Search

Pseudo links are equivalence classes under Reidemeister-type moves of link diagrams containing crossings with undefined over and under information. In this paper, we extend the Kauffman bracket and Jones polynomials from planar pseudo links to pseudo links in the annulus and in the torus, and their respective lifts from the three-space to the solid torus and the thickened torus. Moreover, since annular and toroidal pseudo links can be represented as mixed links in the three-sphere, we also introduce the respective Kauffman bracket and Jones polynomials for their planar mixed link diagrams. Our work provides new tools for the study of annular and toroidal pseudo links.

Federated Learning (FL) enables clients to share learning parameters instead of local data, reducing communication overhead. Traditional wireless networks face latency challenges with FL. In contrast, Cell-Free Massive MIMO (CFmMIMO) can serve multiple clients on shared resources, boosting spectral efficiency and reducing latency for large-scale FL. However, clients' communication resource limitations can hinder the completion of the FL training. To address this challenge, we propose an energy-efficient, low-latency FL framework featuring optimized uplink power allocation for seamless client-server collaboration. Our framework employs an adaptive quantization scheme, dynamically adjusting bit allocation for local gradient updates to reduce communication costs. We formulate a joint optimization problem covering FL model updates, local iterations, and power allocation, solved using sequential quadratic programming (SQP) to balance energy and latency. Additionally, clients use the AdaDelta method for local FL model updates, enhancing local model convergence compared to standard SGD, and we provide a comprehensive analysis of FL convergence with AdaDelta local updates. Numerical results show that, within the same energy and latency budgets, our power allocation scheme outperforms the Dinkelbach and max-sum rate methods by increasing the test accuracy up to $7$\% and $19$\%, respectively. Moreover, for the three power allocation methods, our proposed quantization scheme outperforms AQUILA and LAQ by increasing test accuracy by up to $36$\% and $35$\%, respectively.

Doubly periodic tangles (DP tangles) are configurations of curves embedded in the thickened plane, invariant under translations in two transversal directions. In this paper we extend the classical theory of DP tangles by introducing the theory of {\it doubly periodic pseudo tangles} (pseudo DP tangles), which incorporate undetermined crossings called {\it precrossings}, inspired by the theory of pseudo knots. Pseudo DP tangles are defined as liftings of spatial pseudo links in the thickened torus, called {\it pseudo motifs}, and are analyzed through diagrammatic methods that account for both local and global isotopies. We emphasize on {\it pseudo scale equivalence}, a concept defining equivalence between finite covers of pseudo motif diagrams. We investigate the notion of equivalence for these structures, leading to an analogue of the Reidemeister theorem for pseudo DP tangles. Furthermore, we address the complexities introduced by pseudo scale equivalence in defining minimal pseudo motif diagrams. This work contributes to the broader understanding of periodic entangled structures and can find applications in diverse fields such as textiles, materials science and crystallography due to their periodic nature.

Federated learning (FL) is a distributed learning framework where users train a global model by exchanging local model updates with a server instead of raw datasets, preserving data privacy and reducing communication overhead. However, the latency grows with the number of users and the model size, impeding the successful FL over traditional wireless networks with orthogonal access. Cell-free massive multiple-input multipleoutput (CFmMIMO) is a promising solution to serve numerous users on the same time/frequency resource with similar rates. This architecture greatly reduces uplink latency through spatial multiplexing but does not take application characteristics into account. In this paper, we co-optimize the physical layer with the FL application to mitigate the straggler effect. We introduce a novel adaptive mixed-resolution quantization scheme of the local gradient vector updates, where only the most essential entries are given high resolution. Thereafter, we propose a dynamic uplink power control scheme to manage the varying user rates and mitigate the straggler effect. The numerical results demonstrate that the proposed method achieves test accuracy comparable to classic FL while reducing communication overhead by at least 93% on the CIFAR-10, CIFAR-100, and Fashion-MNIST datasets. We compare our methods against AQUILA, Top-q, and LAQ, using the max-sum rate and Dinkelbach power control schemes. Our approach reduces the communication overhead by 75% and achieves 10% higher test accuracy than these benchmarks within a constrained total latency budget.

