Search

Distributed quantum networks are not merely information conduits but intricate systems that embody the principles of quantum mechanics. In our study, we examine the underlying mechanisms of quantum connectivity within a distributed framework by exploring phenomena such as superposition and entanglement and their influence on information propagation. We investigate how these fundamental quantum effects interact with routing strategies that, while inspired by classical methods, must contend with quantum decoherence and measurement uncertainties. By simulating distributed networks of 10, 20, 50 and 100 nodes, we assess the performance of routing mechanisms through metrics that reflect both quantum fidelity and operational efficiency. Our findings reveal that the quantum coherence inherent in entangled states can enhance routing fidelity under specific conditions, yet also introduce challenges such as increased computational overhead and sensitivity to network scale. This work bridges the gap between the underlying principles of quantum systems and practical routing implementations, offering new insights into the design of robust distributed quantum networks.

Efficient exploration is essential for intelligent systems interacting with their environment, but existing language models often fall short in scenarios that require strategic information gathering. In this paper, we present PAPRIKA, a fine-tuning approach that enables language models to develop general decision-making capabilities that are not confined to particular environments. By training on synthetic interaction data from different tasks that require diverse strategies, PAPRIKA teaches models to explore and adapt their behavior on a new task based on environment feedback in-context without more gradient updates. Experimental results show that models fine-tuned with PAPRIKA can effectively transfer their learned decision-making capabilities to entirely unseen tasks without additional training. Unlike traditional training, our approach's primary bottleneck lies in sampling useful interaction data instead of model updates. To improve sample efficiency, we propose a curriculum learning strategy that prioritizes sampling trajectories from tasks with high learning potential. These results suggest a promising path towards AI systems that can autonomously solve novel sequential decision-making problems that require interactions with the external world., Comment: Project Website: https://paprika-llm.github.io

Sensor attacks compromise the reliability of cyber-physical systems (CPSs) by altering sensor outputs with the objective of leading the system to unsafe system states. This paper studies a probabilistic intrusion detection framework based on $\lambda$-sensor-attack detectability ($\lambda$-sa), a formal measure that evaluates the likelihood of a system being under attack based on observed behaviors. Our framework enhances detection by extending its capabilities to identify multiple sensor attack strategies using probabilistic information, which enables the detection of sensor attacks that were undetected by current detection methodologies. We develop a polynomial-time algorithm that verifies $\lambda$-sa detectability by constructing a weighted verifier automaton and solving the shortest path problem. Additionally, we propose a method to determine the maximum detection confidence level ($\lambda$*) achievable by the system, ensuring the highest probability of identifying attack-induced behaviors.

Centralized electronic health record repositories are critical for advancing disease surveillance, public health research, and evidence-based policymaking. However, developing countries face persistent challenges in achieving this due to fragmented healthcare data sources, inconsistent record-keeping practices, and the absence of standardized patient identifiers, limiting reliable record linkage, compromise data interoperability, and limit scalability-obstacles exacerbated by infrastructural constraints and privacy concerns. To address these barriers, this study proposes a scalable, privacy-preserving clinical data warehouse, NCDW, designed for heterogeneous EHR integration in resource-limited settings and tested with 1.16 million clinical records. The framework incorporates a wrapper-based data acquisition layer for secure, automated ingestion of multisource health data and introduces a soundex algorithm to resolve patient identity mismatches in the absence of unique IDs. A modular data mart is designed for disease-specific analytics, demonstrated through a dengue fever case study in Bangladesh, integrating clinical, demographic, and environmental data for outbreak prediction and resource planning. Quantitative assessment of the data mart underscores its utility in strengthening national decision-support systems, highlighting the model's adaptability for infectious disease management. Comparative evaluation of database technologies reveals NoSQL outperforms relational SQL by 40-69% in complex query processing, while system load estimates validate the architecture's capacity to manage 19 million daily records (34TB over 5 years). The framework can be adapted to various healthcare settings across developing nations by modifying the ingestion layer to accommodate standards like ICD-11 and HL7 FHIR, facilitating interoperability for managing infectious diseases (i.e., COVID, tuberculosis).

In this paper, we propose an efficient implementation of deep policy gradient method (PGM) for optimal control problems in continuous time. The proposed method has the ability to manage the allocation of computational resources, number of trajectories, and complexity of architecture of the neural network. This is, in particular, important for continuous-time problems that require a fine time discretization. Each step of this method focuses on a different time scale and learns a policy, modeled by a neural network, for a discretized optimal control problem. The first step has the coarsest time discretization. As we proceed to other steps, the time discretization becomes finer. The optimal trained policy in each step is also used to provide data for the next step. We accompany the multi-scale deep PGM with a theoretical result on allocation of computational resources to obtain a targeted efficiency and test our methods on the linear-quadratic stochastic optimal control problem., Comment: 20 pages, 4 figures

Human intelligence involves metacognitive abilities like self-regulation, recognizing limitations, and seeking assistance only when needed. While LLM Agents excel in many domains, they often lack this awareness. Overconfident agents risk catastrophic failures, while those that seek help excessively hinder efficiency. A key challenge is enabling agents with a limited intervention budget $C$ is to decide when to request assistance. In this paper, we propose an offline framework that trains a "helper" policy to request interventions, such as more powerful models or test-time compute, by combining LLM-based process reward models (PRMs) with tabular reinforcement learning. Using state transitions collected offline, we score optimal intervention timing with PRMs and train the helper model on these labeled trajectories. This offline approach significantly reduces costly intervention calls during training. Furthermore, the integration of PRMs with tabular RL enhances robustness to off-policy data while avoiding the inefficiencies of deep RL. We empirically find that our method delivers optimal helper behavior.

