015), showing that they form and erupt as a result of flux cancellation. Here HINODE high-resolution photospheric vector magnetic field measurements are integrated with advanced environment reconstruction models: TFRs develop on various scales and are associated with the appearance of mesospots. The developing TFRs contain sufficient free magnetic energy to match the requirements of the recently observed "campfires" discovered by Solar Orbiter in the quiet Sun. The free magnetic energy is found to be large enough to trigger eruptions while the magnetic twist large enough to trigger confined eruptions, heating the atmosphere. TFRs are also connected to larger scale magnetic fields such as supergranulation loops, allowing the generation of Alfv\'en waves at the top of the chromosphere that can propagate along them. High-resolution magnetohydrodynamic simulations, incorporating subsurface dynamo activity at an unprecedented 30 km spatial resolution, confirm that TFRs are ubiquitous products of the permanent small scale dynamo engine that feeds their formation, destabilization, eruption via flux emergence, submergence and cancellation of their chromospheric feet, similar to the dynamics driving large scale eruptive events. Future investigations, especially with the Daniel K. Inouye Solar Telescope (DKIST) and Solar Orbiter will deepen our understanding of TFRs in the context of atmospheric heating.

The Open Radio Access Network (O-RAN) architecture is reshaping the telecommunications landscape by enhancing network flexibility, openness, and intelligence. This paper establishes the requirements, evaluates the design tradeoffs, and introduces a scalable architecture and prototype of an open-source O-RAN experimentation platform within the Aerial Experimentation and Research Platform for Advanced Wireless (AERPAW), an at scale testbed that integrates unmanned aerial vehicles (UAVs) with advanced wireless network technologies, offering experimentation in both outdoor testbed and emulation via a custom digital twin (DT). Through a series of aerial experiments, we evaluate FlexRIC, an open-source RAN Intelligent Controller, within the AERPAW hardware-software platform for network data monitoring, providing valuable insights into the proposed integration and revealing opportunities for leveraging O-RAN to create custom service based optimizations for cellular connected UAVs. We discuss the challenges and potential use cases of this integration and demonstrate the use of a generative artificial intelligence model for generating realistic data based on collected real-world data to support AERPAW's DT., Comment: This article has been accepted for publication in the IEEE Communications Magazine

In the rapidly evolving landscape of 5G technology, safeguarding Radio Frequency (RF) environments against sophisticated intrusions is paramount, especially in dynamic spectrum access and management. This paper presents an enhanced experimental model that integrates a self-attention mechanism with a Recurrent Neural Network (RNN)-based autoencoder for the detection of anomalous spectral activities in 5G networks at the waveform level. Our approach, grounded in time-series analysis, processes in-phase and quadrature (I/Q) samples to identify irregularities that could indicate potential jamming attacks. The model's architecture, augmented with a self-attention layer, extends the capabilities of RNN autoencoders, enabling a more nuanced understanding of temporal dependencies and contextual relationships within the RF spectrum. Utilizing a simulated 5G Radio Access Network (RAN) test-bed constructed with srsRAN 5G and Software Defined Radios (SDRs), we generated a comprehensive stream of data that reflects real-world RF spectrum conditions and attack scenarios. The model is trained to reconstruct standard signal behavior, establishing a normative baseline against which deviations, indicative of security threats, are identified. The proposed architecture is designed to balance between detection precision and computational efficiency, so the LSTM network, enriched with self-attention, continues to optimize for minimal execution latency and power consumption. Conducted on a real-world SDR-based testbed, our results demonstrate the model's improved performance and accuracy in threat detection. Keywords: self-attention, real-time intrusion detection, RNN autoencoder, Transformer architecture, LSTM, time series anomaly detection, 5G Security, spectrum access security., Comment: This article has been accepted for publication in WiOpt 2024