Computer vision tasks such as object detection and segmentation rely on the availability of extensive, accurately annotated datasets. In this work, We present CIA, a modular pipeline, for (1) generating synthetic images for dataset augmentation using Stable Diffusion, (2) filtering out low quality samples using defined quality metrics, (3) forcing the existence of specific patterns in generated images using accurate prompting and ControlNet. In order to show how CIA can be used to search for an optimal augmentation pipeline of training data, we study human object detection in a data constrained scenario, using YOLOv8n on COCO and Flickr30k datasets. We have recorded significant improvement using CIA-generated images, approaching the performances obtained when doubling the amount of real images in the dataset. Our findings suggest that our modular framework can significantly enhance object detection systems, and make it possible for future research to be done on data-constrained scenarios. The framework is available at: github.com/multitel-ai/CIA.

Characterizing domains is essential for models analyzing dynamic environments, as it allows them to adapt to evolving conditions or to hand the task over to backup systems when facing conditions outside their operational domain. Existing solutions typically characterize a domain by solving a regression or classification problem, which limits their applicability as they only provide a limited summarized description of the domain. In this paper, we present a novel approach to domain characterization by characterizing domains as probability distributions. Particularly, we develop a method to predict the likelihood of different weather conditions from images captured by vehicle-mounted cameras by estimating distributions of physical parameters using normalizing flows. To validate our proposed approach, we conduct experiments within the context of autonomous vehicles, focusing on predicting the distribution of weather parameters to characterize the operational domain. This domain is characterized by physical parameters (absolute characterization) and arbitrarily predefined domains (relative characterization). Finally, we evaluate whether a system can safely operate in a target domain by comparing it to multiple source domains where safety has already been established. This approach holds significant potential, as accurate weather prediction and effective domain adaptation are crucial for autonomous systems to adjust to dynamic environmental conditions.

Left ventricular ejection fraction (LVEF) is the most important clinical parameter of cardiovascular function. The accuracy in estimating this parameter is highly dependent upon the precise segmentation of the left ventricle (LV) structure at the end diastole and systole phases. Therefore, it is crucial to develop robust algorithms for the precise segmentation of the heart structure during different phases. Methodology: In this work, an improved 3D UNet model is introduced to segment the myocardium and LV, while excluding papillary muscles, as per the recommendation of the Society for Cardiovascular Magnetic Resonance. For the practical testing of the proposed framework, a total of 8,400 cardiac MRI images were collected and analysed from the military hospital in Tunis (HMPIT), as well as the popular ACDC public dataset. As performance metrics, we used the Dice coefficient and the F1 score for validation/testing of the LV and the myocardium segmentation. Results: The data was split into 70%, 10%, and 20% for training, validation, and testing, respectively. It is worth noting that the proposed segmentation model was tested across three axis views: basal, medio basal and apical at two different cardiac phases: end diastole and end systole instances. The experimental results showed a Dice index of 0.965 and 0.945, and an F1 score of 0.801 and 0.799, at the end diastolic and systolic phases, respectively. Additionally, clinical evaluation outcomes revealed a significant difference in the LVEF and other clinical parameters when the papillary muscles were included or excluded.

Player Experience Modelling (PEM) is the study of AI techniques applied to modelling a player's experience within a video game. PEM development can be labour-intensive, requiring expert hand-authoring or specialized data collection. In this work, we propose a novel PEM development approach, approximating player experience from gameplay video. We evaluate this approach predicting affect in the game Angry Birds via a human subject study. We validate that our PEM can strongly correlate with self-reported and sensor measures of affect, demonstrating the potential of this approach., Comment: 9 pages, 3 figures, AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment

The cosmic birefringence (CB) angle refers to the rotation of the linear polarization plane of Cosmic Microwave Background (CMB) radiations when parity-violating theories are considered. We analyzed the Quantum Boltzmann equation for an ensemble of CMB photons interacting with the right-handed sterile neutrino dark matter (DM) and axion-like particles (ALPs) DM in the presence of the scalar metric perturbation. We used the birefringence angle of CMB to study those probable candidates of DM. It is shown that the CB angle contribution of sterile neutrino is much less that two other sources considered here. Next, we combined the results of the cosmic neutrinos' contribution and the contribution of the ALPs to producing the CMB birefringence and discussed the uncertainty on the parameter space of axions caused by the share of CMB-cosmic neutrino interaction in generating this effect. Finally, we plotted the EB power spectrum of the CMB and showed that this spectrum behaves differently in the presence of cosmic neutrinos and ALPs interactions in small $l$. Hence, future observed data for $C^{l}_{EB}$, will help us to distinguish the CB angle value due to the various sources of its production.

