Search

Recent contrastive multimodal vision-language models like CLIP have demonstrated robust open-world semantic understanding, becoming the standard image backbones for vision-language applications due to their aligned latent space. However, this practice has left powerful unimodal encoders for both vision and language underutilized in multimodal applications which raises a key question: Is there a plausible way to connect unimodal backbones for zero-shot vision-language tasks? To this end, we propose a novel approach that aligns vision and language modalities using only projection layers on pretrained, frozen unimodal encoders. Our method exploits the high semantic similarity between embedding spaces of well-trained vision and language models. It involves selecting semantically similar encoders in the latent space, curating a concept-rich dataset of image-caption pairs, and training simple MLP projectors. We evaluated our approach on 12 zero-shot classification datasets and 2 image-text retrieval datasets. Our best model, utilizing DINOv2 and All-Roberta-Large text encoder, achieves 76\(\%\) accuracy on ImageNet with a 20-fold reduction in data and 65 fold reduction in compute requirements. The proposed framework enhances the accessibility of model development while enabling flexible adaptation across diverse scenarios, offering an efficient approach to building multimodal models by utilizing existing unimodal architectures. Code and datasets will be released soon., Comment: Preprint, 10 pages; First two authors contributed equally

Iterative self-training, or iterative pseudo-labeling (IPL)--using an improved model from the current iteration to provide pseudo-labels for the next iteration--has proven to be a powerful approach to enhance the quality of speaker representations. Recent applications of IPL in unsupervised speaker recognition start with representations extracted from very elaborate self-supervised methods (e.g., DINO). However, training such strong self-supervised models is not straightforward (they require hyper-parameters tuning and may not generalize to out-of-domain data) and, moreover, may not be needed at all. To this end, we show the simple, well-studied, and established i-vector generative model is enough to bootstrap the IPL process for unsupervised learning of speaker representations. We also systematically study the impact of other components on the IPL process, which includes the initial model, the encoder, augmentations, the number of clusters, and the clustering algorithm. Remarkably, we find that even with a simple and significantly weaker initial model like i-vector, IPL can still achieve speaker verification performance that rivals state-of-the-art methods., Comment: Submitted to ICASSP 2025

Speech foundation models, such as HuBERT and its variants, are pre-trained on large amounts of unlabeled speech for various downstream tasks. These models use a masked prediction objective, where the model learns to predict information about masked input segments from the unmasked context. The choice of prediction targets in this framework can influence performance on downstream tasks. For example, targets that encode prosody are beneficial for speaker-related tasks, while targets that encode phonetics are more suited for content-related tasks. Additionally, prediction targets can vary in the level of detail they encode; targets that encode fine-grained acoustic details are beneficial for denoising tasks, while targets that encode higher-level abstractions are more suited for content-related tasks. Despite the importance of prediction targets, the design choices that affect them have not been thoroughly studied. This work explores the design choices and their impact on downstream task performance. Our results indicate that the commonly used design choices for HuBERT can be suboptimal. We propose novel approaches to create more informative prediction targets and demonstrate their effectiveness through improvements across various downstream tasks., Comment: Submitted to ICASSP 2025

This research presents a model that accurately represents the motions of gaseous stars We employ the Navier-Stokes-Poisson system to transform compressible Euler equations into non-compressible ones by combining quasineutral and inviscid conditions. We intend to put Onsager's hypothesis to the test using the Leray-alpha gaseous star model. This conjecture connects energy conservation with the regularity of weak solutions in the Euler equations. The model used in this work functions as an inviscid regularization of the Euler equations. It technically converges to the Euler equations as the regularization length scale $\alpha$ approaches $0^{+}$.

This work investigates the existence and uniqueness of local weak solutions for the d-dimensional $(d \geq 2)$ fractional magnetic B\'enard system without thermal diffusion, integrating the B\'enard equation and MHD system. For $\kappa = 0$ and $1 \leq \alpha=\beta < 1 + \dfrac{d}{4}$, we establish that any starting conditions $(u_{0},B_{0})\in B_{2,1}^{1+\frac{d}{2}-2\alpha}(\mathbb{R}^{d})$ and $\theta_{0}\in B_{2,1}^{1+\frac{d}{2}-\alpha}(\mathbb{R}^{d})$.

Autonomous agent applications powered by large language models (LLMs) have recently risen to prominence as effective tools for addressing complex real-world tasks. At their core, agentic workflows rely on LLMs to plan and execute the use of tools and external Application Programming Interfaces (APIs) in sequence to arrive at the answer to a user's request. Various benchmarks and leaderboards have emerged to evaluate an LLM's capabilities for tool and API use; however, most of these evaluations only track single or multiple isolated API calling capabilities. In this paper, we present NESTFUL, a benchmark to evaluate LLMs on nested sequences of API calls, i.e., sequences where the output of one API call is passed as input to a subsequent call. NESTFUL has a total of 300 human annotated samples divided into two types - executable and non-executable. The executable samples are curated manually by crawling Rapid-APIs whereas the non-executable samples are hand picked by human annotators from data synthetically generated using an LLM. We evaluate state-of-the-art LLMs with function calling abilities on NESTFUL. Our results show that most models do not perform well on nested APIs in NESTFUL as compared to their performance on the simpler problem settings available in existing benchmarks.

