Search

Digital twins (DTs) of wireless environments can be utilized to predict the propagation channel and reduce the overhead of required to estimate the channel statistics. However, direct channel prediction requires data-intensive calibration of the DT to capture the environment properties relevant for propagation of electromagnetic signals. We introduce a framework that starts from a satellite image of the environment to produce an uncalibrated DT, which has no or imprecise information about the materials and their electromagnetic properties. The key idea is to use the uncalibrated DT to implicitly provide a geometric prior for the environment. This is utilized to inform a Gaussian process (GP), which permits the use of few channel measurements to attain an accurate prediction of the channel statistics. Additionally, the framework is able to quantify the uncertainty in channel statistics prediction and select rate in ultra-reliable low-latency communication (URLLC) that complies with statistical guarantees. The efficacy of the proposed geometry-informed GP is validated using experimental data obtained through a measurement campaign. Furthermore, the proposed prediction framework is shown to provide significant improvements compared to the benchmarks where i) direct channel statistics prediction is obtained using an uncalibrated DT and (ii) the GP predicts channel statistics using information about the location., Comment: 6 pages, 10 figures

Physics Informed Machine Learning has emerged as a popular approach in modelling and simulation for digital twins to generate accurate models of processes and behaviours of real-world systems. However, despite their success in generating accurate and reliable models, the existing methods either use simple regularizations in loss functions to offer limited physics integration or are too specific in architectural definitions to be generalized to a wide variety of physical systems. This paper presents a generic approach based on a novel physics-encoded residual neural network architecture to combine data-driven and physics-based analytical models to address these limitations. Our method combines physics blocks as mathematical operators from physics-based models with learning blocks comprising feed-forward layers. Intermediate residual blocks are incorporated for stable gradient flow as they train on physical system observation data. This way, the model learns to comply with the geometric and kinematic aspects of the physical system. Compared to conventional neural network-based methods, our method improves generalizability with substantially low data requirements and model complexity in terms of parameters, especially in scenarios where prior physics knowledge is either elementary or incomplete. We investigate our approach in two application domains. The first is a basic robotic motion model using Euler Lagrangian equations of motion as physics prior. The second application is a complex scenario of a steering model for a self-driving vehicle in a simulation. In both applications, our method outperforms both conventional neural network based approaches as-well as state-of-the-art Physics Informed Machine Learning methods.

Cellular vehicle-to-everything (C-V2X) networks provide a promising solution to improve road safety and traffic efficiency. One key challenge in such systems lies in meeting different quality-of-service (QoS) requirements of coexisting vehicular communication links, particularly under imperfect channel state information (CSI) conditions caused by the highly dynamic environment. In this paper, a novel analytical framework for examining the resilience of C-V2X networks in face of imperfect CSI is proposed. In this framework, the adaptation phase of the C-V2X network is studied, in which an adaptation power scheme is employed and the probability distribution function (PDF) of the imperfect CSI is estimated. Then, the resilience of C-V2X networks is studied through two principal dimensions: remediation capability and adaptation performance, both of which are defined, quantified, and analyzed for the first time. Particularly, an upper bound on the estimation's mean square error (MSE) is explicitly derived to capture the C-V2X's remediation capability, and a novel metric named hazard rate (HR) is exploited to evaluate the C-V2X's adaptation performance. Afterwards, the impact of the adaptation power scheme on the C-V2X's resilience is examined, revealing a tradeoff between the C-V2X's remediation capability and adaptation performance. Simulation results validate the framework's superiority in capturing the interplay between adaptation and remediation, as well as the effectiveness of the two proposed metrics in guiding the design of the adaptation power scheme to enhance the system's resilience., Comment: Submitted to ICC 2025

The effects of a dispersed particulate phase on the onset of Rayleigh-B\'enard convection in a fluid layer is studied theoretically by means of a two-fluid Eulerian modelization. The particles are non-Brownian, spherical, with inertia and heat capacity, and they interact with the surrounding fluid mechanically and thermally. We study both the cases of particles denser and lighter than the fluid that are injected uniformly at the system's horizontal boundaries with their settling terminal velocity and prescribed temperatures. The performed linear stability analysis shows that the onset of thermal convection is stationary, i.e., the system undergoes a pitchfork bifurcation as in the classical single-phase RB problem. Remarkably, the mechanical coupling due to the particle motion always stabilizes the system, increasing the critical Rayleigh number ($Ra_c$) of the convective onset. Furthermore, the particle to fluid heat capacity ratio provides an additional stabilizing mechanism, that we explore in full by addressing both the asymptotic limits of negligible and overwhelming particle thermal inertia. The overall resulting stabilization effect on $Ra_c$ is significant: for a particulate volume fraction of 0.1% it reaches up to a factor 30 for the case of the lightest particle density (i.e. bubbles) and 60 for the heaviest one. The present work extends the analysis performed by Prakhar & Prosperetti (Phys. Rev. Fluids 6, 083901, 2021) where the thermo-mechanical stabilization effect has been first demonstrated for highly dense particles. Here, by including the effect of the added-mass force in the model system, we succeed in exploring the full range of particle densities. Finally, we critically discuss the role of the particle injection boundary conditions which are adopted in this study and how their modification may lead to different dynamics, that deserve to be studied in the future., Comment: 28 pages, 14 figures

We present an innovative approach for solving mathematical problems in Bengali, developed for the DL Sprint 3.0 BUET CSE Fest 2024 Competition. Our method uses advanced deep learning models, notably the Qwen 2.5 series, with improvements made through prompt engineering, model quantization, and Tool Integrated Reasoning (TIR) to handle complex calculations. Initially, we explored various model architectures, including fine-tuned Mistral and quantized Qwen models, refining them with translation techniques, Retrieval-Augmented Generation (RAG), and custom dataset curation. Manual hyperparameter tuning optimized parameters like temperature and top-p to enhance model adaptability and accuracy. Removal of RAG and parameter adjustments further improved robustness. Our approach highlights the potential of advanced NLP techniques in solving Bengali mathematical problems.