The rapid advancement of large language models has opened new avenues for automating complex problem-solving tasks such as algorithmic coding and competitive programming. This paper introduces a novel evaluation technique, LLM-ProS, to assess the performance of state-of-the-art LLMs on International Collegiate Programming Contest (ICPC) problems. Using a curated dataset of 166 World Finals problems from 2011 to 2024, we benchmark the models' reasoning, accuracy, and efficiency. We evaluate the five models-GPT-4o, Mistral Large, Llama-3.1-405B, and the o1 family, consisting of o1-mini and o1-preview, across critical metrics like correctness, resource utilization, and response calibration. Our results reveal significant differences in the models' abilities to generalize, adapt, and solve novel problems. We also investigated the impact of training methodologies, dataset contamination, and chain-of-thought reasoning on model performance. The findings provide new insights into optimizing LLMs for algorithmic tasks, highlighting both strengths and limitations of current models., Comment: To be published in LLM4Code 2025 workshop proceedings

Metasurface-based optical components are becoming increasingly important due to their unparalleled ability to shape and manipulate electromagnetic waves across a wide range of frequencies, from microwave and terahertz to visible light. In particular, planar meta-mirrors with high efficiency and broad operational bandwidth are expected to play a significant role in shaping channel characteristics for 6G communication links due to their reduced SWaP (size, weight, and power) and integrability. However, achieving large fractional bandwidth and high efficiency simultaneously using metasurface-based mirrors or reflectors have been challenging. In this work, highly efficient broadband focusing meta-mirror has been designed, fabricated and experimentally demonstrated. The device operates in the D-Band (110-170 GHz) and beyond (up to 200 GHz), and exhibits unmatched performance in terms of combined fractional bandwidth (58\%) and energy efficiency, exceeding 53\% over the entire bandwidth with a peak of 77.4\% at frequency 135 GHz. The high performance and advantageous design tradeoffs of this meta-mirror are discussed in terms of the principles of low Fresnel number optics.

We present an end-to-end workflow for superconducting qubit readout that embeds co-designed Neural Networks (NNs) into the Quantum Instrumentation Control Kit (QICK). Capitalizing on the custom firmware and software of the QICK platform, which is built on Xilinx RFSoC FPGAs, we aim to leverage machine learning (ML) to address critical challenges in qubit readout accuracy and scalability. The workflow utilizes the hls4ml package and employs quantization-aware training to translate ML models into hardware-efficient FPGA implementations via user-friendly Python APIs. We experimentally demonstrate the design, optimization, and integration of an ML algorithm for single transmon qubit readout, achieving 96% single-shot fidelity with a latency of 32ns and less than 16% FPGA look-up table resource utilization. Our results offer the community an accessible workflow to advance ML-driven readout and adaptive control in quantum information processing applications.

Many real-world decision-making problems are combinatorial in nature, where states (e.g., surrounding traffic of a self-driving car) can be seen as a combination of basic elements (e.g., pedestrians, trees, and other cars). Due to combinatorial complexity, observing all combinations of basic elements in the training set is infeasible, which leads to an essential yet understudied problem of zero-shot generalization to states that are unseen combinations of previously seen elements. In this work, we first formalize this problem and then demonstrate how existing value-based reinforcement learning (RL) algorithms struggle due to unreliable value predictions in unseen states. We argue that this problem cannot be addressed with exploration alone, but requires more expressive and generalizable models. We demonstrate that behavior cloning with a conditioned diffusion model trained on expert trajectory generalizes better to states formed by new combinations of seen elements than traditional RL methods. Through experiments in maze, driving, and multiagent environments, we show that conditioned diffusion models outperform traditional RL techniques and highlight the broad applicability of our problem formulation.

Materials exhibiting a martensitic phase transition are essential for applications in shape memory alloys, actuators and sensors. Hexamethylbenzene (HMB) has long been considered as a classical example of ferroelastic organic crystals since Mnyukh's pioneering work in 1970s. However, the atomistic mechanism underlying this phase transition has never been clarified. In this work, we present a direct molecular dynamics simulation to investigate the phase transition mechanism in HMB. For the first time, we report a simulation results that can accurately reproduce both the transition temperature and hysteresis loop observed in previous experimental studies. By analyzing the MD trajectories, the potential energy surface, we identified that a low-barrier atomic sliding mode along the close-packed (11$\overline{1}$) plane of the low-temperature phase is the key to trigger the phase transition at the critical temperature window. This is further confirmed by the observed continuous softening of shear modulus around the transition window. Our results demonstrate that the integration of various atomistic modeling techniques can provide invaluable insight into the martensitic phase transition mechanisms in organic crystals and guide the development of new organic martensites., Comment: 9 pages, 9 figures

We present Fanar, a platform for Arabic-centric multimodal generative AI systems, that supports language, speech and image generation tasks. At the heart of Fanar are Fanar Star and Fanar Prime, two highly capable Arabic Large Language Models (LLMs) that are best in the class on well established benchmarks for similar sized models. Fanar Star is a 7B (billion) parameter model that was trained from scratch on nearly 1 trillion clean and deduplicated Arabic, English and Code tokens. Fanar Prime is a 9B parameter model continually trained on the Gemma-2 9B base model on the same 1 trillion token set. Both models are concurrently deployed and designed to address different types of prompts transparently routed through a custom-built orchestrator. The Fanar platform provides many other capabilities including a customized Islamic Retrieval Augmented Generation (RAG) system for handling religious prompts, a Recency RAG for summarizing information about current or recent events that have occurred after the pre-training data cut-off date. The platform provides additional cognitive capabilities including in-house bilingual speech recognition that supports multiple Arabic dialects, voice and image generation that is fine-tuned to better reflect regional characteristics. Finally, Fanar provides an attribution service that can be used to verify the authenticity of fact based generated content. The design, development, and implementation of Fanar was entirely undertaken at Hamad Bin Khalifa University's Qatar Computing Research Institute (QCRI) and was sponsored by Qatar's Ministry of Communications and Information Technology to enable sovereign AI technology development.