The automated and timely conversion of cybersecurity information from unstructured online sources, such as blogs and articles to more formal representations has become a necessity for many applications in the domain nowadays. Named Entity Recognition (NER) is one of the early phases towards this goal. It involves the detection of the relevant domain entities, such as product, version, attack name, etc. in technical documents. Although generally considered a simple task in the information extraction field, it is quite challenging in some domains like cybersecurity because of the complex structure of its entities. The state of the art methods require time-consuming and labor intensive feature engineering that describes the properties of the entities, their context, domain knowledge, and linguistic characteristics. The model demonstrated in this paper is domain independent and does not rely on any features specific to the entities in the cybersecurity domain, hence does not require expert knowledge to perform feature engineering. The method used relies on a type of recurrent neural networks called Long Short-Term Memory (LSTM) and the Conditional Random Fields (CRFs) method. The results we obtained showed that this method outperforms the state of the art methods given an annotated corpus of a decent size.

In recent years, numerous large-scale cyberattacks have exploited Internet of Things (IoT) devices, a phenomenon that is expected to escalate with the continuing proliferation of IoT technology. Despite considerable efforts in attack detection, intrusion detection systems remain mostly reactive, responding to specific patterns or observed anomalies. This work proposes a proactive approach to anticipate and mitigate malicious activities before they cause damage. This paper proposes a novel network intrusion prediction framework that combines Large Language Models (LLMs) with Long Short Term Memory (LSTM) networks. The framework incorporates two LLMs in a feedback loop: a fine-tuned Generative Pre-trained Transformer (GPT) model for predicting network traffic and a fine-tuned Bidirectional Encoder Representations from Transformers (BERT) for evaluating the predicted traffic. The LSTM classifier model then identifies malicious packets among these predictions. Our framework, evaluated on the CICIoT2023 IoT attack dataset, demonstrates a significant improvement in predictive capabilities, achieving an overall accuracy of 98%, offering a robust solution to IoT cybersecurity challenges.

Real-time visual feedback from catheterization analysis is crucial for enhancing surgical safety and efficiency during endovascular interventions. However, existing datasets are often limited to specific tasks, small scale, and lack the comprehensive annotations necessary for broader endovascular intervention understanding. To tackle these limitations, we introduce CathAction, a large-scale dataset for catheterization understanding. Our CathAction dataset encompasses approximately 500,000 annotated frames for catheterization action understanding and collision detection, and 25,000 ground truth masks for catheter and guidewire segmentation. For each task, we benchmark recent related works in the field. We further discuss the challenges of endovascular intentions compared to traditional computer vision tasks and point out open research questions. We hope that CathAction will facilitate the development of endovascular intervention understanding methods that can be applied to real-world applications. The dataset is available at https://airvlab.github.io/cathaction/., Comment: 10 pages. Webpage: https://airvlab.github.io/cathaction/

This paper establishes the fundamental limits of a multi-access system where multiple users communicate to a legitimate receiver in presence of an external warden. Only a specific subset of the users, called covert users, needs their communication to remain undetected to the warden, while the remaining non-covert users have no such constraint. The fundamental limits show a tradeoff between the different rates that are simultaneously achievable at the various users in function of the secret-key rates that the different users share with the legitimate receiver. Interestingly, the presence of the non-covert users can enhance the capacities of the covert users, especially under stringent secret-key budgets. Our findings underscore the essential requirement of employing a multiplexing (coded time-sharing) strategy to exhaust the fundamental region of all rates that are simultaneously achievable at the covert and the non-covert users. As a side-product of our results, we also establish the covert-capacity secret-key tradeoff for standard single-user and multi-access covert communication systems (without non-covert users), i.e., the largest covert rates that are achievable under given secret-key rate budgets. Previous works had only established the minimum secret-key rates required at largest covert rates., Comment: Submitted for review to IEEE Transactions on Information Theory

Assuming a weighted Nash type inequality for the generator $-A$ of a Markov semigroup, we prove a weighted Nash type inequality for its fractional power and deduce non-uniform bounds on the transition kernel corresponding to the Markov semigroup generated by $-A^\alpha$.

End-to-end wireless communication is new concept expected to be widely used in the physical layer of future wireless communication systems (6G). It involves the substitution of transmitter and receiver block components with a deep neural network (DNN), aiming to enhance the efficiency of data transmission. This will ensure the transition of autonomous vehicles (AVs) from self-autonomy to full collaborative autonomy, that requires vehicular connectivity with high data throughput and minimal latency. In this article, we propose a novel neural network receiver based on transformer architecture, named TransRx, designed for vehicle-to-network (V2N) communications. The TransRx system replaces conventional receiver block components in traditional communication setups. We evaluated our proposed system across various scenarios using different parameter sets and velocities ranging from 0 to 120 km/h over Urban Macro-cell (UMa) channels as defined by 3GPP. The results demonstrate that TransRx outperforms the state-of-the-art systems, achieving a 3.5dB improvement in convergence to low Bit Error Rate (BER) compared to convolutional neural network (CNN)-based neural receivers, and an 8dB improvement compared to traditional baseline receiver configurations. Furthermore, our proposed system exhibits robust generalization capabilities, making it suitable for deployment in large-scale environments.