We study second-order electromagnetic perturbations in the Schwarzschild background and derive the effective source terms for Regge-Wheeler equation which are quadratic in first-order gravitational and electromagnetic perturbations. In addition to the induced mixed quadratic modes, we find that linear gravitational modes are also excited, with amplitudes dependent on the electromagnetic potential. A toy model involving a Dirac delta function potential demonstrates mixing of linear gravitational and electromagnetic perturbations with frequencies \( \omega^{(1)} \) and \( \Omega^{(1)} \), resulting in the second-order QNM mixing in the electromagnetic field at \( \Omega^{(2)} =\Omega^{(1)} + \omega^{(1)} \). This complements prior work in [1] on the second-order gravitational perturbation mixing and highlights potential applications in multi-messenger astrophysics for systems observed by LIGO and upcoming LISA. We also study first-order perturbations due to a point charge and show it could be reduced to a one-dimensional path integral. Within the toy model, we investigate the first-order electromagnetic perturbation due to a radially free-falling single charge \( q \) and radial dipole moment \( p = q \eta \), employing semi-analytical and numerical methods. For the dipole case, we show that the QNM perturbation is excited with a nearly constant amplitude. Future work will focus on incorporating mixing in more realistic potentials and exploring numerical approach in the context of rotating spacetimes., Comment: 20 pages, 25 figures

Fact verification on tabular evidence incentivises the use of symbolic reasoning models where a logical form is constructed (e.g. a LISP-style program), providing greater verifiability than fully neural approaches. However, these systems typically rely on well-formed tables, restricting their use in many scenarios. An emerging symbolic reasoning paradigm for textual evidence focuses on natural logic inference, which constructs proofs by modelling set-theoretic relations between a claim and its evidence in natural language. This approach provides flexibility and transparency but is less compatible with tabular evidence since the relations do not extend to arithmetic functions. We propose a set-theoretic interpretation of numerals and arithmetic functions in the context of natural logic, enabling the integration of arithmetic expressions in deterministic proofs. We leverage large language models to generate arithmetic expressions by generating questions about salient parts of a claim which are answered by executing appropriate functions on tables. In a few-shot setting on FEVEROUS, we achieve an accuracy of 71.4, outperforming both fully neural and symbolic reasoning models by 3.4 points. When evaluated on TabFact without any further training, our method remains competitive with an accuracy lead of 0.5 points., Comment: Accepted to TACL. This is a slightly extended version

The Automated Verification of Textual Claims (AVeriTeC) shared task asks participants to retrieve evidence and predict veracity for real-world claims checked by fact-checkers. Evidence can be found either via a search engine, or via a knowledge store provided by the organisers. Submissions are evaluated using AVeriTeC score, which considers a claim to be accurately verified if and only if both the verdict is correct and retrieved evidence is considered to meet a certain quality threshold. The shared task received 21 submissions, 18 of which surpassed our baseline. The winning team was TUDA_MAI with an AVeriTeC score of 63%. In this paper we describe the shared task, present the full results, and highlight key takeaways from the shared task.

Dyslexia is one of the most common learning disorders, often characterized by distinct features in handwriting. Early detection is essential for effective intervention. In this paper, we propose an explainable AI (XAI) framework for dyslexia detection through handwriting analysis, utilizing transfer learning and transformer-based models. Our approach surpasses state-of-the-art methods, achieving a test accuracy of 0.9958, while ensuring model interpretability through Grad-CAM visualizations that highlight the critical handwriting features influencing model decisions. The main contributions of this work include the integration of XAI for enhanced interpretability, adaptation to diverse languages and writing systems, and demonstration of the method's global applicability. This framework not only improves diagnostic accuracy but also fosters trust and understanding among educators, clinicians, and parents, supporting earlier diagnoses and the development of personalized educational strategies., Comment: 4 pages, 4 figures, JAC-ECC Conference