Quantum security improves cryptographic protocols by applying quantum mechanics principles, assuring resistance to both quantum and conventional computer attacks. This work addresses these issues by integrating Quantum Key Distribution (QKD) utilizing the E91 method with Multi-Layer Chaotic Encryption, which employs a variety of patterns to detect eavesdropping, resulting in a highly secure image-transmission architecture. The method leverages entropy calculations to determine the unpredictability and integrity of encrypted and decrypted pictures, guaranteeing strong security. Extensive statistical scenarios illustrate the framework's effectiveness in image encryption while preserving high entropy and sensitivity to the original visuals. The findings indicate significant improvement in encryption and decryption performance, demonstrating the framework's potential as a robust response to weaknesses introduced by advances in quantum computing. Several metrics, such as Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index (SSIM), Normalized Cross-Correlation (NCC), Bit Error Rate (BER), entropy values for original, encrypted, and decrypted images, and the correlation between original and decrypted images, validate the framework's effectiveness. The combination of QKD with Multi-Layer Chaotic Encryption provides a scalable and resilient technique to secure image communication. As quantum computing advances, this framework offers a future-proof approach for defining secure communication protocols in crucial sectors such as medical treatment, forensic computing, and national security, where information confidentiality is valuable., Comment: no

We construct new classes of the dynamical black hole solutions in five or higher dimensional Einstein-Maxwell theory, coupled to a dilaton field, in the presence of arbitrary cosmological constant. The dilaton field interacts non-trivially with the Maxwell field, as well as the cosmological constant, with two arbitrary coupling constants. The solutions are non-stationary, and almost conformally regular everywhere. To construct the solutions, we use the four-dimensional Bianchi type IX geometry, as the base space. We find three different classes of solutions, based on the values of the coupling constants. We notice that our solutions could be asymptotically de-Sitter, anti-de-Sitter or flat. We find the relevant quantities of the solutions, and discuss the properties of the solutions., Comment: 26 pages, 13 figures

The Yamada-Watanabe theory provides a robust framework for understanding stochastic equations driven by Wiener processes. Despite its comprehensive treatment in the literature, the applicability of the theory to SPDEs driven by Poisson random measures or, more generally, L\'evy processes remains significantly less explored, with only a handful of results addressing this context. In this work, we leverage a result by Kurtz to demonstrate that the existence of a martingale solution combined with pathwise uniqueness implies the existence of a unique strong solution for SPDEs driven by both a Wiener process and a Poisson random measure. Our discussion is set within the variational framework, where the SPDE under consideration may be nonlinear. This work is influenced by earlier research conducted by the second author alongside de Bouard and Ondrej\'at.

We find the non-extremal charged rotating black holes in quadratic $f(T)$ gravity are holographically dual to two different hidden conformal field theories. The two conformal field theories can be merged to find a very general hidden conformal field theory, which is generated by the $SL(2,\mathbb{Z})$ modular group. We also carry out the the calculation to the extremal limit of the black holes, and find the corresponding dual quantities. Contrary to the existence of two different dual conformal field theories for the extremal charged rotating black holes in Einstein gravity, we find only one dual theory exists for the extremal charged rotating black holes in quadratic $f(T)$ gravity., Comment: 31 pages, 8 figures

Deploying local AI models, such as Large Language Models (LLMs), to edge devices can substantially enhance devices' independent capabilities, alleviate the server's burden, and lower the response time. Owing to these tremendous potentials, many big tech companies have released several lightweight Small Language Models (SLMs) to bridge this gap. However, we still have huge motivations to deploy more powerful (LLMs) AI models on edge devices and enhance their smartness level. Unlike the conventional approaches for AI model compression, we investigate activation sparsity. The activation sparsity method is orthogonal and combinable with existing techniques to maximize the compression rate while maintaining great accuracy. LLMs' Feed-Forward Network (FFN) components, which typically comprise a large proportion of parameters (around 2/3), ensure that our FFN optimizations would have a better chance of achieving effective compression. Moreover, our findings are beneficial to general LLMs and are not restricted to ReLU-based models. This work systematically investigates the tradeoff between enforcing activation sparsity and perplexity (accuracy) on state-of-the-art LLMs. Our empirical analysis demonstrates that we can obtain around 50% of main memory and computing reductions for critical FFN components with negligible accuracy degradation. This extra 50% sparsity does not naturally exist in the current LLMs, which require tuning LLMs' activation outputs by injecting zero-enforcing thresholds. To obtain the benefits of activation sparsity, we provide a guideline for the system architect for LLM prediction and prefetching. The success prediction allows the system to prefetch the necessary weights while omitting the inactive ones and their successors, therefore lowering cache and memory pollution and reducing LLM execution time on resource-constrained edge devices., Comment: pp. 1-9, doi: 10.1109/IPCCC59868.2024.10850382. keywords: {Accuracy;Prefetching;Large language models;Computational modeling;Companies;Transformers;User experience;Time factors;Tuning;Guidelines;Large Language Models (LLMs);AI Compression;Activation Sparsity;Edge LLM}