Personalized virtual reality exposure therapy is a therapeutic practice that can adapt to an individual patient, leading to better health outcomes. Measuring a patient's mental state to adjust the therapy is a critical but difficult task. Most published studies use subjective methods to estimate a patient's mental state, which can be inaccurate. This article proposes a virtual reality exposure therapy (VRET) platform capable of assessing a patient's mental state using non-intrusive and widely available physiological signals such as photoplethysmography (PPG). In a case study, we evaluate how PPG signals can be used to detect two binary classifications: peaceful and stressful states. Sixteen healthy subjects were exposed to the two VR environments (relaxed and stressful). Using LOSO cross-validation, our best classification model could predict the two states with a 70.6% accuracy which outperforms many more complex approaches., Comment: Updated code and data available at https://github.com/athar70/Stress-Estimation

The need to generate a spider to provoke a desired anxiety response arises in the context of personalized virtual reality exposure therapy (VRET), a treatment approach for arachnophobia. This treatment involves patients observing virtual spiders in order to become desensitized and decrease their phobia, which requires that the spiders elicit specific anxiety responses. However, VRET approaches tend to require therapists to hand-select the appropriate spider for each patient, which is a time-consuming process and takes significant technical knowledge and patient insight. While automated methods exist, they tend to employ rules-based approaches with minimal ability to adapt to specific users. To address these challenges, we present a framework for VRET utilizing procedural content generation (PCG) and reinforcement learning (RL), which automatically adapts a spider to elicit a desired anxiety response. We demonstrate the superior performance of this system compared to a more common rules-based VRET method., Comment: Under review at ACM Transactions on Interactive Intelligent Systems (TIIS). Code and data available at https://github.com/athar70/EDPCGRL4Spider

Due to intensive genetic selection for rapid growth rates and high broiler yields in recent years, the global poultry industry has faced a challenging problem in the form of woody breast (WB) conditions. This condition has caused significant economic losses as high as $200 million annually, and the root cause of WB has yet to be identified. Human palpation is the most common method of distinguishing a WB from others. However, this method is time-consuming and subjective. Hyperspectral imaging (HSI) combined with machine learning algorithms can evaluate the WB conditions of fillets in a non-invasive, objective, and high-throughput manner. In this study, 250 raw chicken breast fillet samples (normal, mild, severe) were taken, and spatially heterogeneous hardness distribution was first considered when designing HSI processing models. The study not only classified the WB levels from HSI but also built a regression model to correlate the spectral information with sample hardness data. To achieve a satisfactory classification and regression model, a neural network architecture search (NAS) enabled a wide-deep neural network model named NAS-WD, which was developed. In NAS-WD, NAS was first used to automatically optimize the network architecture and hyperparameters. The classification results show that NAS-WD can classify the three WB levels with an overall accuracy of 95%, outperforming the traditional machine learning model, and the regression correlation between the spectral data and hardness was 0.75, which performs significantly better than traditional regression models.

The stacking order of two-dimensional transition metal dichalcogenides (TMDs) is attracting tremendous interest as an essential component of van der Waals heterostructures. A common and fast approach to distinguish between the AAprime (2H) and AB (3R) configurations uses the relative edge orientation of each triangular layer (theta) from optical images. Here, we highlight that this method alone is not sufficient to fully identify the stacking order. Instead we propose a model and methodology to accurately determine the bilayer configuration of WSe2 using second harmonic generation (SHG) and Raman spectroscopy. We demonstrate that the SHG response of the AB phase (theta = 0) deg layers is more intense than the signal from the single layer structure. However, the SHG totally vanishes in the AAprime and ABprime phases (theta = 60 deg) and 0 deg respectively) of homo-bilayer WSe2. Also, several optical features of homo-bilayer WSe2 are found to depend on the details of the stacking order, with the difference being the clearest in the low frequency (LF) Raman frequencies, as confirmed by DFT simulation. This allows unambiguous, high-throughput, nondestructive identification of stacking order in TMDs, which is not robustly addressed in this emerging research area.