The introduction of 6G technology into the Internet of Vehicles (IoV) promises to revolutionize connectivity with ultra-high data rates and seamless network coverage. However, this technological leap also brings significant challenges, particularly for the dynamic and diverse IoV landscape, which must meet the rigorous reliability and security requirements of 6G networks. Furthermore, integrating 6G will likely increase the IoV's susceptibility to a spectrum of emerging cyber threats. Therefore, it is crucial for security mechanisms to dynamically adapt and learn new attack patterns, keeping pace with the rapid evolution and diversification of these threats - a capability currently lacking in existing systems. This paper presents a novel intrusion detection system leveraging the paradigm of life-long (or continual) learning. Our methodology combines class-incremental learning with federated learning, an approach ideally suited to the distributed nature of the IoV. This strategy effectively harnesses the collective intelligence of Connected and Automated Vehicles (CAVs) and edge computing capabilities to train the detection system. To the best of our knowledge, this study is the first to synergize class-incremental learning with federated learning specifically for cyber attack detection. Through comprehensive experiments on a recent network traffic dataset, our system has exhibited a robust adaptability in learning new cyber attack patterns, while effectively retaining knowledge of previously encountered ones. Additionally, it has proven to maintain high accuracy and a low false positive rate.

In the rapidly evolving landscape of cyber threats targeting the Internet of Things (IoT) ecosystem, and in light of the surge in botnet-driven Distributed Denial of Service (DDoS) and brute force attacks, this study focuses on the early detection of IoT bots. It specifically addresses the detection of stealth bot communication that precedes and orchestrates attacks. This study proposes a comprehensive methodology for analyzing IoT network traffic, including considerations for both unidirectional and bidirectional flow, as well as packet formats. It explores a wide spectrum of network features critical for representing network traffic and characterizing benign IoT traffic patterns effectively. Moreover, it delves into the modeling of traffic using various semi-supervised learning techniques. Through extensive experimentation with the IoT-23 dataset - a comprehensive collection featuring diverse botnet types and traffic scenarios - we have demonstrated the feasibility of detecting botnet traffic corresponding to different operations and types of bots, specifically focusing on stealth command and control (C2) communications. The results obtained have demonstrated the feasibility of identifying C2 communication with a 100% success rate through packet-based methods and 94% via flow based approaches, with a false positive rate of 1.53%.

Incorporating rate splitting multiple access (RSMA) into integrated sensing and communication (ISAC) presents a significant security challenge, particularly in scenarios where the location of a potential eavesdropper (Eve) is unidentified. Splitting users' messages into common and private streams exposes them to eavesdropping, with the common stream dedicated for sensing and accessible to multiple users. In response to this challenge, this paper proposes a novel approach that leverages active reconfigurable intelligent surface (RIS) aided beamforming and artificial noise (AN) to enhance the security of RSMA-enabled ISAC. Specifically, we first derive the ergodic private secrecy rate (EPSR) based on mathematical approximation of the average Eve channel gain. An optimization problem is then formulated to maximize the minimum EPSR, while satisfying the minimum required thresholds on ergodic common secrecy rate, radar sensing and RIS power budget. To address this non-convex problem, a novel optimization strategy is developed, whereby we alternatively optimize the transmit beamforming matrix for the common and private streams, rate splitting, AN, RIS reflection coefficient matrix, and radar receive beamformer. Successive convex approximation (SCA) and Majorization-Minimization (MM) are employed to convexify the beamforming and RIS sub-problems. Simulations are conducted to showcase the effectiveness of the proposed framework against established benchmarks.

We introduce Falcon2-11B, a foundation model trained on over five trillion tokens, and its multimodal counterpart, Falcon2-11B-vlm, which is a vision-to-text model. We report our findings during the training of the Falcon2-11B which follows a multi-stage approach where the early stages are distinguished by their context length and a final stage where we use a curated, high-quality dataset. Additionally, we report the effect of doubling the batch size mid-training and how training loss spikes are affected by the learning rate. The downstream performance of the foundation model is evaluated on established benchmarks, including multilingual and code datasets. The foundation model shows strong generalization across all the tasks which makes it suitable for downstream finetuning use cases. For the vision language model, we report the performance on several benchmarks and show that our model achieves a higher average score compared to open-source models of similar size. The model weights and code of both Falcon2-11B and Falcon2-11B-vlm are made available under a permissive license.

The application of Riemannian geometry in the decoding of brain-computer interfaces (BCIs) has swiftly garnered attention because of its straightforwardness, precision, and resilience, along with its aptitude for transfer learning, which has been demonstrated through significant achievements in global BCI competitions. This paper presents a comprehensive review of recent advancements in the integration of deep learning with Riemannian geometry to enhance EEG signal decoding in BCIs. Our review updates the findings since the last major review in 2017, comparing modern approaches that utilize deep learning to improve the handling of non-Euclidean data structures inherent in EEG signals. We discuss how these approaches not only tackle the traditional challenges of noise sensitivity, non-stationarity, and lengthy calibration times but also introduce novel classification frameworks and signal processing techniques to reduce these limitations significantly. Furthermore, we identify current shortcomings and propose future research directions in manifold learning and riemannian-based classification, focusing on practical implementations and theoretical expansions, such as feature tracking on manifolds, multitask learning, feature extraction, and transfer learning. This review aims to bridge the gap between theoretical research and practical, real-world applications, making sophisticated mathematical approaches accessible and actionable for BCI enhancements.