A Content Delivery Network (CDN) is a powerful system of distributed caching servers that aims to accelerate content delivery, like high-definition video, IoT applications, and ultra-low-latency services, efficiently and with fast velocity. This has become of paramount importance in the post-pandemic era. Challenges arise when exponential content volume growth and scalability across different geographic locations are required. This paper investigates data-driven evaluations of CDN algorithms in dynamic server selection for latency reduction, bandwidth throttling for efficient resource management, real-time Round Trip Time analysis for adaptive routing, and programmatic network delay simulation to emulate various conditions. Key performance metrics, such as round-trip time (RTT) and CPU usage, are carefully analyzed to evaluate scalability and algorithmic efficiency through two experimental setups: a constrained edge-like local system and a scalable FABRIC testbed. The statistical validation of RTT trends, alongside CPU utilization, is presented in the results. The optimization process reveals significant trade-offs between scalability and resource consumption, providing actionable insights for effectively deploying and enhancing CDN algorithms in edge and distributed computing environments., Comment: 6 Pages, 10 Figures

Tiny machine learning (TinyML) aims to run ML models on small devices and is increasingly favored for its enhanced privacy, reduced latency, and low cost. Recently, the advent of tiny AI accelerators has revolutionized the TinyML field by significantly enhancing hardware processing power. These accelerators, equipped with multiple parallel processors and dedicated per-processor memory instances, offer substantial performance improvements over traditional microcontroller units (MCUs). However, their limited data memory often necessitates downsampling input images, resulting in accuracy degradation. To address this challenge, we propose Data channel EXtension (DEX), a novel approach for efficient CNN execution on tiny AI accelerators. DEX incorporates additional spatial information from original images into input images through patch-wise even sampling and channel-wise stacking, effectively extending data across input channels. By leveraging underutilized processors and data memory for channel extension, DEX facilitates parallel execution without increasing inference latency. Our evaluation with four models and four datasets on tiny AI accelerators demonstrates that this simple idea improves accuracy on average by 3.5%p while keeping the inference latency the same on the AI accelerator. The source code is available at https://github.com/Nokia-Bell-Labs/data-channel-extension., Comment: Accepted to NeurIPS 2024

Recent progresses in magnetoionics offer exciting potentials to leverage its non-linearity, short-term memory, and energy-efficiency to uniquely advance the field of physical reservoir computing. In this work, we experimentally demonstrate the classification of temporal data using a magneto-ionic (MI) heterostructure. The device was specifically engineered to induce non-linear ion migration dynamics, which in turn imparted non-linearity and short-term memory (STM) to the magnetization. These capabilities, key features for enabling reservoir computing, were investigated, and the role of the ion migration mechanism, along with its history-dependent influence on STM, was explained. These attributes were utilized to distinguish between sine and square waveforms within a randomly distributed set of pulses. Additionally, two important performance metrics, short-term memory and parity check capacity (PC), were quantified, yielding promising values of 1.44 and 2, respectively, comparable to those of other state-of-the-art reservoirs. Our work paves the way for exploiting the relaxation dynamics of solid-state magneto-ionic platforms and developing energy-efficient magneto-ionic reservoir computing devices.

The analysis of retinal images for the diagnosis of various diseases is one of the emerging areas of research. Recently, the research direction has been inclined towards investigating several changes in retinal blood vessels in subjects with many neurological disorders, including dementia. This research focuses on detecting diseases early by improving the performance of models for segmentation of retinal vessels with fewer parameters, which reduces computational costs and supports faster processing. This paper presents a novel lightweight encoder-decoder model that segments retinal vessels to improve the efficiency of disease detection. It incorporates multi-scale convolutional blocks in the encoder to accurately identify vessels of various sizes and thicknesses. The bottleneck of the model integrates the Focal Modulation Attention and Spatial Feature Refinement Blocks to refine and enhance essential features for efficient segmentation. The decoder upsamples features and integrates them with the corresponding feature in the encoder using skip connections and the spatial feature refinement block at every upsampling stage to enhance feature representation at various scales. The estimated computation complexity of our proposed model is around 29.60 GFLOP with 0.71 million parameters and 2.74 MB of memory size, and it is evaluated using public datasets, that is, DRIVE, CHASE\_DB, and STARE. It outperforms existing models with dice scores of 86.44\%, 84.22\%, and 87.88\%, respectively.

Strong lensing directly probes the internal structure of the lensing galaxies. In this paper, we investigate the relation between the internal structure of massive elliptical galaxies and their environment using a sample of 15 strong lensing systems. We performed lens modeling for them using Lenstronomy and constrained the mass and light distributions of the deflector galaxies. We adopt the local galaxy density as a metric for the environment and test our results against several alternative definitions of it. We robustly find that the centroid offset between the mass and light is not correlated with the local galaxy density. This result supports using centroid offsets as a probe of dark matter theories since the environment's impact on it can be treated as negligible. Although we find a strong correlation between the position angle offset and the standard definition of the local galaxy density, consistent with previous studies, the correlation becomes weaker for alternative definitions of the local galaxy density. This result weakens the support for interpreting the position angle misalignment as having originated from interaction with the environment. Furthermore, we find the 'residual shear' magnitude in the lens model to be uncorrelated with the local galaxy density, supporting the interpretation of the residual shear originating, in part, from the inadequacy in modeling the angular structure of the lensing galaxy and not solely from the structures present in the environment or along the line of sight., Comment: 15 pages, 9 figures, 3 tables. Submitted to A&A