We investigate the possibility of electromagnetic fields leaving imprints on gravitational wave (GW) signals from Neutron Star-Black hole (NSBH) mergers, specifically in the context of extreme mass ratio inspirals (EMRIs). Using black hole perturbation theory (BHPT) in the context of a minimally coupled Einstein-Maxwell system, we demonstrate that electromagnetic quasi normal modes(QNMs) can excite gravitational QNMs with frequencies that are linear or quadratic in the electromagnetic QNMs, at first level of mixing. Moreover, We then study the electromagnetism-gravity coupling by approximating the Regge-Wheeler and Zerilli potentials with Dirac delta functions. In this example, we examine gravitational perturbations induced by the electromagnetic field of an ideal dipole radially free fall towards the blackhole, building on calculations from a companion paper [1]. Our results show that both linear and quadratic electromagnetic QNMs appear in gravitational perturbations. In addition, linear gravitational QNMs are also excited due to the electromagnetic source, with their amplitudes depending on the details of the electromagnetic and gravitational potentials, analogous to gravitational mode mixing analysis. Furthermore, at late stages, gravitational perturbations might exhibit polynomial tails induced by electromagnetic perturbations. This article sets the stage for future numerical investigations aimed at identifying such modes in various scenarios., Comment: 27 pages, and 3 figures

Thousands of individuals succumb annually to leukemia alone. This study explores the application of image processing and deep learning techniques for detecting Acute Lymphoblastic Leukemia (ALL), a severe form of blood cancer responsible for numerous annual fatalities. As artificial intelligence technologies advance, the research investigates the reliability of these methods in real-world scenarios. The study focuses on recent developments in ALL detection, particularly using the latest YOLO series models, to distinguish between malignant and benign white blood cells and to identify different stages of ALL, including early stages. Additionally, the models are capable of detecting hematogones, which are often misclassified as ALL. By utilizing advanced deep learning models like YOLOv8 and YOLOv11, the study achieves high accuracy rates reaching 98.8%, demonstrating the effectiveness of these algorithms across multiple datasets and various real-world situations., Comment: 4 pages, 6 figures, 3 tables

Vehicle-to-everything (V2X) networks support a variety of safety, entertainment, and commercial applications. This is realized by applying the principles of the Internet of Vehicles (IoV) to facilitate connectivity among vehicles and between vehicles and roadside units (RSUs). Network congestion management is essential for IoVs and it represents a significant concern due to its impact on improving the efficiency of transportation systems and providing reliable communication among vehicles for the timely delivery of safety-critical packets. This paper introduces a framework for proactive congestion management for IoV networks. We generate congestion scenarios and a data set to predict the congestion using LSTM. We present the framework and the packet congestion dataset. Simulation results using SUMO with NS3 demonstrate the effectiveness of the framework for forecasting IoV network congestion and clustering/prioritizing packets employing recurrent neural networks., Comment: This article has been accepted for publication in the IEEE VTC Fall 2024

The rapid evolution of 5G and beyond has advanced space-air-terrestrial networks, with unmanned aerial vehicles (UAVs) offering enhanced coverage, flexible configurations, and cost efficiency. However, deploying UAV-based systems presents challenges including varying propagation conditions and hardware limitations. While simulators and theoretical models have been developed, real-world experimentation is critically important to validate the research. Digital twins, virtual replicas of physical systems, enable emulation that bridge theory and practice. This paper presents our experimental results from AERPAW's digital twin, showcasing its ability to simulate UAV communication scenarios and providing insights into system performance and reliability., Comment: This article has been accepted for publication in the IEEE VTC Fall 2024--UAV Communication and Experimentation Workshop

With the rise of 5G and open radio access networks (O-RAN), there is a growing demand for customizable experimental platforms dedicated to security testing, as existing testbeds do not prioritize this area. Traditional, hardware-dependent testing methods pose challenges for smaller companies and research institutions. The growing wireless threat landscape highlights the critical need for proactive security testing, as 5G and O-RAN deployments are appealing targets for cybercriminals. To address these challenges, this article introduces the Soft Tester UE (soft T-UE), a software-defined test equipment designed to evaluate the security of 5G and O-RAN deployments via the Uu air interface between the user equipment (UE) and the network. The outcome is to deliver a free, open-source, and expandable test instrument to address the need for both standardized and customizable automated security testing. By extending beyond traditional security metrics, the soft T-UE promotes the development of new security measures and enhances the capability to anticipate and mitigate potential security breaches. The tool's automated testing capabilities are demonstrated through a scenario where the Radio Access Network (RAN) under test is evaluated when it receives fuzzed data when initiating a connection with an UE., Comment: This article has been accepted for publication in the IEEE VTC Fall 2024--RitiRAN Workshop