The interpretability of random forest (RF) models is a research topic of growing interest in the machine learning (ML) community. In the state of the art, RF is considered a powerful learning ensemble given its predictive performance, flexibility, and ease of use. Furthermore, the inner process of the RF model is understandable because it uses an intuitive and intelligible approach for building the RF decision tree ensemble. However, the RF resulting model is regarded as a "black box" because of its numerous deep decision trees. Gaining visibility over the entire process that induces the final decisions by exploring each decision tree is complicated, if not impossible. This complexity limits the acceptance and implementation of RF models in several fields of application. Several papers have tackled the interpretation of RF models. This paper aims to provide an extensive review of methods used in the literature to interpret RF resulting models. We have analyzed these methods and classified them based on different axes. Although this review is not exhaustive, it provides a taxonomy of various techniques that should guide users in choosing the most appropriate tools for interpreting RF models, depending on the interpretability aspects sought. It should also be valuable for researchers who aim to focus their work on the interpretability of RF or ML black boxes in general., Comment: 37 pages, 11 figures

Intrusion Detection Systems (IDS) play a crucial role in ensuring the security of computer networks. Machine learning has emerged as a popular approach for intrusion detection due to its ability to analyze and detect patterns in large volumes of data. However, current ML-based IDS solutions often struggle to keep pace with the ever-changing nature of attack patterns and the emergence of new attack types. Additionally, these solutions face challenges related to class imbalance, where the number of instances belonging to different classes (normal and intrusions) is significantly imbalanced, which hinders their ability to effectively detect minor classes. In this paper, we propose a novel multi-agent reinforcement learning (RL) architecture, enabling automatic, efficient, and robust network intrusion detection. To enhance the capabilities of the proposed model, we have improved the DQN algorithm by implementing the weighted mean square loss function and employing cost-sensitive learning techniques. Our solution introduces a resilient architecture designed to accommodate the addition of new attacks and effectively adapt to changes in existing attack patterns. Experimental results realized using CIC-IDS-2017 dataset, demonstrate that our approach can effectively handle the class imbalance problem and provide a fine grained classification of attacks with a very low false positive rate. In comparison to the current state-of-the-art works, our solution demonstrates a significant superiority in both detection rate and false positive rate.

The Advanced Metering Infrastructure (AMI) is one of the key components of the smart grid. It provides interactive services for managing billing and electricity consumption, but it also introduces new vectors for cyberattacks. Although, the devastating and severe impact of power overloading cyberattacks on smart grid AMI, few researches in the literature have addressed them. In the present paper, we propose a two-level anomaly detection framework based on regression decision trees. The introduced detection approach leverages the regularity and predictability of energy consumption to build reference consumption patterns for the whole neighborhood and each household within it. Using a reference consumption pattern enables detecting power overloading cyberattacks regardless of the attacker's strategy as they cause a drastic change in the consumption pattern. The continuous two-level monitoring of energy consumption load allows efficient and early detection of cyberattacks. We carried out an extensive experiment on a real-world publicly available energy consumption dataset of 500 customers in Ireland. We extracted, from the raw data, the relevant attributes for training the energy consumption patterns. The evaluation shows that our approach achieves a high detection rate, a low false alarm rate, and superior performances compared to existing solutions.

Deploying Connected and Automated Vehicles (CAVs) on top of 5G and Beyond networks (5GB) makes them vulnerable to increasing vectors of security and privacy attacks. In this context, a wide range of advanced machine/deep learning based solutions have been designed to accurately detect security attacks. Specifically, supervised learning techniques have been widely applied to train attack detection models. However, the main limitation of such solutions is their inability to detect attacks different from those seen during the training phase, or new attacks, also called zero-day attacks. Moreover, training the detection model requires significant data collection and labeling, which increases the communication overhead, and raises privacy concerns. To address the aforementioned limits, we propose in this paper a novel detection mechanism that leverages the ability of the deep auto-encoder method to detect attacks relying only on the benign network traffic pattern. Using federated learning, the proposed intrusion detection system can be trained with large and diverse benign network traffic, while preserving the CAVs privacy, and minimizing the communication overhead. The in-depth experiment on a recent network traffic dataset shows that the proposed system achieved a high detection rate while minimizing the false positive rate, and the detection delay.

The Internet of Vehicles (IoV) is a crucial technology for Intelligent Transportation Systems (ITS) that integrates vehicles with the Internet and other entities. The emergence of 5G and the forthcoming 6G networks presents an enormous potential to transform the IoV by enabling ultra-reliable, low-latency, and high-bandwidth communications. Nevertheless, as connectivity expands, cybersecurity threats have become a significant concern. The issue has been further exacerbated by the rising number of zero-day (0-day) attacks, which can exploit unknown vulnerabilities and bypass existing Intrusion Detection Systems (IDSs). In this paper, we propose Zero-X, an innovative security framework that effectively detects both 0-day and N-day attacks. The framework achieves this by combining deep neural networks with Open-Set Recognition (OSR). Our approach introduces a novel scheme that uses blockchain technology to facilitate trusted and decentralized federated learning (FL) of the ZeroX framework. This scheme also prioritizes privacy preservation, enabling both CAVs and Security Operation Centers (SOCs) to contribute their unique knowledge while protecting the privacy of their sensitive data. To the best of our knowledge, this is the first work to leverage OSR in combination with privacy-preserving FL to identify both 0-day and N-day attacks in the realm of IoV. The in-depth experiments on two recent network traffic datasets show that the proposed framework achieved a high detection rate while minimizing the false positive rate. Comparison with related work showed that the Zero-X framework outperforms existing solutions.