Software security is crucial in any field where breaches can exploit sensitive data, and lead to financial losses. As a result, vulnerability detection becomes an essential part of the software development process. One of the key steps in maintaining software integrity is identifying vulnerabilities in the source code before deployment. A security breach like CWE-476, which stands for NULL pointer dereferences (NPD), is crucial because it can cause software crashes, unpredictable behavior, and security vulnerabilities. In this scientific era, there are several vulnerability checkers, where, previous tools often fall short in analyzing specific feature connections of the source code, which weakens the tools in real-world scenarios. In this study, we propose another novel approach using a fine-tuned Large Language Model (LLM) termed "DeLLNeuN". This model leverages the advantage of various layers to reduce both overfitting and non-linearity, enhancing its performance and reliability. Additionally, this method provides dropout and dimensionality reduction to help streamline the model, making it faster and more efficient. Our model showed 87% accuracy with 88% precision using the Draper VDISC dataset. As software becomes more complex and cyber threats continuously evolve, the need for proactive security measures will keep growing. In this particular case, the proposed model looks promising to use as an early vulnerability checker in software development.

Extractive Text Summarization is the process of selecting the most representative parts of a larger text without losing any key information. Recent attempts at extractive text summarization in Bengali, either relied on statistical techniques like TF-IDF or used naive sentence similarity measures like the word averaging technique. All of these strategies suffer from expressing semantic relationships correctly. Here, we propose a novel Word pair-based Gaussian Sentence Similarity (WGSS) algorithm for calculating the semantic relation between two sentences. WGSS takes the geometric means of individual Gaussian similarity values of word embedding vectors to get the semantic relationship between sentences. It compares two sentences on a word-to-word basis which rectifies the sentence representation problem faced by the word averaging method. The summarization process extracts key sentences by grouping semantically similar sentences into clusters using the Spectral Clustering algorithm. After clustering, we use TF-IDF ranking to pick the best sentence from each cluster. The proposed method is validated using four different datasets, and it outperformed other recent models by 43.2% on average ROUGE scores (ranging from 2.5% to 95.4%). It is also experimented on other low-resource languages i.e. Turkish, Marathi, and Hindi language, where we find that the proposed method performs as similar as Bengali for these languages. In addition, a new high-quality Bengali dataset is curated which contains 250 articles and a pair of summaries for each of them. We believe this research is a crucial addition to Bengali Natural Language Processing (NLP) research and it can easily be extended into other low-resource languages. We made the implementation of the proposed model and data public on https://github.com/FMOpee/WGSS.

In software, a vulnerability is a defect in a program that attackers might utilize to acquire unauthorized access, alter system functions, and acquire information. These vulnerabilities arise from programming faults, design flaws, incorrect setups, and a lack of security protective measures. To mitigate these vulnerabilities, regular software upgrades, code reviews, safe development techniques, and the use of security tools to find and fix problems have been important. Several ways have been delivered in recent studies to address difficulties related to software vulnerabilities. However, previous approaches have significant limitations, notably in feature embedding and precisely recognizing specific vulnerabilities. To overcome these drawbacks, we present CodeGraphNet, an experimental method that combines GraphCodeBERT and Graph Convolutional Network (GCN) approaches, where, CodeGraphNet reveals data in a high-dimensional vector space, with comparable or related properties grouped closer together. This method captures intricate relationships between features, providing for more exact identification and separation of vulnerabilities. Using this feature embedding approach, we employed four machine learning models, applying both independent testing and 10-fold cross-validation. The DeepTree model, which is a hybrid of a Decision Tree and a Neural Network, outperforms state-of-the-art approaches. In additional validation, we evaluated our model using feature embeddings from LSA, GloVe, FastText, CodeBERT and GraphCodeBERT, and found that the CodeGraphNet method presented improved vulnerability identification with 98% of accuracy. Our model was tested on a real-time dataset to determine its capacity to handle real-world data and to focus on defect localization, which might influence future studies.

Modern deep policy gradient methods achieve effective performance on simulated robotic tasks, but they all require large replay buffers or expensive batch updates, or both, making them incompatible for real systems with resource-limited computers. We show that these methods fail catastrophically when limited to small replay buffers or during incremental learning, where updates only use the most recent sample without batch updates or a replay buffer. We propose a novel incremental deep policy gradient method -- Action Value Gradient (AVG) and a set of normalization and scaling techniques to address the challenges of instability in incremental learning. On robotic simulation benchmarks, we show that AVG is the only incremental method that learns effectively, often achieving final performance comparable to batch policy gradient methods. This advancement enabled us to show for the first time effective deep reinforcement learning with real robots using only incremental updates, employing a robotic manipulator and a mobile robot., Comment: In The Thirty-eighth Annual Conference on Neural Information Processing Systems. Source code at https://github.com/gauthamvasan/avg and companion video at https://youtu.be/cwwuN6Hyew0

Sensing human motions through Inertial Measurement Units (IMUs) embedded in personal devices has enabled significant applications in health and wellness. Labeled IMU data is scarce, however, unlabeled or weakly labeled IMU data can be used to model human motions. For video or text modalities, the "pretrain and adapt" approach utilizes large volumes of unlabeled or weakly labeled data to build a strong feature extractor, followed by adaptation to specific tasks using limited labeled data. However, pretraining methods are poorly understood for IMU data, and pipelines are rarely evaluated on out-of-domain tasks. We propose PRIMUS: a method for PRetraining IMU encoderS that uses a novel pretraining objective that is empirically validated based on downstream performance on both in-domain and out-of-domain datasets. The PRIMUS objective effectively enhances downstream performance by combining self-supervision, multimodal, and nearest-neighbor supervision. With fewer than 500 labeled samples per class, PRIMUS improves test accuracy by up to 15%, compared to state-of-the-art baselines. To benefit the broader community, we have open-sourced our code at github.com/nokia-bell-labs/pretrained-imu-encoders., Comment: Presented at ICASSP 2025. Also presented under the title "PRIMUS: Pretraining IMU Encoders with Multimodal and Self-Supervised Learning" at NeurIPS 2024 TSALM Workshop (Time Series in the Age of Large Models)