The muon's magnetic moment is a fundamental quantity in particle physics and the deviation of its value from quantum electrodynamics (QED), motivates research beyond the standard models (SM). In this study, we utilize the effective coupling of right-handed sterile neutrinos with SM gauge bosons to calculate the muon anomalous magnetic moment ($\boldsymbol{\mu}$AMM) at one-loop level. The contribution of the sterile neutrino interactions on the $\boldsymbol{\mu}$AMM is calculated by considering the standard and non-standard neutrino interactions. Our results show that the standard sterile neutrino interactions give a negligible contribution to $\Delta a_{\boldsymbol{\mu}}$ while the non-standard neutrino interactions can play a significant role in explaining the muon $(g-2)$ anomaly. In the context of the non-standard neutrino interaction, our calculation shows that a Dirac mass scale $M_D$ around $100\,\text{GeV}$ could explain the muon anomaly if the right handed sterile neutrino's coupling with SM particles is about $\mathcal{G}_R\approx 10^{-3}$. We have also plotted the allowed region of the model parameters that satisfy the experimental data on $\Delta a_{{\boldsymbol{\mu}}}^{SN}$ and discuss the percentage of the ${\boldsymbol{\mu}}$ anomaly compensation in terms of the coupling constant $\mathcal{G}_R$.

Kaon physics is at an important experimental juncture with respect to the ongoing measurements of several observables. This work will build on the existing status by formulating different phenomenological analyses corresponding to different paths that may lie ahead. Beginning with the golden channels, $K^+\rightarrow\pi^+\nu\bar\nu$ and $K_L\rightarrow\pi^0\nu\bar\nu$, the paper will eventually cast the spotlight on the importance of a precise measurement of BR($K_L\rightarrow\pi^0\ell\bar\ell$). The phenomenological analyses involve sequentially adding kaon physics observables at the projected final precision of their respective measurements to the global fit. More specifically, we consider three different scenarios with different sets of observables assumed at their final precisions. Beginning with BR($K^+\rightarrow\pi^+\nu\bar\nu$) and BR($K_L\rightarrow\pi^0\nu\bar\nu$), we sequentially add BR($K_L\rightarrow\pi^0 e\bar e$) and BR($K_L\rightarrow\pi^0 \mu\bar\mu$) to the global fit. The evolution of the result from one scenario to the next makes a strong case for the consideration of future measurement of BR($K_L\rightarrow\pi^0\ell\bar\ell$)., Comment: 16 pages, 5 figures, 1 table. v2: Appendix A added

In this paper, we extend the theory of planar pseudo knots to the theories of annular and toroidal pseudo knots. Pseudo knots are defined as equivalence classes under Reidemeister-like moves of knot diagrams characterized by crossings with undefined over/under information. In the theories of annular and toroidal pseudo knots we introduce their respective lifts to the solid and the thickened torus. Then, we interlink these theories by representing annular and toroidal pseudo knots as planar ${\rm O}$-mixed and ${\rm H}$-mixed pseudo links. We also explore the inclusion relations between planar, annular and toroidal pseudo knots, as well as of ${\rm O}$-mixed and ${\rm H}$-mixed pseudo links. Finally, we extend the planar weighted resolution set to annular and toroidal pseudo knots, defining new invariants for classifying pseudo knots and links in the solid and in the thickened torus., Comment: 22 pages, 20 figures

Vision-Language Models for remote sensing have shown promising uses thanks to their extensive pretraining. However, their conventional usage in zero-shot scene classification methods still involves dividing large images into patches and making independent predictions, i.e., inductive inference, thereby limiting their effectiveness by ignoring valuable contextual information. Our approach tackles this issue by utilizing initial predictions based on text prompting and patch affinity relationships from the image encoder to enhance zero-shot capabilities through transductive inference, all without the need for supervision and at a minor computational cost. Experiments on 10 remote sensing datasets with state-of-the-art Vision-Language Models demonstrate significant accuracy improvements over inductive zero-shot classification. Our source code is publicly available on Github: https://github.com/elkhouryk/RS-TransCLIP

We introduce a novel method to simultaneously perform variable selection and estimation in the joint frailty model of recurrent and terminal events using the Broken Adaptive Ridge Regression penalty. The BAR penalty can be summarized as an iteratively reweighted squared $L_2$-penalized regression, which approximates the $L_0$-regularization method. Our method allows for the number of covariates to diverge with the sample size. Under certain regularity conditions, we prove that the BAR estimator implemented under the model framework is consistent and asymptotically normally distributed, which are known as the oracle properties in the variable selection literature. In our simulation studies, we compare our proposed method to the Minimum Information Criterion (MIC) method. We apply our method on the Medical Information Mart for Intensive Care (MIMIC-III) database, with the aim of investigating which variables affect the risks of repeated ICU admissions and death during ICU stay.