Starshade is one of the technologies that will enable the observation and characterization of small planets around nearby stars through direct imaging. The Starshade Exoplanetary Data Challenge (SEDC) was designed to validate starshade-imaging's noise budget and evaluate the capabilities of image-processing techniques, by inviting community participating teams to analyze >1000 simulated images of hypothetical exoplanetary systems observed through a starshade. Because the starshade would suppress the starlight so well, the dominant noise source and the main challenge for the planet detection becomes the exozodiacal disks and their structures. In this paper, we summarize the techniques used by the participating teams and compare their findings with the truth. With an independent component analysis to remove the background, about 70% of the inner planets (close to the inner working angle) have been detected and ~40% of the outer planet (fainter than the inner counterparts) have been identified. Planet detection becomes more difficult in the cases of higher disk inclination, as the false negative and false positive counts increase. Interestingly, we found little difference in the planet detection ability between 1e-10 and 1e-9 instrument contrast, confirming that the dominant limitations are from the astrophysical background and not due to the performance of the starshade. Finally, we find that a non-parametric background calibration scheme, such as the independent component analysis reported here, results in a mean residual of 10% the background brightness. This background estimation error leads to substantial false positives and negatives and systematic bias in the planet flux estimation, and should be included in the estimation of the planet detection signal-to-noise ratio for imaging using a starshade and also a coronagraph that delivers exozodi-limited imaging., Comment: Accepted for publication in Journal of Astronomical Telescopes, Instruments, and Systems

Lymphoma diagnosis, particularly distinguishing between subtypes, is critical for effective treatment but remains challenging due to the subtle morphological differences in histopathological images. This study presents a novel hybrid deep learning framework that combines DenseNet201 for feature extraction with a Dense Neural Network (DNN) for classification, optimized using the Harris Hawks Optimization (HHO) algorithm. The model was trained on a dataset of 15,000 biopsy images, spanning three lymphoma subtypes: Chronic Lymphocytic Leukemia (CLL), Follicular Lymphoma (FL), and Mantle Cell Lymphoma (MCL). Our approach achieved a testing accuracy of 99.33\%, demonstrating significant improvements in both accuracy and model interpretability. Comprehensive evaluation using precision, recall, F1-score, and ROC-AUC underscores the model's robustness and potential for clinical adoption. This framework offers a scalable solution for improving diagnostic accuracy and efficiency in oncology., Comment: 6 pages, 5 figures, 4 tables, IEEE ICCA

We investigate the formation of dust traffic jams in polar-aligning circumbinary discs. In our first paper, we found as the circumbinary disc evolves towards a polar configuration perpendicular to the binary orbital plane, the differential precession between the gas and dust components leads to multiple dust traffic jams. These dust traffic jams evolve to form a coherent dust ring. In part two, we use 3D smoothed particle hydrodynamical simulations of gas and dust to model an initially highly misaligned circumbinary disc around the 99 Herculis (99 Her) binary system. Our results reveal that the formation of these dust rings is observed across various disc parameters, including the disc aspect ratio, viscosity, surface density power law index, and temperature power law index. The dust traffic jams are long-lived and persist even when the disc is fully aligned polar. The midplane dust-to-gas ratio within the rings can surpass unity, which may be a favourable environment for planetesimal formation. Using 2D inviscid shearing box calculations with parameters from our 3D simulations, we find streaming instability modes with significant growth rates. The streaming instability growth timescale is less than the tilt oscillation timescale during the alignment process. Therefore, the dust ring will survive once the gas disc aligns polar, suggesting that the streaming instability may aid in forming polar planets around 99 Her., Comment: 13 pages, 11 figures, Accepted for publication in MNRAS