Backscatter Communication (BackCom) technology has emerged as a promising paradigm for the Green Internet of Things (IoT) ecosystem, offering advantages such as low power consumption, cost-effectiveness, and ease of deployment. While traditional BackCom systems, such as RFID technology, have found widespread applications, the advent of ambient backscatter presents new opportunities for expanding applications and enhancing capabilities. Moreover, ongoing standardization efforts are actively focusing on BackCom technologies, positioning them as a potential solution to meet the near-zero power consumption and massive connectivity requirements of next-generation wireless systems. 0G networks have the potential to provide advanced solutions by leveraging BackCom technology to deliver ultra-low-power, ubiquitous connectivity for the expanding IoT ecosystem, supporting billions of devices with minimal energy consumption. This paper investigates the integration of BackCom and 0G networks to enhance the capabilities of traditional BackCom systems and enable Green IoT. We conduct an in-depth analysis of BackCom-enabled 0G networks, exploring their architecture and operational objectives, and also explore the Waste Factor (WF) metric for evaluating energy efficiency and minimizing energy waste within integrated systems. By examining both structural and operational aspects, we demonstrate how this synergy enhances the performance, scalability, and sustainability of next-generation wireless networks. Moreover, we highlight possible applications, open challenges, and future directions, offering valuable insights for guiding future research and practical implementations aimed at achieving large-scale, sustainable IoT deployments.

Here we present a technique for using machine learning (ML) for single-qubit gate synthesis on field programmable logic for a superconducting transmon-based quantum computer based on simulated studies. Our approach is multi-stage. We first bootstrap a model based on simulation with access to the full statevector for measuring gate fidelity. We next present an algorithm, named adapted randomized benchmarking (ARB), for fine-tuning the gate on hardware based on measurements of the devices. We also present techniques for deploying the model on programmable devices with care to reduce the required resources. While the techniques here are applied to a transmon-based computer, many of them are portable to other architectures.

There has been little systematic study on how dialectal differences affect toxicity detection by modern LLMs. Furthermore, although using LLMs as evaluators ("LLM-as-a-judge") is a growing research area, their sensitivity to dialectal nuances is still underexplored and requires more focused attention. In this paper, we address these gaps through a comprehensive toxicity evaluation of LLMs across diverse dialects. We create a multi-dialect dataset through synthetic transformations and human-assisted translations, covering 10 language clusters and 60 varieties. We then evaluated three LLMs on their ability to assess toxicity across multilingual, dialectal, and LLM-human consistency. Our findings show that LLMs are sensitive in handling both multilingual and dialectal variations. However, if we have to rank the consistency, the weakest area is LLM-human agreement, followed by dialectal consistency. Code repository: \url{https://github.com/ffaisal93/dialect_toxicity_llm_judge}

Nowadays, the rapid diffusion of fake news poses a significant problem, as it can spread misinformation and confusion. This paper aims to develop an advanced machine learning solution for detecting fake news articles. Leveraging a comprehensive dataset of news articles, including 23,502 fake news articles and 21,417 accurate news articles, we implemented and evaluated three machine-learning models. Our dataset, curated from diverse sources, provides rich textual content categorized into title, text, subject, and Date features. These features are essential for training robust classification models to distinguish between fake and authentic news articles. The initial model employed a Long Short-Term Memory (LSTM) network, achieving an accuracy of 94%. The second model improved upon this by incorporating additional regularization techniques and fine-tuning hyperparameters, resulting in a 97% accuracy. The final model combined the strengths of previous architectures with advanced optimization strategies, achieving a peak accuracy of 98%. These results demonstrate the effectiveness of our approach in identifying fake news with high precision. Implementing these models showcases significant advancements in natural language processing and machine learning techniques, contributing valuable tools for combating misinformation. Our work highlights the potential for deploying such models in real-world applications, providing a reliable method for automated fake news detection and enhancing the credibility of news dissemination., Comment: 6 pages, 7 figures, 2024 IEEE International Conference on Signal Processing, Information, Communication and Systems (SPICSCON)

In the span of four decades, quantum computation has evolved from an intellectual curiosity to a potentially realizable technology. Today, small-scale demonstrations have become possible for quantum algorithmic primitives on hundreds of physical qubits and proof-of-principle error-correction on a single logical qubit. Nevertheless, despite significant progress and excitement, the path toward a full-stack scalable technology is largely unknown. There are significant outstanding quantum hardware, fabrication, software architecture, and algorithmic challenges that are either unresolved or overlooked. These issues could seriously undermine the arrival of utility-scale quantum computers for the foreseeable future. Here, we provide a comprehensive review of these scaling challenges. We show how the road to scaling could be paved by adopting existing semiconductor technology to build much higher-quality qubits, employing system engineering approaches, and performing distributed quantum computation within heterogeneous high-performance computing infrastructures. These opportunities for research and development could unlock certain promising applications, in particular, efficient quantum simulation/learning of quantum data generated by natural or engineered quantum systems. To estimate the true cost of such promises, we provide a detailed resource and sensitivity analysis for classically hard quantum chemistry calculations on surface-code error-corrected quantum computers given current, target, and desired hardware specifications based on superconducting qubits, accounting for a realistic distribution of errors. Furthermore, we argue that, to tackle industry-scale classical optimization and machine learning problems in a cost-effective manner, heterogeneous quantum-probabilistic computing with custom-designed accelerators should be considered as a complementary path toward scalability., Comment: 76 pages, 46 figures. General revision, added figures, added references, added appendices