Recent developments in computer graphics, machine learning, and sensor technologies enable numerous opportunities for extended reality (XR) setups for everyday life, from skills training to entertainment. With large corporations offering consumer-grade head-mounted displays (HMDs) in an affordable way, it is likely that XR will become pervasive, and HMDs will develop as personal devices like smartphones and tablets. However, having intelligent spaces and naturalistic interactions in XR is as important as technological advances so that users grow their engagement in virtual and augmented spaces. To this end, large language model (LLM)--powered non-player characters (NPCs) with speech-to-text (STT) and text-to-speech (TTS) models bring significant advantages over conventional or pre-scripted NPCs for facilitating more natural conversational user interfaces (CUIs) in XR. In this paper, we provide the community with an open-source, customizable, extensible, and privacy-aware Unity package, CUIfy, that facilitates speech-based NPC-user interaction with various LLMs, STT, and TTS models. Our package also supports multiple LLM-powered NPCs per environment and minimizes the latency between different computational models through streaming to achieve usable interactions between users and NPCs. We publish our source code in the following repository: https://gitlab.lrz.de/hctl/cuify, Comment: This work has been submitted to the IEEE for possible publication

Field work is an essential component not just for organismal biology, but also for the expanding umbrella of disciplines which have turned their attention towards living systems. Observing organisms in naturalistic contexts is a critical component of discovery; however, conducting field research can be a massive barrier for scientists who do not have experience working with organisms in a naturalistic context under challenging field conditions. Here we propose 8 critical steps for organizing and executing interdisciplinary curiosity-driven field research, drawing on the insights from The in Situ Jungle Biomechanics Lab (JBL). The JBL program is a field research course that helps early-career scientists gain experience in organizing and conducting interdisciplinary field research. JBL uses a curiosity-driven approach to field science education by encouraging early-career researchers to explore scientific questions in the Peruvian Amazon with a non-prescriptive approach to research output. We achieve an inclusive research space by bringing scientists from across disciplines together, with local communities to collaborate and spark new questions and ideas. To stoke curiosity, the JBL imparts a naturalist tradition set forth by organismal biologists of the 20th century who have extolled the merits of observing the natural world as a form of scientific exploration.

We present a search for long-lived particles (LLPs) produced from kaon decay that decay to two muons inside the ICARUS neutrino detector. This channel would be a signal of hidden sector models that can address outstanding issues in particle physics such as the strong CP problem and the microphysical origin of dark matter. The search is performed with data collected in the Neutrinos at the Main Injector (NuMI) beam at Fermilab corresponding to $2.41\times 10^{20}$ protons-on-target. No new physics signal is observed, and we set world-leading limits on heavy QCD axions, as well as for the Higgs portal scalar among dedicated searches. Limits are also presented in a model-independent way applicable to any new physics model predicting the process $K\to \pi+S(\to\mu\mu)$, for a long-lived particle S. This result is the first search for new physics performed with the ICARUS detector at Fermilab. It paves the way for the future program of long-lived particle searches at ICARUS.

Deep Neural Networks (DNNs) suffer from significant retraining costs when adapting to evolving requirements. Modularizing DNNs offers the promise of improving their reusability. Previous work has proposed techniques to decompose DNN models into modules both during and after training. However, these strategies yield several shortcomings, including significant weight overlaps and accuracy losses across modules, restricted focus on convolutional layers only, and added complexity and training time by introducing auxiliary masks to control modularity. In this work, we propose MODA, an activation-driven modular training approach. MODA promotes inherent modularity within a DNN model by directly regulating the activation outputs of its layers based on three modular objectives: intra-class affinity, inter-class dispersion, and compactness. MODA is evaluated using three well-known DNN models and three datasets with varying sizes. This evaluation indicates that, compared to the existing state-of-the-art, using MODA yields several advantages: (1) MODA accomplishes modularization with 29% less training time; (2) the resultant modules generated by MODA comprise 2.4x fewer weights and 3.5x less weight overlap while (3) preserving the original model's accuracy without additional fine-tuning; in module replacement scenarios, (4) MODA improves the accuracy of a target class by 12% on average while ensuring minimal impact on the accuracy of other classes.

Social dilemmas where the good of a group is at odds with individual interests are usually considered as static -- the dilemma does not change over time. In the COVID-19 pandemic, social dilemmas occurred in the mitigation of epidemic spread: Should I reduce my contacts or wear a mask to protect others? In the context of respiratory diseases, which are predominantly spreading during the winter months, some of these situations re-occur seasonally. We couple a game theoretical model, where individuals can adjust their behavior, to an epidemiological model with seasonal forcing. We find that social dilemmas can occur annually and that behavioral reactions to them can either decrease or increase the peaks of infections in a population. Our work has not only implications for seasonal infectious diseases, but also more generally for oscillatory social dilemmas: A complex interdependence between behavior and external dynamics emerges. To be effective and to exploit behavioral dynamics, intervention measures to mitigate re-occuring social dilemmas have to be timed carefully.