Connected cars offer safety and efficiency for both individuals and fleets of private vehicles and public transportation companies. However, equipping vehicles with information and communication technologies raises privacy and security concerns, which significantly threaten the user's data and life. Using bot malware, a hacker may compromise a vehicle and control it remotely, for instance, he can disable breaks or start the engine remotely. In this paper, besides in-vehicle attacks existing in the literature, we consider new zeroday bot malware attacks specific to the vehicular context, WSMP-Flood, and Geo-WSMP Flood. Then, we propose AntibotV, a multilevel behaviour-based framework for vehicular botnets detection in vehicular networks. The proposed framework combines two main modules for attack detection, the first one monitors the vehicle's activity at the network level, whereas the second one monitors the in-vehicle activity. The two intrusion detection modules have been trained on a historical network and in-vehicle communication using decision tree algorithms. The experimental results showed that the proposed framework outperforms existing solutions, it achieves a detection rate higher than 97% and a false positive rate lower than 0.14%.

Enterprises have a growing need to identify relevant tables in data lakes; e.g. tables that are unionable, joinable, or subsets of each other. Tabular neural models can be helpful for such data discovery tasks. In this paper, we present TabSketchFM, a neural tabular model for data discovery over data lakes. First, we propose novel pre-training: a sketch-based approach to enhance the effectiveness of data discovery in neural tabular models. Second, we finetune the pretrained model for identifying unionable, joinable, and subset table pairs and show significant improvement over previous tabular neural models. Third, we present a detailed ablation study to highlight which sketches are crucial for which tasks. Fourth, we use these finetuned models to perform table search; i.e., given a query table, find other tables in a corpus that are unionable, joinable, or that are subsets of the query. Our results demonstrate significant improvements in F1 scores for search compared to state-of-the-art techniques. Finally, we show significant transfer across datasets and tasks establishing that our model can generalize across different tasks and over different data lakes.

Large language models (LLMs) have recently shown tremendous promise in serving as the backbone to agentic systems, as demonstrated by their performance in multi-faceted, challenging benchmarks like SWE-Bench and Agent-Bench. However, to realize the true potential of LLMs as autonomous agents, they must learn to identify, call, and interact with external tools and application program interfaces (APIs) to complete complex tasks. These tasks together are termed function calling. Endowing LLMs with function calling abilities leads to a myriad of advantages, such as access to current and domain-specific information in databases and knowledge sources, and the ability to outsource tasks that can be reliably performed by tools, e.g., a Python interpreter or calculator. While there has been significant progress in function calling with LLMs, there is still a dearth of open models that perform on par with proprietary LLMs like GPT, Claude, and Gemini. Therefore, in this work, we introduce the GRANITE-20B-FUNCTIONCALLING model under an Apache 2.0 license. The model is trained using a multi-task training approach on seven fundamental tasks encompassed in function calling, those being Nested Function Calling, Function Chaining, Parallel Functions, Function Name Detection, Parameter-Value Pair Detection, Next-Best Function, and Response Generation. We present a comprehensive evaluation on multiple out-of-domain datasets comparing GRANITE-20B-FUNCTIONCALLING to more than 15 other best proprietary and open models. GRANITE-20B-FUNCTIONCALLING provides the best performance among all open models on the Berkeley Function Calling Leaderboard and fourth overall. As a result of the diverse tasks and datasets used for training our model, we show that GRANITE-20B-FUNCTIONCALLING has better generalizability on multiple tasks in seven different evaluation datasets.

The fundamental limit of Semantic Communications (joint source-channel coding) is established when the transmission needs to be kept covert from an external warden. We derive information-theoretic achievability and matching converse results and we show that source and channel coding separation holds for this setup. Furthermore, we show through an experimental setup that one can train a deep neural network to achieve covert semantic communication for the classification task. Our numerical experiments confirm our theoretical findings, which indicate that for reliable joint source-channel coding the number of transmitted source symbols can only scale as the square-root of the number of channel uses., Comment: Submitted to GLOBECOM'2024 for review

In this paper, we propose the "Kinetics Observer", a novel estimator addressing the challenge of state estimation for legged robots using proprioceptive sensors (encoders, IMU and force/torque sensors). Based on a Multiplicative Extended Kalman Filter, the Kinetics Observer allows the real-time simultaneous estimation of contact and perturbation forces, and of the robot's kinematics, which are accurate enough to perform proprioceptive odometry. Thanks to a visco-elastic model of the contacts linking their kinematics to the ones of the centroid of the robot, the Kinetics Observer ensures a tight coupling between the whole-body kinematics and dynamics of the robot. This coupling entails a redundancy of the measurements that enhances the robustness and the accuracy of the estimation. This estimator was tested on two humanoid robots performing long distance walking on even terrain and non-coplanar multi-contact locomotion.