Recent advances in image generation, particularly via diffusion models, have led to impressive improvements in image synthesis quality. Despite this, diffusion models are still challenged by model-induced artifacts and limited stability in image fidelity. In this work, we hypothesize that the primary cause of this issue is the improper resampling operation that introduces aliasing in the diffusion model and a careful alias-free resampling dictated by image processing theory can improve the model's performance in image synthesis. We propose the integration of alias-free resampling layers into the UNet architecture of diffusion models without adding extra trainable parameters, thereby maintaining computational efficiency. We then assess whether these theory-driven modifications enhance image quality and rotational equivariance. Our experimental results on benchmark datasets, including CIFAR-10, MNIST, and MNIST-M, reveal consistent gains in image quality, particularly in terms of FID and KID scores. Furthermore, we propose a modified diffusion process that enables user-controlled rotation of generated images without requiring additional training. Our findings highlight the potential of theory-driven enhancements such as alias-free resampling in generative models to improve image quality while maintaining model efficiency and pioneer future research directions to incorporate them into video-generating diffusion models, enabling deeper exploration of the applications of alias-free resampling in generative modeling., Comment: 13 pages, 7 figures

Functional near-infrared spectroscopy (fNIRS) is a non-invasive optical technique that measures brain activity by estimating blood oxygenation using near-infrared light. Traditionally, PsychoPy is used in many studies to send task-specific markers, requiring a separate device to interface with the fNIRS data collection system. In this work, we present a Python-based implementation to send markers directly, eliminating the need for an additional device. This approach allows researchers to run both marker transmission and fNIRS data collection on the same computer, simplifying the setup and enhancing accessibility. This streamlined solution reduces hardware requirements and makes fNIRS studies more efficient.

Developing new machine learning applications often requires the collection of new datasets. However, existing datasets may already contain relevant information to train models for new purposes. We propose SoundCollage: a framework to discover new classes within audio datasets by incorporating (1) an audio pre-processing pipeline to decompose different sounds in audio samples, and (2) an automated model-based annotation mechanism to identify the discovered classes. Furthermore, we introduce the clarity measure to assess the coherence of the discovered classes for better training new downstream applications. Our evaluations show that the accuracy of downstream audio classifiers within discovered class samples and a held-out dataset improves over the baseline by up to 34.7% and 4.5%, respectively. These results highlight the potential of SoundCollage in making datasets reusable by labeling with newly discovered classes. To encourage further research in this area, we open-source our code at https://github.com/nokia-bell-labs/audio-class-discovery., Comment: 5 pages, 2 figures. Accepted in IEEE ICASSP 2025

Photoplethysmography (PPG) is the leading non-invasive technique for monitoring biosignals and cardiovascular health, with widespread adoption in both clinical settings and consumer wearable devices. While machine learning models trained on PPG signals have shown promise, they tend to be task-specific and struggle with generalization. Current research is limited by the use of single-device datasets, insufficient exploration of out-of-domain generalization, and a lack of publicly available models, which hampers reproducibility. To address these limitations, we present PaPaGei, the first open foundation model for PPG signals. The model is pre-trained on over 57,000 hours of data, comprising 20 million unlabeled PPG segments from publicly available datasets. We introduce a novel representation learning approach that leverages domain knowledge of PPG signal morphology across individuals, enabling the capture of richer representations compared to traditional contrastive learning methods. We evaluate PaPaGei against state-of-the-art time-series foundation models and self-supervised learning benchmarks across 20 tasks from 10 diverse datasets, spanning cardiovascular health, sleep disorders, pregnancy monitoring, and wellbeing assessment. Our model demonstrates superior performance, improving classification and regression metrics by 6.3% and 2.9% respectively in at least 14 tasks. Notably, PaPaGei achieves these results while being more data- and parameter-efficient, outperforming models that are 70x larger. Beyond accuracy, we examine model robustness across different skin tones, establishing a benchmark for bias evaluation in future models. PaPaGei can serve as both a feature extractor and an encoder for multimodal models, opening up new opportunities for multimodal health monitoring., Comment: Accepted at ICLR 2025. Improved version with new experiments and results. Code and models: https://github.com/nokia-bell-labs/papagei-foundation-model

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

Traffic forecasting in Intelligent Transportation Systems (ITS) is vital for intelligent traffic prediction. Yet, ITS often relies on data from traffic sensors or vehicle devices, where certain cities might not have all those smart devices or enabling infrastructures. Also, recent studies have employed meta-learning to generalize spatial-temporal traffic networks, utilizing data from multiple cities for effective traffic forecasting for data-scarce target cities. However, collecting data from multiple cities can be costly and time-consuming. To tackle this challenge, we introduce Single Source Meta-Transfer Learning (SSMT) which relies only on a single source city for traffic prediction. Our method harnesses this transferred knowledge to enable few-shot traffic forecasting, particularly when the target city possesses limited data. Specifically, we use memory-augmented attention to store the heterogeneous spatial knowledge from the source city and selectively recall them for the data-scarce target city. We extend the idea of sinusoidal positional encoding to establish meta-learning tasks by leveraging diverse temporal traffic patterns from the source city. Moreover, to capture a more generalized representation of the positions we introduced a meta-positional encoding that learns the most optimal representation of the temporal pattern across all the tasks. We experiment on five real-world benchmark datasets to demonstrate that our method outperforms several existing methods in time series traffic prediction., Comment: ICPR 2024