The recent development of fact verification systems with natural logic has enhanced their explainability by aligning claims with evidence through set-theoretic operators, providing faithful justifications. Despite these advancements, such systems often rely on a large amount of training data annotated with natural logic. To address this issue, we propose a zero-shot method that utilizes the generalization capabilities of instruction-tuned large language models. To comprehensively assess the zero-shot capabilities of our method and other fact verification systems, we evaluate all models on both artificial and real-world claims, including multilingual datasets. We also compare our method against other fact verification systems in two setups. First, in the zero-shot generalization setup, we demonstrate that our approach outperforms other systems that were not specifically trained on natural logic data, achieving an average accuracy improvement of 8.96 points over the best-performing baseline. Second, in the zero-shot transfer setup, we show that current systems trained on natural logic data do not generalize well to other domains, and our method outperforms these systems across all datasets with real-world claims., Comment: Accepted to EMNLP 2024

The scarcity of subspecialist medical expertise, particularly in rare, complex and life-threatening diseases, poses a significant challenge for healthcare delivery. This issue is particularly acute in cardiology where timely, accurate management determines outcomes. We explored the potential of AMIE (Articulate Medical Intelligence Explorer), a large language model (LLM)-based experimental AI system optimized for diagnostic dialogue, to potentially augment and support clinical decision-making in this challenging context. We curated a real-world dataset of 204 complex cases from a subspecialist cardiology practice, including results for electrocardiograms, echocardiograms, cardiac MRI, genetic tests, and cardiopulmonary stress tests. We developed a ten-domain evaluation rubric used by subspecialists to evaluate the quality of diagnosis and clinical management plans produced by general cardiologists or AMIE, the latter enhanced with web-search and self-critique capabilities. AMIE was rated superior to general cardiologists for 5 of the 10 domains (with preference ranging from 9% to 20%), and equivalent for the rest. Access to AMIE's response improved cardiologists' overall response quality in 63.7% of cases while lowering quality in just 3.4%. Cardiologists' responses with access to AMIE were superior to cardiologist responses without access to AMIE for all 10 domains. Qualitative examinations suggest AMIE and general cardiologist could complement each other, with AMIE thorough and sensitive, while general cardiologist concise and specific. Overall, our results suggest that specialized medical LLMs have the potential to augment general cardiologists' capabilities by bridging gaps in subspecialty expertise, though further research and validation are essential for wide clinical utility.

This paper presents an Arabic Alphabet Sign Language recognition approach, using deep learning methods in conjunction with transfer learning and transformer-based models. We study the performance of the different variants on two publicly available datasets, namely ArSL2018 and AASL. This task will make full use of state-of-the-art CNN architectures like ResNet50, MobileNetV2, and EfficientNetB7, and the latest transformer models such as Google ViT and Microsoft Swin Transformer. These pre-trained models have been fine-tuned on the above datasets in an attempt to capture some unique features of Arabic sign language motions. Experimental results present evidence that the suggested methodology can receive a high recognition accuracy, by up to 99.6\% and 99.43\% on ArSL2018 and AASL, respectively. That is far beyond the previously reported state-of-the-art approaches. This performance opens up even more avenues for communication that may be more accessible to Arabic-speaking deaf and hard-of-hearing, and thus encourages an inclusive society., Comment: 6 pages, 8 figures

Early diagnosis and intervention for Autism Spectrum Disorder (ASD) has been shown to significantly improve the quality of life of autistic individuals. However, diagnostics methods for ASD rely on assessments based on clinical presentation that are prone to bias and can be challenging to arrive at an early diagnosis. There is a need for objective biomarkers of ASD which can help improve diagnostic accuracy. Deep learning (DL) has achieved outstanding performance in diagnosing diseases and conditions from medical imaging data. Extensive research has been conducted on creating models that classify ASD using resting-state functional Magnetic Resonance Imaging (fMRI) data. However, existing models lack interpretability. This research aims to improve the accuracy and interpretability of ASD diagnosis by creating a DL model that can not only accurately classify ASD but also provide explainable insights into its working. The dataset used is a preprocessed version of the Autism Brain Imaging Data Exchange (ABIDE) with 884 samples. Our findings show a model that can accurately classify ASD and highlight critical brain regions differing between ASD and typical controls, with potential implications for early diagnosis and understanding of the neural basis of ASD. These findings are validated by studies in the literature that use different datasets and modalities, confirming that the model actually learned characteristics of ASD and not just the dataset. This study advances the field of explainable AI in medical imaging by providing a robust and interpretable model, thereby contributing to a future with objective and reliable ASD diagnostics.