Although Large Language Models (LLMs) have shown promise for human-like conversations, they are primarily pre-trained on text data. Incorporating audio or video improves performance, but collecting large-scale multimodal data and pre-training multimodal LLMs is challenging. To this end, we propose a Fusion Low Rank Adaptation (FLoRA) technique that efficiently adapts a pre-trained unimodal LLM to consume new, previously unseen modalities via low rank adaptation. For device-directed speech detection, using FLoRA, the multimodal LLM achieves 22% relative reduction in equal error rate (EER) over the text-only approach and attains performance parity with its full fine-tuning (FFT) counterpart while needing to tune only a fraction of its parameters. Furthermore, with the newly introduced adapter dropout, FLoRA is robust to missing data, improving over FFT by 20% lower EER and 56% lower false accept rate. The proposed approach scales well for model sizes from 16M to 3B parameters., Comment: Accepted at Interspeech 2024

Voice activity detection (VAD) is a critical component in various applications such as speech recognition, speech enhancement, and hands-free communication systems. With the increasing demand for personalized and context-aware technologies, the need for effective personalized VAD systems has become paramount. In this paper, we present a comparative analysis of Personalized Voice Activity Detection (PVAD) systems to assess their real-world effectiveness. We introduce a comprehensive approach to assess PVAD systems, incorporating various performance metrics such as frame-level and utterance-level error rates, detection latency and accuracy, alongside user-level analysis. Through extensive experimentation and evaluation, we provide a thorough understanding of the strengths and limitations of various PVAD variants. This paper advances the understanding of PVAD technology by offering insights into its efficacy and viability in practical applications using a comprehensive set of metrics.

This work presents an extensive and detailed study on Audio-Visual Speech Recognition (AVSR) for five widely spoken languages: Chinese, Spanish, English, Arabic, and French. We have collected large-scale datasets for each language except for English, and have engaged in the training of supervised learning models. Our model, ViSpeR, is trained in a multi-lingual setting, resulting in competitive performance on newly established benchmarks for each language. The datasets and models are released to the community with an aim to serve as a foundation for triggering and feeding further research work and exploration on Audio-Visual Speech Recognition, an increasingly important area of research. Code available at \href{https://github.com/YasserdahouML/visper}{https://github.com/YasserdahouML/visper}.

Using Unmanned Aerial Vehicles (UAVs) in Search and rescue operations (SAR) to navigate challenging terrain while maintaining reliable communication with the cellular network is a promising approach. This paper suggests a novel technique employing a reinforcement learning multi Q-learning algorithm to optimize UAV connectivity in such scenarios. We introduce a Strategic Planning Agent for efficient path planning and collision awareness and a Real-time Adaptive Agent to maintain optimal connection with the cellular base station. The agents trained in a simulated environment using multi Q-learning, encouraging them to learn from experience and adjust their decision-making to diverse terrain complexities and communication scenarios. Evaluation results reveal the significance of the approach, highlighting successful navigation in environments with varying obstacle densities and the ability to perform optimal connectivity using different frequency bands. This work paves the way for enhanced UAV autonomy and enhanced communication reliability in search and rescue operations.

Locating an object in a sequence of frames, given its appearance in the first frame of the sequence, is a hard problem that involves many stages. Usually, state-of-the-art methods focus on bringing novel ideas in the visual encoding or relational modelling phases. However, in this work, we show that bounding box regression from learned joint search and template features is of high importance as well. While previous methods relied heavily on well-learned features representing interactions between search and template, we hypothesize that the receptive field of the input convolutional bounding box network plays an important role in accurately determining the object location. To this end, we introduce two novel bounding box regression networks: inception and deformable. Experiments and ablation studies show that our inception module installed on the recent ODTrack outperforms the latter on three benchmarks: the GOT-10k, the UAV123 and the OTB2015.

Single object tracking is a vital task of many applications in critical fields. However, it is still considered one of the most challenging vision tasks. In recent years, computer vision, especially object tracking, witnessed the introduction or adoption of many novel techniques, setting new fronts for performance. In this survey, we visit some of the cutting-edge techniques in vision, such as Sequence Models, Generative Models, Self-supervised Learning, Unsupervised Learning, Reinforcement Learning, Meta-Learning, Continual Learning, and Domain Adaptation, focusing on their application in single object tracking. We propose a novel categorization of single object tracking methods based on novel techniques and trends. Also, we conduct a comparative analysis of the performance reported by the methods presented on popular tracking benchmarks. Moreover, we analyze the pros and cons of the presented approaches and present a guide for non-traditional techniques in single object tracking. Finally, we suggest potential avenues for future research in single-object tracking.

The Schr\"odinger equation for diatomic molecules in deSitter and anti-deSitter spaces is studied using the extended uncertainty principle formulation. The equations are solved by the Nikiforov-Uvarov method for both the Kratzer potential and the pseudoharmonic oscillator. The energy eigenvalues of the system have been derived analytically, and the exact expressions of the eigenfunctions are provided in terms of Romanovski and Jacobi polynomials. The impact of the spatial deformation parameter on the bound states is also examined, with experimental results used to establish an upper limit for this parameter., Comment: 13 pages, 7 figures and 4 tables

Large Language Models (LLMs) trained on code are revolutionizing the software development process. Increasingly, code LLMs are being integrated into software development environments to improve the productivity of human programmers, and LLM-based agents are beginning to show promise for handling complex tasks autonomously. Realizing the full potential of code LLMs requires a wide range of capabilities, including code generation, fixing bugs, explaining and documenting code, maintaining repositories, and more. In this work, we introduce the Granite series of decoder-only code models for code generative tasks, trained with code written in 116 programming languages. The Granite Code models family consists of models ranging in size from 3 to 34 billion parameters, suitable for applications ranging from complex application modernization tasks to on-device memory-constrained use cases. Evaluation on a comprehensive set of tasks demonstrates that Granite Code models consistently reaches state-of-the-art performance among available open-source code LLMs. The Granite Code model family was optimized for enterprise software development workflows and performs well across a range of coding tasks (e.g. code generation, fixing and explanation), making it a versatile all around code model. We release all our Granite Code models under an Apache 2.0 license for both research and commercial use., Comment: Corresponding Authors: Rameswar Panda, Ruchir Puri; Equal Contributors: Mayank Mishra, Matt Stallone, Gaoyuan Zhang