For specialized domains, there is often not a wealth of data with which to train large machine learning models. In such limited data / compute settings, various methods exist aiming to $\textit{do more with less}$, such as finetuning from a pretrained model, modulating difficulty levels as data are presented to a model (curriculum learning), and considering the role of model type / size. Approaches to efficient $\textit{machine}$ learning also take inspiration from $\textit{human}$ learning by considering use cases where machine learning systems have access to approximately the same number of words experienced by a 13 year old child (100M words). We investigate the role of 3 primary variables in a limited data regime as part of the multimodal track of the BabyLM challenge. We contrast: (i) curriculum learning, (ii), pretraining (with text-only data), (iii) model type. We modulate these variables and assess them on two types of tasks: (a) multimodal (text+image), and (b) unimodal (text-only) tasks. We find that curriculum learning benefits multimodal evaluations over non-curriclum learning models, particularly when combining text-only pretraining. On text-only tasks, curriculum learning appears to help models with smaller trainable parameter counts. We suggest possible reasons based on architectural differences and training designs as to why one might observe such results., Comment: CoNLL BabyLM Challenge 2024 camera ready

The current work is a qualitative study that aims to explore the implementation of Twisted and Coiled Artificial Muscles (TCAMs) for actuating and replicating the bending motion of an octopus-like soft robot arm underwater. Additionally, it investigates the impact of hydrostatic and dynamic forces from steady-state fluid flow on the arm's motion. The artificial muscles are lightweight and low-cost actuators that generate a high power-to-weight ratio, producing tensile force up to 12,600 times their own weight, which is close to the functionality of biological muscles. The "extended" Cosserat theory of rods is employed to formulate a quasi-static continuum model of arm motion, where the arm's cross-section is not only capable of rigid rotation but also deforms within its plane. This planar deformation of the arm cross-section aligns with the biological behavior of the octopus arm, where the stiffness of the hydrostat is directly induced by the incompressibility of the tissues. In line with the main goal, a constitutive model is derived for the material of the octopus arm to capture its characteristic behavior., Comment: 12 pages, Under review at the journal "Robotics Reports"

Recent advancements in speech-language models have yielded significant improvements in speech tokenization and synthesis. However, effectively mapping the complex, multidimensional attributes of speech into discrete tokens remains challenging. This process demands acoustic, semantic, and contextual information for precise speech representations. Existing speech representations generally fall into two categories: acoustic tokens from audio codecs and semantic tokens from speech self-supervised learning models. Although recent efforts have unified acoustic and semantic tokens for improved performance, they overlook the crucial role of contextual representation in comprehensive speech modeling. Our empirical investigations reveal that the absence of contextual representations results in elevated Word Error Rate (WER) and Word Information Lost (WIL) scores in speech transcriptions. To address these limitations, we propose two novel distillation approaches: (1) a language model (LM)-guided distillation method that incorporates contextual information, and (2) a combined LM and self-supervised speech model (SM)-guided distillation technique that effectively distills multimodal representations (acoustic, semantic, and contextual) into a comprehensive speech tokenizer, termed DM-Codec. The DM-Codec architecture adopts a streamlined encoder-decoder framework with a Residual Vector Quantizer (RVQ) and incorporates the LM and SM during the training process. Experiments show DM-Codec significantly outperforms state-of-the-art speech tokenization models, reducing WER by up to 13.46%, WIL by 9.82%, and improving speech quality by 5.84% and intelligibility by 1.85% on the LibriSpeech benchmark dataset. The code, samples, and model checkpoints are available at https://github.com/mubtasimahasan/DM-Codec.

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

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

Sensory earables have evolved from basic audio enhancement devices into sophisticated platforms for clinical-grade health monitoring and wellbeing management. This paper introduces OmniBuds, an advanced sensory earable platform integrating multiple biosensors and onboard computation powered by a machine learning accelerator, all within a real-time operating system (RTOS). The platform's dual-ear symmetric design, equipped with precisely positioned kinetic, acoustic, optical, and thermal sensors, enables highly accurate and real-time physiological assessments. Unlike conventional earables that rely on external data processing, OmniBuds leverage real-time onboard computation to significantly enhance system efficiency, reduce latency, and safeguard privacy by processing data locally. This capability includes executing complex machine learning models directly on the device. We provide a comprehensive analysis of OmniBuds' design, hardware and software architecture demonstrating its capacity for multi-functional applications, accurate and robust tracking of physiological parameters, and advanced human-computer interaction.

Detectors at future high energy colliders will face enormous technical challenges. Disentangling the unprecedented numbers of particles expected in each event will require highly granular silicon pixel detectors with billions of readout channels. With event rates as high as 40 MHz, these detectors will generate petabytes of data per second. To enable discovery within strict bandwidth and latency constraints, future trackers must be capable of fast, power efficient, and radiation hard data-reduction at the source. We are developing a radiation hard readout integrated circuit (ROIC) in 28nm CMOS with on-chip machine learning (ML) for future intelligent pixel detectors. We will show track parameter predictions using a neural network within a single layer of silicon and hardware tests on the first tape-outs produced with TSMC. Preliminary results indicate that reading out featurized clusters from particles above a modest momentum threshold could enable using pixel information at 40 MHz., Comment: Contribution to the 42nd International Conference on High Energy Physics (ICHEP)