Treatment selection in breast cancer is guided by molecular subtypes and clinical characteristics. Recurrence risk assessment plays a crucial role in personalizing treatment. Current methods, including genomic assays, have limited accuracy and clinical utility, leading to suboptimal decisions for many patients. We developed a test for breast cancer patient stratification based on digital pathology and clinical characteristics using novel AI methods. Specifically, we utilized a vision transformer-based pan-cancer foundation model trained with self-supervised learning to extract features from digitized H&E-stained slides. These features were integrated with clinical data to form a multi-modal AI test predicting cancer recurrence and death. The test was developed and evaluated using data from a total of 8,161 breast cancer patients across 15 cohorts originating from seven countries. Of these, 3,502 patients from five cohorts were used exclusively for evaluation, while the remaining patients were used for training. Our test accurately predicted our primary endpoint, disease-free interval, in the five external cohorts (C-index: 0.71 [0.68-0.75], HR: 3.63 [3.02-4.37, p<0.01]). In a direct comparison (N=858), the AI test was more accurate than Oncotype DX, the standard-of-care 21-gene assay, with a C-index of 0.67 [0.61-0.74] versus 0.61 [0.49-0.73], respectively. Additionally, the AI test added independent information to Oncotype DX in a multivariate analysis (HR: 3.11 [1.91-5.09, p<0.01)]). The test demonstrated robust accuracy across all major breast cancer subtypes, including TNBC (C-index: 0.71 [0.62-0.81], HR: 3.81 [2.35-6.17, p=0.02]), where no diagnostic tools are currently recommended by clinical guidelines. These results suggest that our AI test can improve accuracy, extend applicability to a wider range of patients, and enhance access to treatment selection tools.

Dual-arm manipulation is an area of growing interest in the robotics community. Enabling robots to perform tasks that require the coordinated use of two arms, is essential for complex manipulation tasks such as handling large objects, assembling components, and performing human-like interactions. However, achieving effective dual-arm manipulation is challenging due to the need for precise coordination, dynamic adaptability, and the ability to manage interaction forces between the arms and the objects being manipulated. We propose a novel pipeline that combines the advantages of policy learning based on environment feedback and gradient-based optimization to learn controller gains required for the control outputs. This allows the robotic system to dynamically modulate its impedance in response to task demands, ensuring stability and dexterity in dual-arm operations. We evaluate our pipeline on a trajectory-tracking task involving a variety of large, complex objects with different masses and geometries. The performance is then compared to three other established methods for controlling dual-arm robots, demonstrating superior results.

In this work, we approach certain black hole issues, including remnants, by providing a statistical description based on the weak gravity conjecture in the swampland program. Inspired by the Pauli exclusion principal in the context of the Fermi sphere, we derive an inequality which can be exploited to verify the instability manifestation of the black holes via a characteristic function. For several species, we show that this function is in agreement with the weak gravity swampland conjecture. Then, we deal with the cutoff issue as an interval estimation problem by putting an upper bound on the black hole mass scale matching with certain results reported in the literature. Using the developed formalism for the proposed instability scenarios, we provide a suppression mechanism to the remnant production rate. Furthermore, we reconsider the stability study of the Reissner-Nordstrom black holes. Among others, we show that the proposed instabilities prohibit naked singularity behaviors, Comment: Latex, 15 pages

Aquatic mammals, such as pinnipeds, utilize their whiskers to detect and discriminate objects and analyze water movements, inspiring the development of robotic whiskers for sensing contacts, surfaces, and water flows. We present the design and application of underwater whisker sensors based on Fiber Bragg Grating (FBG) technology. These passive whiskers are mounted along the robot$'$s exterior to sense its surroundings through light, non-intrusive contacts. For contact tracking, we employ a sim-to-real learning framework, which involves extensive data collection in simulation followed by a sim-to-real calibration process to transfer the model trained in simulation to the real world. Experiments with whiskers immersed in water indicate that our approach can track contact points with an accuracy of $<2$ mm, without requiring precise robot proprioception. We demonstrate that the approach also generalizes to unseen objects.

For the validation and verification of automotive radars, datasets of realistic traffic scenarios are required, which, how ever, are laborious to acquire. In this paper, we introduce radar scene synthesis using GANs as an alternative to the real dataset acquisition and simulation-based approaches. We train a PointNet++ based GAN model to generate realistic radar point cloud scenes and use a binary classifier to evaluate the performance of scenes generated using this model against a test set of real scenes. We demonstrate that our GAN model achieves similar performance (~87%) to the real scenes test set., Comment: ICMIM 2024; 7th IEEE MTT Conference

Data acquisition in array signal processing (ASP) is costly, as high angular and range resolutions require large antenna apertures and wide frequency bandwidths. Data requirements grow multiplicatively with viewpoints and frequencies, increasing collection burdens. Implicit Neural Representations (INRs)--neural network models of 3D scenes--offer compact, continuous representations with minimal data, interpolating to unseen viewpoints, potentially reducing sampling costs in ASP. We propose the Radon Implicit Field Transform (RIFT), combining a radar forward model (Generalized Radon Transform, GRT) with an INR-based scene representation learned from radar signals. This method extends to other ASP problems by replacing the GRT with appropriate algorithms. In experiments, we synthesize radar data using the GRT and train the INR model by minimizing radar signal reconstruction error. We render the scene using the trained INR and evaluate it against ground truth. We introduce new error metrics: phase-Root Mean Square Error (p-RMSE) and magnitude-Structural Similarity Index Measure (m-SSIM). Compared to traditional scene models, our RIFT model achieves up to 188% improvement in scene reconstruction with only 10% of the data. Using the same amount of data, RIFT achieves 3x better reconstruction and shows a 10% improvement when generalizing to unseen viewpoints., Comment: A version of this work is under review as a submission to ICLR 2025 Conference

In this paper we assess how well users know biometric authentication methods, how they perceive them, and if they have misconceptions about them. We present the results of an online survey that we conducted in two rounds (2019, N=57; and 2023, N=47) to understand the impact of the increasing availability of biometrics on their use and perception. The survey covered participants' general understanding of physiological and behavioral biometrics and their perceived usability and security. While most participants were able to name examples and stated that they use biometrics in their daily lives, they still had difficulties explaining the concepts behind them. We shed light on participants' misconceptions, their coping strategies with authentication failures and potential attacks, as well as their perception of the usability and security of biometrics in general. As such, our results can support the design of both further studies to gain deeper insights and future biometric interfaces to foster the informed use of biometrics.