In this paper, we characterize the fundamental limits of a communication system with three users (i.e., three transmitters) and a single receiver where communication from two covert users must remain undetectable to an external warden. Our results show a tradeoff between the highest rates that are simultaneously achievable for the three users. They further show that the presence of a non-covert user in the system can enhance the capacities of the covert users under stringent secret-key constraints. To derive our fundamental limits, we provide an information-theoretic converse proof and present a coding scheme that achieves the performance of our converse result. Our coding scheme is based on multiplexing different code phases, which seems to be essential to exhaust the entire tradeoff region between the rates at the covert and the two non-covert users. This property is reminiscent of the setup with multiple non-covert users, where multiplexing is also required to exhaust the entire rate-region., Comment: To be presented at ICC2024

In this paper, we first propose a coupled numerical model of unsaturated flow in soils and plant root water uptake. The Richards equation and different formulations are used in the developed numerical model to describe infiltration in root zone and to investigate the impact of the plant root on the distribution of soil moisture. The Kirchhoff transformed Richards equation is used and the Gardner model is considered for capillary pressure. In our approach, we employ a meshless method based on localized radial basis functions (LRBF) to solve the resulting system of equations. The LRBF approach is an accurate and computationally efficient method that does not require mesh generation and is flexible in addressing high-dimensional problems with complex geometries. Furthermore, this method leads to a sparse matrix system, which avoids ill-conditioning issues. We implement the coupled numerical model of infiltration and plant root water uptake for one, two, and three-dimensional soils. Numerical experiments are performed using nontrivial analytical solutions and available experimental data to validate the coupled numerical model. The numerical results demonstrate the performance and ability of the proposed numerical method to predict soil moisture dynamics in root zone.

Modeling unsaturated flow through soils with water uptake by plan root has many applications in agriculture and water resources management. In this study, our aim is to develop efficient numerical techniques for solving the Richards equation with a sink term due to plant root water uptake. The Feddes model is used for water absorption by plant roots, and the van-Genuchten model is employed for capillary pressure. We introduce a numerical approach that combines the localized exponential radial basis function (EXP-RBF) method for space and the second-order backward differentiation formula (BDF2) for temporal discretization. The localized RBF methods eliminate the need for mesh generation and avoid ill-conditioning problems. This approach yields a sparse matrix for the global system, optimizing memory usage and computational time. The proposed implicit EXP-RBF techniques have advantages in terms of accuracy and computational efficiency thanks to the use of BDF2 and the localized RBF method. Modified Picards iteration method for the mixed form of the Richards equation is employed to linearize the system. Various numerical experiments are conducted to validate the proposed numerical model of infiltration with plant root water absorption. The obtained results conclusively demonstrate the effectiveness of the proposed numerical model in accurately predicting soil moisture dynamics under water uptake by plant roots. The proposed numerical techniques can be incorporated in the numerical models where unsaturated flows and water uptake by plant roots are involved such as in hydrology, agriculture, and water management.

In this paper, we examine the stability of a heat-conducting copper rod and a magnetizable piezoelectric beam, with fractional damping affecting the longitudinal displacement of the piezoelectric material's centerline. We establish a polynomial stability result that is dependent on the order of the fractional derivative., Comment: arXiv admin note: text overlap with arXiv:2204.00283, arXiv:2208.12012, arXiv:2102.07538 by other authors

In this study, a novel semi-implicit second-order temporal scheme combined with the finite element method for space discretization is proposed to solve the coupled system of infiltration and solute transport in unsaturated porous media. The Richards equation is used to describe unsaturated flow, while the advection-dispersion equation (ADE) is used for modeling solute transport. The proposed approach is used to linearize the system of equations in time, eliminating the need of iterative processes. A free parameter is introduced to ensure the stability of the scheme. Numerical tests are conducted to analyze the accuracy of the proposed method in comparison with a family of second-order iterative schemes. The proposed numerical technique based on the optimal free parameter is accurate and performs better in terms of efficiency since it offers a considerable gain in computational time compared to the other methods. For reliability and effectiveness evaluation of the developed semi-implicit scheme, four showcase scenarios are used. The first two numerical tests focus on modeling water flow in heterogeneous soil and transient flow in variably saturated zones. The last numerical tests are carried out to simulate the salt and nitrate transport through unsaturated soils. The simulation results are compared with reference solutions and laboratory data, and demonstrate the effectiveness of the proposed scheme in simulating infiltration and solute transport through unsaturated soils.