Intrinsic motivation (IM) and reward shaping are common methods for guiding the exploration of reinforcement learning (RL) agents by adding pseudo-rewards. Designing these rewards is challenging, however, and they can counter-intuitively harm performance. To address this, we characterize them as reward shaping in Bayes-Adaptive Markov Decision Processes (BAMDPs), which formalizes the value of exploration by formulating the RL process as updating a prior over possible MDPs through experience. RL algorithms can be viewed as BAMDP policies; instead of attempting to find optimal algorithms by solving BAMDPs directly, we use it at a theoretical framework for understanding how pseudo-rewards guide suboptimal algorithms. By decomposing BAMDP state value into the value of the information collected plus the prior value of the physical state, we show how psuedo-rewards can help by compensating for RL algorithms' misestimation of these two terms, yielding a new typology of IM and reward shaping approaches. We carefully extend the potential-based shaping theorem to BAMDPs to prove that when pseudo-rewards are BAMDP Potential-based shaping Functions (BAMPFs), they preserve optimal, or approximately optimal, behavior of RL algorithms; otherwise, they can corrupt even optimal learners. We finally give guidance on how to design or convert existing pseudo-rewards to BAMPFs by expressing assumptions about the environment as potential functions on BAMDP states.

In the realm of cyber-security, detecting Advanced Persistent Threats (APTs) remains a formidable challenge due to their stealthy and sophisticated nature. This research paper presents an innovative approach that leverages Convolutional Neural Networks (CNNs) with a 2D baseline model, enhanced by the cutting-edge Cat Swarm Optimization (CSO) algorithm, to significantly improve APT detection accuracy. By seamlessly integrating the 2D-CNN baseline model with CSO, we unlock the potential for unprecedented accuracy and efficiency in APT detection. The results unveil an impressive accuracy score of $98.4\%$, marking a significant enhancement in APT detection across various attack stages, illuminating a path forward in combating these relentless and sophisticated threats., Comment: 6 pages, 5 figures

Distilling conversational skills into Small Language Models (SLMs) with approximately 1 billion parameters presents significant challenges. Firstly, SLMs have limited capacity in their model parameters to learn extensive knowledge compared to larger models. Secondly, high-quality conversational datasets are often scarce, small, and domain-specific. Addressing these challenges, we introduce a novel data distillation framework named CoDi (short for Conversational Distillation, pronounced "Cody"), allowing us to synthesize large-scale, assistant-style datasets in a steerable and diverse manner. Specifically, while our framework is task agnostic at its core, we explore and evaluate the potential of CoDi on the task of conversational grounded reasoning for question answering. This is a typical on-device scenario for specialist SLMs, allowing for open-domain model responses, without requiring the model to "memorize" world knowledge in its limited weights. Our evaluations show that SLMs trained with CoDi-synthesized data achieve performance comparable to models trained on human-annotated data in standard metrics. Additionally, when using our framework to generate larger datasets from web data, our models surpass larger, instruction-tuned models in zero-shot conversational grounded reasoning tasks., Comment: 13 pages

Existing multimodal-based human action recognition approaches are either computationally expensive, which limits their applicability in real-time scenarios, or fail to exploit the spatial temporal information of multiple data modalities. In this work, we present an efficient pose-driven attention-guided multimodal network (EPAM-Net) for action recognition in videos. Specifically, we adapted X3D networks for both RGB and pose streams to capture spatio-temporal features from RGB videos and their skeleton sequences. Then skeleton features are utilized to help the visual network stream focusing on key frames and their salient spatial regions using a spatial temporal attention block. Finally, the scores of the two streams of the proposed network are fused for final classification. The experimental results show that our method achieves competitive performance on NTU-D 60 and NTU RGB-D 120 benchmark datasets. Moreover, our model provides a 6.2--9.9x reduction in FLOPs (floating-point operation, in number of multiply-adds) and a 9--9.6x reduction in the number of network parameters. The code will be available at https://github.com/ahmed-nady/Multimodal-Action-Recognition.