Vehicle-road collaboration is a promising approach for enhancing the safety and efficiency of autonomous driving by extending the intelligence of onboard systems to smart roadside infrastructures. The introduction of digital twins (DTs), particularly local DTs (LDTs) at the edge, in smart mobility presents a new embodiment of augmented intelligence, which could enhance information exchange and extract human driving expertise to improve onboard intelligence. This paper presents a novel LDT-assisted hybrid autonomous driving system for improving safety and efficiency in traffic intersections. By leveraging roadside units (RSUs) equipped with sensory and computing capabilities, the proposed system continuously monitors traffic, extracts human driving knowledge, and generates intersection-specific local driving agents through an offline reinforcement learning (RL) framework. When connected and automated vehicles (CAVs) pass through RSU-equipped intersections, RSUs can provide local agents to support safe and efficient driving in local areas. Meanwhile, they provide real-time cooperative perception (CP) to broaden onboard sensory horizons. The proposed LDT-assisted hybrid system is implemented with state-of-the-art products, e.g., CAVs and RSUs, and technologies, e.g., millimeter-wave (mmWave) communications. Hardware-in-the-loop (HiL) simulations and proof-of-concept (PoC) tests validate system performance from two standpoints: (i) The peak latency for CP and local agent downloading are 8.51 ms and 146 ms, respectively, aligning with 3GPP requirements for vehicle-to-everything (V2X) and model transfer use cases. Moreover, (ii) local driving agents can improve safety measures by 10% and reduce travel time by 15% compared with conventional onboard systems. The implemented prototype also demonstrates reliable real-time performance, fulfilling the targets of the proposed system design., Comment: 14 pages, 9 figures

We propose new algorithms to efficiently average a collection of points on a Grassmannian manifold in both the centralized and decentralized settings. Grassmannian points are used ubiquitously in machine learning, computer vision, and signal processing to represent data through (often low-dimensional) subspaces. While averaging these points is crucial to many tasks (especially in the decentralized setting), existing methods unfortunately remain computationally expensive due to the non-Euclidean geometry of the manifold. Our proposed algorithms, Rapid Grassmannian Averaging (RGrAv) and Decentralized Rapid Grassmannian Averaging (DRGrAv), overcome this challenge by leveraging the spectral structure of the problem to rapidly compute an average using only small matrix multiplications and QR factorizations. We provide a theoretical guarantee of optimality and present numerical experiments which demonstrate that our algorithms outperform state-of-the-art methods in providing high accuracy solutions in minimal time. Additional experiments showcase the versatility of our algorithms to tasks such as K-means clustering on video motion data, establishing RGrAv and DRGrAv as powerful tools for generic Grassmannian averaging., Comment: Submitted to ICLR 2025

We have carried out a detailed study of the GRB photospheric emission model predicting a quasi-blackbody spectrum slightly broader than a Planck function. This model was suggested within the relativistic fireball dynamics for interpreting a still not well understood thermal component in the GRB prompt emission, recently observed by the GBM on board the Fermi space telescope. We propose a Monte Carlo (M C) code for elucidating the observed spectrum, the outflow dynamics and its geometry for a basic and a structured plasma jets whose parameters are implemented. The code involves a simulation part describing the photon propagation assuming an unpolarized, non-dissipative relativistic outflow and a data analysis part for exploring main photospheric emission properties such as the energy, arrival time and observed flux of the simulated seed photons and the photospheric radius. Computing the latter two observables by numerical integration, we obtained values very concordant with the M C simulated results. Fitting Band functions to the photon spectra generated by this method, we derived best-fit values of the photon indices matching well those featuring the observed spectra for most typical GRBs, but corresponding to fit functions inconciliable with blackbody spectral shapes. Various derived results are reported, compared to previous ones and discussed. They show to be very sensitive to the structure of the Lorentz factor that plays a crucial role in determining the presence and strength of geometrical effects. The latter manifest themselves by large broadenings of the simulated spectra featured by multiple peak energies consistently with GRB observations. They are assumed, with multiple Compton scattering, to produce bumps pointed out at very low photon energies. Finally, developments of this work are put into perspective., Comment: Referring to the arXiv pdf file : 40 pages, 14 figures, 01 diagram (Appendix)

Local search is a common method for solving combinatorial optimisation problems. We focus on general-purpose local search solvers that accept as input a constraint model - a declarative description of a problem consisting of a set of decision variables under a set of constraints. Existing approaches typically take as input models written in solver-independent constraint modelling languages like MiniZinc. The Athanor solver we describe herein differs in that it begins from a specification of a problem in the abstract constraint specification language Essence, which allows problems to be described without commitment to low-level modelling decisions through its support for a rich set of abstract types. The advantage of proceeding from Essence is that the structure apparent in a concise, abstract specification of a problem can be exploited to generate high quality neighbourhoods automatically, avoiding the difficult task of identifying that structure in an equivalent constraint model. Based on the twin benefits of neighbourhoods derived from high level types and the scalability derived by searching directly over those types, our empirical results demonstrate strong performance in practice relative to existing solution methods., Comment: 48 pages