We present the first CCD observations of an eclipsing binary, ZTFJ015003.88+534734.1, which is a member in the open star cluster UBC 188. The observations were taken by the 1.88 m telescope at the Kottamia Astronomical Observatory (KAO) in SDSS griz bands. The latest version of the Wilson- Devinney (W-D) code was employed for photometric analysis and light curve modeling of the eclipsing binary. The results indicate that the binary system is in an over-contact configuration. The mass of the primary star (M1) is determined to be 1.293 Msun, and the mass of the secondary star (M2) is directly derived from the system's estimated mass ratio (q= M2/M1) as 0.340 times the solar mass (Msun). We investigated the color-magnitude diagram and the membership probability of the open cluster UBC 188 using the Gaia DR3 data. We determined the membership probability of the eclipsing binary ZTFJ015003.88+534734.1 using the pyUPMASK algorithm and found that its membership probability is one., Comment: 25 pages, 9 figures, 5 tables, accepted in New Astronomy

Random Forest (RF) is well-known as an efficient ensemble learning method in terms of predictive performance. It is also considered a Black Box because of its hundreds of deep decision trees. This lack of interpretability can be a real drawback for acceptance of RF models in several real-world applications, especially those affecting one's lives, such as in healthcare, security, and law. In this work, we present Forest-ORE, a method that makes RF interpretable via an optimized rule ensemble (ORE) for local and global interpretation. Unlike other rule-based approaches aiming at interpreting the RF model, this method simultaneously considers several parameters that influence the choice of an interpretable rule ensemble. Existing methods often prioritize predictive performance over interpretability coverage and do not provide information about existing overlaps or interactions between rules. Forest-ORE uses a mixed-integer optimization program to build an ORE that considers the trade-off between predictive performance, interpretability coverage, and model size (size of the rule ensemble, rule lengths, and rule overlaps). In addition to providing an ORE competitive in predictive performance with RF, this method enriches the ORE through other rules that afford complementary information. It also enables monitoring of the rule selection process and delivers various metrics that can be used to generate a graphical representation of the final model. This framework is illustrated through an example, and its robustness is assessed through 36 benchmark datasets. A comparative analysis of well-known methods shows that Forest-ORE provides an excellent trade-off between predictive performance, interpretability coverage, and model size., Comment: 48 pages, 11 figures

Quantum steering refers to the apparent possibility of exploiting nonseparable quantum correlations to remotely influence the quantum state of an observer via local measurements. Different from entanglement and Bell nonlocality, quantum steering exhibits an inherent asymmetric property, which makes it relevant for many asymmetric quantum information processing tasks. Here, we study Gaussian quantum steering between two mechanical modes in an optomechanical ring cavity. Using experimentally feasible parameters, we show that the state of the two considered modes can exhibit two-way steering and even one-way steering. Instead of using unbalanced losses or noises, we propose a simple practical way to control the direction of one-way steering. A comparative study between the steering and entanglement of the studied modes shows that both steering and entanglement undergo a sudden death-like behavior. In particular, steering is found more fragile against thermal noise remaining constantly upper bounded by entanglement. The proposed scheme may be meaningful for one-sided device-independent quantum key distribution, where the security of such protocol depends fundamentally on the direction of steering.

This research will present a hybrid approach to accelerate convergence in a second order optimization. An online finite difference approximation of the diagonal Hessian matrix will be introduced, along with fuzzy inferencing of several hyperparameters. Competitive results have been achieved

Cardiovascular diseases (CVDs) and congenital heart diseases (CHD) pose significant global health challenges. Fluoroscopy-guided endovascular interventions, though effective, are accompanied by ionizing radiation concerns, especially in pediatric cases. Magnetic resonance imaging (MRI) emerges as a radiation-free alternative, offering superior soft tissue visualization and functional insights. However, the lack of compatible instruments remains a hurdle. We present two novel catheter systems, a tendon-driven steerable catheter and an active tracking Tiger-shaped catheter, fabricated using a unique fiber drawing technique. These catheters, showcasing mechanical properties similar to commercial counterparts, have undergone rigorous in-vitro and in-vivo testing, yielding promising outcomes. This innovative approach has the potential to streamline medical device development, thus enhancing patient care in MR-guided interventions.

There is a growing need for Large Language Models (LLMs) to effectively use tools and external Application Programming Interfaces (APIs) to plan and complete tasks. As such, there is tremendous interest in methods that can acquire sufficient quantities of train and test data that involve calls to tools / APIs. Two lines of research have emerged as the predominant strategies for addressing this challenge. The first has focused on synthetic data generation techniques, while the second has involved curating task-adjacent datasets which can be transformed into API / Tool-based tasks. In this paper, we focus on the task of identifying, curating, and transforming existing datasets and, in turn, introduce API-BLEND, a large corpora for training and systematic testing of tool-augmented LLMs. The datasets mimic real-world scenarios involving API-tasks such as API / tool detection, slot filling, and sequencing of the detected APIs. We demonstrate the utility of the API-BLEND dataset for both training and benchmarking purposes., Comment: Accepted at ACL'24-main conference

Lipreading involves using visual data to recognize spoken words by analyzing the movements of the lips and surrounding area. It is a hot research topic with many potential applications, such as human-machine interaction and enhancing audio speech recognition. Recent deep-learning based works aim to integrate visual features extracted from the mouth region with landmark points on the lip contours. However, employing a simple combination method such as concatenation may not be the most effective approach to get the optimal feature vector. To address this challenge, firstly, we propose a cross-attention fusion-based approach for large lexicon Arabic vocabulary to predict spoken words in videos. Our method leverages the power of cross-attention networks to efficiently integrate visual and geometric features computed on the mouth region. Secondly, we introduce the first large-scale Lipreading in the Wild for Arabic (LRW-AR) dataset containing 20,000 videos for 100-word classes, uttered by 36 speakers. The experimental results obtained on LRW-AR and ArabicVisual databases showed the effectiveness and robustness of the proposed approach in recognizing Arabic words. Our work provides insights into the feasibility and effectiveness of applying lipreading techniques to the Arabic language, opening doors for further research in this field. Link to the project page: https://crns-smartvision.github.io/lrwar, Comment: submitted for review