In this paper, a novel zonal machine learning (ML) approach for Reynolds-averaged Navier-Stokes (RANS) turbulence modelling based on the divide-and-conquer technique is introduced. This approach involves partitioning the flow domain into regions of flow physics called zones, training one ML model in each zone, then validating and testing them on their respective zones. The approach was demonstrated with the tensor basis neural network (TBNN) and another neural net called the turbulent kinetic energy neural network (TKENN). These were used to predict Reynolds stress anisotropy and turbulent kinetic energy respectively in test cases of flow over a solid block, which contain regions of different flow physics including separated flows. The results show that the combined predictions given by the zonal TBNNs and TKENNs were significantly more accurate than their corresponding standard non-zonal models. Most notably, shear anisotropy component in the test cases was predicted at least 20% and 55% more accurately on average by the zonal TBNNs compared to the non-zonal TBNN and RANS, respectively. The Reynolds stress constructed with the zonal predictions was also found to be at least 23% more accurate than those obtained with the non-zonal approach and 30% more accurate than the Reynolds stress predicted by RANS on average. These improvements were attributed to the shape of the zones enabling the zonal models to become highly locally optimized at predicting the output.

This paper investigates the generation and properties of gravitational radiation within the framework of Generalized Brans-Dicke (GBD) theory, with a specific emphasis on its manifestation in compact binary systems. The primary focus of this study lies in the comprehensive exploration of gravitational radiation generated by compact binaries. The energy momentum tensor and the associated gravitational wave (GW) radiation power in GBD theory are investigated, elucidating the relationship between these fundamental concepts. Furthermore, detailed calculations are provided for the GW radiation power originating from both tensor fields and scalar fields. Based on our calculations, both scalar fields contribute to GW radiation by producing dipole radiation. We also study the period derivative of compact binaries in this theory. By comparing with the observational data of the orbital period derivative of the quasicircular white dwarf-neutron star binary PSR J1012+5307, we put bounds on the two parameters of the theory: the Brans-Dicke coupling parameter $\omega_{0}$ and the mass of geometrical scalar field $m_f$, \textcolor{black}{resulting in a lower bound $\omega_{0}>6.09723\times10^6$ for a massless BD scalar field and the geometrical field whose mass is smaller than $10^{-29} \text{GeV}$. The obtained bound on $\omega_0$ is two orders of magnitude stricter than those derived from solar system data.} Finally, we find the phase shift that GWs experience in the frequency domain during their propagation., Comment: Accepted for Eur. Phys. J. C

The search for New Physics (NP) beyond the Standard Model (SM) has been a central focus of particle physics, including in the context of $B$-meson decays involving $b \to s \ell \ell$ transitions. These transitions, mediated by flavour-changing neutral currents, are highly sensitive to small NP effects due to their suppression in the SM. While direct searches at colliders have not yet led to NP discoveries, indirect probes through semi-leptonic decays have revealed anomalies in observables such as the branching fraction $\mathcal{B}(B \to K \mu \mu)$ and the angular observable $P_5'(B \to K^* \mu \mu)$. In order to assess the observed tensions, it is essential to ensure an accurate SM prediction. In this review, we examine the theoretical basis of the $B \to K^{(*)} \ell \ell$ decays, addressing in particular key uncertainties arising from local and non-local form factors. We also discuss the impact of QED corrections to the Wilson coefficients, as well as the effect of CKM matrix elements on the predictions and the tension with the experimental measurements. We discuss the most recent results, highlighting ongoing efforts to refine predictions and to constrain potential signs of NP in these critical decay processes., Comment: 29 pages, 4 figures, 3 tables. Contribution to the "Symmetry" Special Issue "Symmetries and Anomalies in Flavour Physics"

Objective: This study aims to develop and validate an evaluation framework to ensure the safety and reliability of mental health chatbots, which are increasingly popular due to their accessibility, human-like interactions, and context-aware support. Materials and Methods: We created an evaluation framework with 100 benchmark questions and ideal responses, and five guideline questions for chatbot responses. This framework, validated by mental health experts, was tested on a GPT-3.5-turbo-based chatbot. Automated evaluation methods explored included large language model (LLM)-based scoring, an agentic approach using real-time data, and embedding models to compare chatbot responses against ground truth standards. Results: The results highlight the importance of guidelines and ground truth for improving LLM evaluation accuracy. The agentic method, dynamically accessing reliable information, demonstrated the best alignment with human assessments. Adherence to a standardized, expert-validated framework significantly enhanced chatbot response safety and reliability. Discussion: Our findings emphasize the need for comprehensive, expert-tailored safety evaluation metrics for mental health chatbots. While LLMs have significant potential, careful implementation is necessary to mitigate risks. The superior performance of the agentic approach underscores the importance of real-time data access in enhancing chatbot reliability. Conclusion: The study validated an evaluation framework for mental health chatbots, proving its effectiveness in improving safety and reliability. Future work should extend evaluations to accuracy, bias, empathy, and privacy to ensure holistic assessment and responsible integration into healthcare. Standardized evaluations will build trust among users and professionals, facilitating broader adoption and improved mental health support through technology.