We review our results on a new class of $2 \to 3$ exclusive processes, as a probe of both chiral-even and chiral-odd quark GPDs. We consider the exclusive photoproduction of a photon-meson pair, in the kinematics where the pair has a large invariant mass, described in the collinear factorization framework. We cover the whole kinematical range from medium energies in fixed target experiments to very large energies of colliders, by considering the experimental conditions of JLab 12-GeV, COMPASS, future EIC and LHC (in ultra-peripheral collisions) cases. Our analysis covers neutral and charged rho-mesons, as well as charged pions. The case of the rho-meson, depending on its polarization, provides access to either chiral-even or chiral-odd GPDs, at leading twist. We find that the order of magnitude of the obtained cross sections are sufficiently large for a dedicated experimental analysis to be performed, especially at JLab. Furthermore, we compute the linear photon beam polarization asymmetry, which we find to be sizeable, in the case of a longitudinally polarized $\rho$-meson or of a charged pion. These predictions are obtained for both asymptotic distribution amplitude (DA) and holographic DA., Comment: 7 pages, 1 figure, Presented by S. Wallon at the 31st International Workshop on Deep Inelastic Scattering (DIS2024). arXiv admin note: substantial text overlap with arXiv:2401.17656

In Query-driven Travel Recommender Systems (RSs), it is crucial to understand the user intent behind challenging natural language(NL) destination queries such as the broadly worded "youth-friendly activities" or the indirect description "a high school graduation trip". Such queries are challenging due to the wide scope and subtlety of potential user intents that confound the ability of retrieval methods to infer relevant destinations from available textual descriptions such as WikiVoyage. While query reformulation (QR) has proven effective in enhancing retrieval by addressing user intent, existing QR methods tend to focus only on expanding the range of potentially matching query subtopics (breadth) or elaborating on the potential meaning of a query (depth), but not both. In this paper, we introduce Elaborative Subtopic Query Reformulation (EQR), a large language model-based QR method that combines both breadth and depth by generating potential query subtopics with information-rich elaborations. We also release TravelDest, a novel dataset for query-driven travel destination RSs. Experiments on TravelDest show that EQR achieves significant improvements in recall and precision over existing state-of-the-art QR methods., Comment: 9 pages, 7 figures,The 1st Workshop on Risks, Opportunities, and Evaluation of Generative Models in Recommender Systems (ROEGEN@RecSys 2024), October 2024, Bari, Italy

Causal models seek to unravel the cause-effect relationships among variables from observed data, as opposed to mere mappings among them, as traditional regression models do. This paper introduces a novel causal discovery algorithm designed for settings in which variables exhibit linearly sparse relationships. In such scenarios, the causal links represented by directed acyclic graphs (DAGs) can be encapsulated in a structural matrix. The proposed approach leverages the structural matrix's ability to reconstruct data and the statistical properties it imposes on the data to identify the correct structural matrix. This method does not rely on independence tests or graph fitting procedures, making it suitable for scenarios with limited training data. Simulation results demonstrate that the proposed method outperforms the well-known PC, GES, BIC exact search, and LINGAM-based methods in recovering linearly sparse causal structures.

We discuss the exclusive photoproduction of a heavy vector quarkonium, namely $J/\psi$ and $\Upsilon$ at 1-loop in $\alpha_s$. In collinear factorisation (CF), the amplitude for such a process is obtained by the convolution of a hard partonic sub-amplitude, with a universal generalised parton distribution (GPD). For the first time, we perform a complete calculation at 1-loop including full leading-log (LL) GPD evolution. We first demonstrate the huge instability of the cross section at high energies when the factorisation scale $\mu_F$ is varied. This instability has been reported previously in the literature, and occurs due to the large logarithms generated by the huge difference between the hard scale of the process, which is the mass of the heavy quarkonium here, and the centre-of-mass energy of the process. This problem was also reported in inclusive heavy vector quarkonium photoproduction. We show that this issue can be resolved by resumming these large logarithms using high-energy factorisation (HEF), by performing a matching with the result in CF using the doubly logarithmic approximation (DLA) in order to be consistent with the fixed order evolution of GPDs. Finally, we show that the cross sections obtained from such a matching, besides being free from the previously mentioned factorisation scale variation instabilities, are consistent with the H1 data for $J/\psi$ production and with the ZEUS data for $\Upsilon$ production., Comment: 7 pages, 3 figures, Presented by S. Nabeebaccus at the 31st International Workshop on Deep Inelastic Scattering (DIS2024)

We identify a $ 2 \to 3 $ exclusive process, where collinear factorisation is broken, namely the exclusive photoproduction of a $ \pi ^{0}\gamma $ pair with large invariant mass. This occurs because the process suffers from gluon exchanges trapped in the Glauber region. Using an explicit example, we show that the Glauber gluon, which is exchanged between a collinear spectator parton from the nucleon sector and a soft spectator parton from the outgoing pion, has both of its lightcone plus and minus components pinched. Therefore, it cannot be deformed to collinear/soft regions, as is often the case for processes that do factorise. We further confirm the leading power behaviour of the identified Glauber region, highlighting that this is the case although it relies on extracting a soft parton from the outgoing pion. We stress that the Glauber pinch for this process is of the leading power, due to the possibility of having two-gluon exchanges between the collinear nucleon sector and hard partonic scattering sub-process. In fact, the Glauber gluon that we identify is one of these two active gluons, and therefore, its effects are observed already at leading order. A direct consequence of our work is that collinear factorisation breaks in the same way for other $ 2 \to 3 $ exclusive processes, where two-gluon exchanges in the $ t $-channel are possible, like in the exclusive production of a photon pair from $ \pi ^{0} N $ collisions. However, we highlight that in cases where such two-gluon exchanges do not exist, like in the exclusive $ \pi ^{\pm}\gamma $ photoproduction, the Glauber exchanges that we discuss here do not occur, and hence they do not suffer from factorisation breaking effects., Comment: 7 pages, 1 figure, Presented by S. Nabeebaccus at the 31st International Workshop on Deep Inelastic Scattering (DIS2024)

In this paper, a novel generative adversarial imitation learning (GAIL)-powered policy learning approach is proposed for optimizing beamforming, spectrum allocation, and remote user equipment (RUE) association in NTNs. Traditional reinforcement learning (RL) methods for wireless network optimization often rely on manually designed reward functions, which can require extensive parameter tuning. To overcome these limitations, we employ inverse RL (IRL), specifically leveraging the GAIL framework, to automatically learn reward functions without manual design. We augment this framework with an asynchronous federated learning approach, enabling decentralized multi-satellite systems to collaboratively derive optimal policies. The proposed method aims to maximize spectrum efficiency (SE) while meeting minimum information rate requirements for RUEs. To address the non-convex, NP-hard nature of this problem, we combine the many-to-one matching theory with a multi-agent asynchronous federated IRL (MA-AFIRL) framework. This allows agents to learn through asynchronous environmental interactions, improving training efficiency and scalability. The expert policy is generated using the Whale optimization algorithm (WOA), providing data to train the automatic reward function within GAIL. Simulation results show that the proposed MA-AFIRL method outperforms traditional RL approaches, achieving a $14.6\%$ improvement in convergence and reward value. The novel GAIL-driven policy learning establishes a novel benchmark for 6G NTN optimization., Comment: Submitted to IEEE Transactions on Mobile Computing (16 pages, 10 figures)

Semantic communications (SC) is an emerging communication paradigm in which wireless devices can send only relevant information from a source of data while relying on computing resources to regenerate missing data points. However, the design of a multi-user SC system becomes more challenging because of the computing and communication overhead required for coordination. Existing solutions for learning the semantic language and performing resource allocation often fail to capture the computing and communication tradeoffs involved in multiuser SC. To address this gap, a novel framework for decentralized computing and communication resource allocation in multiuser SC systems is proposed. The challenge of efficiently allocating communication and computing resources (for reasoning) in a decentralized manner to maximize the quality of task experience for the end users is addressed through the application of Stackelberg hyper game theory. Leveraging the concept of second-level hyper games, novel analytical formulations are developed to model misperceptions of the users about each other's communication and control strategies. Further, equilibrium analysis of the learned resource allocation protocols examines the convergence of the computing and communication strategies to a local Stackelberg equilibria, considering misperceptions. Simulation results show that the proposed Stackelberg hyper game results in efficient usage of communication and computing resources while maintaining a high quality of experience for the users compared to state-of-the-art that does not account for the misperceptions.

Integrating deep learning into medical imaging is poised to greatly advance diagnostic methods but it faces challenges with generalizability. Foundation models, based on self-supervised learning, address these issues and improve data efficiency. Natural domain foundation models show promise for medical imaging, but systematic research evaluating domain adaptation, especially using self-supervised learning and parameter-efficient fine-tuning, remains underexplored. Additionally, little research addresses the issue of catastrophic forgetting during fine-tuning of foundation models. We adapted the DINOv2 vision transformer for retinal imaging classification tasks using self-supervised learning and generated two novel foundation models termed DINORET and BE DINORET. Publicly available color fundus photographs were employed for model development and subsequent fine-tuning for diabetic retinopathy staging and glaucoma detection. We introduced block expansion as a novel domain adaptation strategy and assessed the models for catastrophic forgetting. Models were benchmarked to RETFound, a state-of-the-art foundation model in ophthalmology. DINORET and BE DINORET demonstrated competitive performance on retinal imaging tasks, with the block expanded model achieving the highest scores on most datasets. Block expansion successfully mitigated catastrophic forgetting. Our few-shot learning studies indicated that DINORET and BE DINORET outperform RETFound in terms of data-efficiency. This study highlights the potential of adapting natural domain vision models to retinal imaging using self-supervised learning and block expansion. BE DINORET offers robust performance without sacrificing previously acquired capabilities. Our findings suggest that these methods could enable healthcare institutions to develop tailored vision models for their patient populations, enhancing global healthcare inclusivity., Comment: J.Zoellin, C. Merk and M. Buob contributed equally as shared-first authors. D. Cabrera DeBuc, M. D. Becker and G. M. Somfai contributed equally as senior authors for this work

In this paper, we investigate the impact of denial-of-service attacks on the status updating of a cyber-physical system with one or more sensors connected to a remote monitor via unreliable channels. We approach the problem from the perspective of an adversary that can strategically jam a subset of the channels. The sources are modeled as Markov chains, and the performance of status updating is measured based on the age of incorrect information at the monitor. Our objective is to derive jamming policies that strike a balance between the degradation of the system's performance and the conservation of the adversary's energy. For a single-source scenario, we formulate the problem as a partially-observable Markov decision process, and rigorously prove that the optimal jamming policy is of a threshold form. We then extend the problem to a multi-source scenario. We formulate this problem as a restless multi-armed bandit, and provide a jamming policy based on the Whittle's index. Our numerical results highlight the performance of our policies compared to baseline policies., Comment: arXiv admin note: text overlap with arXiv:2403.04489

We exhibit an exclusive process, namely the photoproduction of a $\pi^{0}\gamma$ pair with large invariant mass, which violates collinear factorization. We explicitly demonstrate that this is due to the fact that there exists diagrams with gluon exchange in $t$ channel, contributing at the leading power, for which Glauber gluons are trapped. This is caused by the pinching of the contour integration of both the plus and minus light-cone components of the Glauber gluon momentum. We argue that this leads to the observed ``endpoint-like'' divergence of the convolution integral at leading order and leading power when collinear factorization is na\"ively assumed., Comment: 7 pages, 3 figures. arXiv admin note: text overlap with arXiv:2311.09146