In this article, we discuss two algorithms tailored to discrete-time deterministic finite-horizon nonlinear optimal control problems or so-called deterministic trajectory optimization problems. Both algorithms can be derived from an emerging theoretical paradigm that we refer to as probabilistic optimal control. The paradigm reformulates stochastic optimal control as an equivalent probabilistic inference problem and can be viewed as a generalisation of the former. The merit of this perspective is that it allows to address the problem using the Expectation-Maximization algorithm. It is shown that the application of this algorithm results in a fixed point iteration of probabilistic policies that converge to the deterministic optimal policy. Two strategies for policy evaluation are discussed, using state-of-the-art uncertainty quantification methods resulting into two distinct algorithms. The algorithms are structurally closest related to the differential dynamic programming algorithm and related methods that use sigma-point methods to avoid direct gradient evaluations. The main advantage of the algorithms is an improved balance between exploration and exploitation over the iterations, leading to improved numerical stability and accelerated convergence. These properties are demonstrated on different nonlinear systems.

Optimal information and knowledge management is crucial for organizations to achieve their objectives efficiently. As a rare and valuable resource, effective knowledge management provides a strategic advantage and has become a key determinant of organizational success. The study aims to identify critical success and failure factors for implementing knowledge management systems in government organizations. This research employs a descriptive survey methodology, collecting data through random interviews and questionnaires. The study highlights the critical success factors for knowledge management systems in government organizations, including cooperation, an open atmosphere, staff training, creativity and innovation, removal of organizational constraints, reward policies, role modeling, and focus. Conversely, failure to consider formality, staff participation, collaboration technologies, network and hardware infrastructure, complexity, IT staff, and trust can pose significant obstacles to successful implementation.

Stance detection, the classification of attitudes expressed in a text towards a specific topic, is vital for applications like fake news detection and opinion mining. However, the scarcity of labeled data remains a challenge for this task. To address this problem, we propose Dynamic Model Adaptation with Contextual Data Generation (DyMoAdapt) that combines Few-Shot Learning and Large Language Models. In this approach, we aim to fine-tune an existing model at test time. We achieve this by generating new topic-specific data using GPT-3. This method could enhance performance by allowing the adaptation of the model to new topics. However, the results did not increase as we expected. Furthermore, we introduce the Multi Generated Topic VAST (MGT-VAST) dataset, which extends VAST using GPT-3. In this dataset, each context is associated with multiple topics, allowing the model to understand the relationship between contexts and various potential topics, Comment: 5 pages, AAAI-2024 Workshop on Public Sector LLMs

Coreference resolution, critical for identifying textual entities referencing the same entity, faces challenges in pronoun resolution, particularly identifying pronoun antecedents. Existing methods often treat pronoun resolution as a separate task from mention detection, potentially missing valuable information. This study proposes the first end-to-end neural network system for Persian pronoun resolution, leveraging pre-trained Transformer models like ParsBERT. Our system jointly optimizes both mention detection and antecedent linking, achieving a 3.37 F1 score improvement over the previous state-of-the-art system (which relied on rule-based and statistical methods) on the Mehr corpus. This significant improvement demonstrates the effectiveness of combining neural networks with linguistic models, potentially marking a significant advancement in Persian pronoun resolution and paving the way for further research in this under-explored area.

The rapid growth of e-commerce has placed considerable pressure on customs representatives, prompting advanced methods. In tackling this, Artificial intelligence (AI) systems have emerged as a promising approach to minimize the risks faced. Given that the Harmonized System (HS) code is a crucial element for an accurate customs declaration, we propose a novel multimodal HS code prediction approach using deep learning models exploiting both image and text features obtained through the customs declaration combined with e-commerce platform information. We evaluated two early fusion methods and introduced our MultConcat fusion method. To the best of our knowledge, few studies analyze the featurelevel combination of text and image in the state-of-the-art for HS code prediction, which heightens interest in our paper and its findings. The experimental results prove the effectiveness of our approach and fusion method with a top-3 and top-5 accuracy of 93.5% and 98.2% respectively, Comment: Accepted for poster presentation at ESANN 2023