Inference-time techniques are emerging as highly effective tools to enhance large language model (LLM) capabilities. However, best practices for developing systems that combine these techniques remain underdeveloped due to our limited understanding of the utility of individual inference-time techniques and the interactions between them. Additionally, efficiently and automatically searching the space of model choices, inference-time techniques, and their compositions is challenging due to the large design space. To address these challenges, we introduce Archon, a modular framework for selecting, combining, and stacking layers of inference-time techniques to construct optimized LLM systems for target benchmarks. Rather than relying on a single LLM called once, we leverage a diverse set of LLMs and inference-time techniques, creating LLM systems greater than the sum of their parts. Archon defines an extensible design space, encompassing techniques such as generation ensembling, repeated sampling, ranking, fusion, critiquing, verification, and unit testing. It transforms the problem of building LLM systems into a hyperparameter optimization objective. Given the available LLMs, inference-time techniques, and compute budget, Archon utilizes hyperparameter search techniques to discover optimized architectures for target benchmark(s). We evaluate Archon architectures across a range of instruction-following, reasoning, and coding benchmarks, including MT-Bench, Arena-Hard-Auto, AlpacaEval 2.0, MixEval, MixEval Hard, MATH, and CodeContests. Archon architectures outperform frontier models, such as GPT-4o and Claude 3.5 Sonnet, on these benchmarks, achieving an average accuracy increase of 15.1 percentage points by using all available LLMs. We make our code and datasets available publicly on Github: https://github.com/ScalingIntelligence/Archon.

This paper describes the software package Cucheb, a GPU implementation of the filtered Lanczos procedure for the solution of large sparse symmetric eigenvalue problems. The filtered Lanczos procedure uses a carefully chosen polynomial spectral transformation to accelerate convergence of the Lanczos method when computing eigenvalues within a desired interval. This method has proven particularly effective for eigenvalue problems that arise in electronic structure calculations and density functional theory. We compare our implementation against an equivalent CPU implementation and show that using the GPU can reduce the computation time by more than a factor of 10.

Large language models have shown promise in various domains, including healthcare. In this study, we conduct a comprehensive evaluation of LLMs in the context of mental health tasks using social media data. We explore the zero-shot (ZS) and few-shot (FS) capabilities of various LLMs, including GPT-4, Llama 3, Gemini, and others, on tasks such as binary disorder detection, disorder severity evaluation, and psychiatric knowledge assessment. Our evaluation involved 33 models testing 9 main prompt templates across the tasks. Key findings revealed that models like GPT-4 and Llama 3 exhibited superior performance in binary disorder detection, with accuracies reaching up to 85% on certain datasets. Moreover, prompt engineering played a crucial role in enhancing model performance. Notably, the Mixtral 8x22b model showed an improvement of over 20%, while Gemma 7b experienced a similar boost in performance. In the task of disorder severity evaluation, we observed that FS learning significantly improved the model's accuracy, highlighting the importance of contextual examples in complex assessments. Notably, the Phi-3-mini model exhibited a substantial increase in performance, with balanced accuracy improving by over 6.80% and mean average error dropping by nearly 1.3 when moving from ZS to FS learning. In the psychiatric knowledge task, recent models generally outperformed older, larger counterparts, with the Llama 3.1 405b achieving an accuracy of 91.2%. Despite promising results, our analysis identified several challenges, including variability in performance across datasets and the need for careful prompt engineering. Furthermore, the ethical guards imposed by many LLM providers hamper the ability to accurately evaluate their performance, due to tendency to not respond to potentially sensitive queries.

In this paper, a space-air-ground quantum (SPARQ) network is developed as a means for providing a seamless on-demand entanglement distribution. The node mobility in SPARQ poses significant challenges to entanglement routing. Existing quantum routing algorithms focus on stationary ground nodes and utilize link distance as an optimality metric, which is unrealistic for dynamic systems like SPARQ. Moreover, in contrast to the prior art that assumes homogeneous nodes, SPARQ encompasses heterogeneous nodes with different functionalities further complicates the entanglement distribution. To solve the entanglement routing problem, a deep reinforcement learning (RL) framework is proposed and trained using deep Q-network (DQN) on multiple graphs of SPARQ to account for the network dynamics. Subsequently, an entanglement distribution policy, third-party entanglement distribution (TPED), is proposed to establish entanglement between communication parties. A realistic quantum network simulator is designed for performance evaluation. Simulation results show that the TPED policy improves entanglement fidelity by 3% and reduces memory consumption by 50% compared with benchmark. The results also show that the proposed DQN algorithm improves the number of resolved teleportation requests by 39% compared with shortest path baseline and the entanglement fidelity by 2% compared with an RL algorithm that is based on long short-term memory (LSTM). It also improved entanglement fidelity by 6% and 9% compared with two state-of-the-art benchmarks. Moreover, the entanglement fidelity is improved by 15% compared with DQN trained on a snapshot of SPARQ. Additionally, SPARQ enhances the average entanglement fidelity by 23.5% compared with existing networks spanning only space and ground layers.

This paper introduces the concept of an education tool that utilizes Generative Artificial Intelligence (GenAI) to enhance storytelling for children. The system combines GenAI-driven narrative co-creation, text-to-speech conversion, and text-to-video generation to produce an engaging experience for learners. We describe the co-creation process, the adaptation of narratives into spoken words using text-to-speech models, and the transformation of these narratives into contextually relevant visuals through text-to-video technology. Our evaluation covers the linguistics of the generated stories, the text-to-speech conversion quality, and the accuracy of the generated visuals.

The SoccerNet 2024 challenges represent the fourth annual video understanding challenges organized by the SoccerNet team. These challenges aim to advance research across multiple themes in football, including broadcast video understanding, field understanding, and player understanding. This year, the challenges encompass four vision-based tasks. (1) Ball Action Spotting, focusing on precisely localizing when and which soccer actions related to the ball occur, (2) Dense Video Captioning, focusing on describing the broadcast with natural language and anchored timestamps, (3) Multi-View Foul Recognition, a novel task focusing on analyzing multiple viewpoints of a potential foul incident to classify whether a foul occurred and assess its severity, (4) Game State Reconstruction, another novel task focusing on reconstructing the game state from broadcast videos onto a 2D top-view map of the field. Detailed information about the tasks, challenges, and leaderboards can be found at https://www.soccer-net.org, with baselines and development kits available at https://github.com/SoccerNet., Comment: 7 pages, 1 figure

Learning semantically meaningful image transformations (i.e. rotation, thickness, blur) directly from examples can be a challenging task. Recently, the Manifold Autoencoder (MAE) proposed using a set of Lie group operators to learn image transformations directly from examples. However, this approach has limitations, as the learned operators are not guaranteed to be disentangled and the training routine is prohibitively expensive when scaling up the model. To address these limitations, we propose MANGO (transformation Manifolds with Grouped Operators) for learning disentangled operators that describe image transformations in distinct latent subspaces. Moreover, our approach allows practitioners the ability to define which transformations they aim to model, thus improving the semantic meaning of the learned operators. Through our experiments, we demonstrate that MANGO enables composition of image transformations and introduces a one-phase training routine that leads to a 100x speedup over prior works., Comment: Submitted to IEEE ICASSP 2025. This work has been submitted to the IEEE for possible publication

Digital twins in manufacturing serve as a crucial bridge between the industrial age and the digital age, offering immense value. Current additive manufacturing processes are able to generate vast amounts of in-process data, which, when effectively ingested, can be transformed into insightful decisions. Data-driven methods from reduced order modeling and system identification are particularly promising in managing this data deluge. This study focuses on Laser Powder Directed Energy Deposition (LP-DED) equipped with in-situ process measurements to develop a compact virtual test-bed. This test-bed can accurately ingest arbitrary process inputs and report in-process observables as outputs. This virtual test-bed is derived using Dynamic Mode Decomposition with Control (DMDc) and is coupled with uncertainty quantification techniques to ensure robust predictions., Comment: 22 pages, 12 figures

RGB-D cameras supply rich and dense visual and spatial information for various robotics tasks such as scene understanding, map reconstruction, and localization. Integrating depth and visual information can aid robots in localization and element mapping, advancing applications like 3D scene graph generation and Visual Simultaneous Localization and Mapping (VSLAM). While point cloud data containing such information is primarily used for enhanced scene understanding, exploiting their potential to capture and represent rich semantic information has yet to be adequately targeted. This paper presents a real-time pipeline for localizing building components, including wall and ground surfaces, by integrating geometric calculations for pure 3D plane detection followed by validating their semantic category using point cloud data from RGB-D cameras. It has a parallel multi-thread architecture to precisely estimate poses and equations of all the planes detected in the environment, filters the ones forming the map structure using a panoptic segmentation validation, and keeps only the validated building components. Incorporating the proposed method into a VSLAM framework confirmed that constraining the map with the detected environment-driven semantic elements can improve scene understanding and map reconstruction accuracy. It can also ensure (re-)association of these detected components into a unified 3D scene graph, bridging the gap between geometric accuracy and semantic understanding. Additionally, the pipeline allows for the detection of potential higher-level structural entities, such as rooms, by identifying the relationships between building components based on their layout., Comment: 6 pages, 5 figures. 3 tables

In recent years robots have become an important part of our day-to-day lives with various applications. Human-robot interaction creates a positive impact in the field of robotics to interact and communicate with the robots. Gesture recognition techniques combined with machine learning algorithms have shown remarkable progress in recent years, particularly in human-robot interaction (HRI). This paper comprehensively reviews the latest advancements in gesture recognition methods and their integration with machine learning approaches to enhance HRI. Furthermore, this paper represents the vision-based gesture recognition for safe and reliable human-robot-interaction with a depth-sensing system, analyses the role of machine learning algorithms such as deep learning, reinforcement learning, and transfer learning in improving the accuracy and robustness of gesture recognition systems for effective communication between humans and robots., Comment: 19 pages,1 Figure

Robot manipulators have become a significant tool for production industries due to their advantages in high speed, accuracy, safety, and repeatability. This paper simulates and optimizes the design of a 3-DOF articulated robotic manipulator (RRR Configuration). The forward and inverse dynamic models are utilized. The trajectory is planned using the end effector's required initial position. A torque compute model is used to calculate the physical end effector's trajectory, position, and velocity. The MATLAB Simulink platform is used for all simulations of the RRR manipulator. With the aid of MATLAB, we primarily focused on manipulator control of the robot using a calculated torque control strategy to achieve the required position.

In prior studies, a very minimal Yukawa sector within the $SO(10)$ Grand Unified Theory framework has been identified, comprising of Higgs fields belonging to a real $10_H$, a real $120_H$, and a $\overline{126}_H$ dimensional representations. In this work, within this minimal framework, we have obtained fits to fermion masses and mixings while successfully reproducing the cosmological baryon asymmetry via leptogenesis.The right-handed neutrino ($N_i$) mass spectrum obtained from the fit is strongly hierarchical, suggesting that $B-L$ asymmetry is dominantly produced from $N_2$ dynamics while $N_1$ is responsible for erasing the excess asymmetry. With this rather constrained Yukawa sector, fits are obtained both for normal and inverted ordered neutrino mass spectra, consistent with leptonic CP-violating phase $\delta_\mathrm{CP}$ indicated by global fits of neutrino oscillation data, while also satisfying the current limits from neutrinoless double beta decay experiments. In particular, the the leptonic CP-violating phase has a preference to be in the range $\delta_\mathrm{CP}\simeq (230-300)^\circ$. We also show the consistency of the framework with gauge coupling unification and proton lifetime limits., Comment: 30 pages + references, 3 figures; comments are welcome