Search

GPT-4o is an autoregressive omni model that accepts as input any combination of text, audio, image, and video, and generates any combination of text, audio, and image outputs. It's trained end-to-end across text, vision, and audio, meaning all inputs and outputs are processed by the same neural network. GPT-4o can respond to audio inputs in as little as 232 milliseconds, with an average of 320 milliseconds, which is similar to human response time in conversation. It matches GPT-4 Turbo performance on text in English and code, with significant improvement on text in non-English languages, while also being much faster and 50\% cheaper in the API. GPT-4o is especially better at vision and audio understanding compared to existing models. In line with our commitment to building AI safely and consistent with our voluntary commitments to the White House, we are sharing the GPT-4o System Card, which includes our Preparedness Framework evaluations. In this System Card, we provide a detailed look at GPT-4o's capabilities, limitations, and safety evaluations across multiple categories, focusing on speech-to-speech while also evaluating text and image capabilities, and measures we've implemented to ensure the model is safe and aligned. We also include third-party assessments on dangerous capabilities, as well as discussion of potential societal impacts of GPT-4o's text and vision capabilities.

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.

Background: This study aimed to evaluate and compare the performance of classical machine learning models (CMLs) and large language models (LLMs) in predicting mortality associated with COVID-19 by utilizing a high-dimensional tabular dataset. Materials and Methods: We analyzed data from 9,134 COVID-19 patients collected across four hospitals. Seven CML models, including XGBoost and random forest (RF), were trained and evaluated. The structured data was converted into text for zero-shot classification by eight LLMs, including GPT-4 and Mistral-7b. Additionally, Mistral-7b was fine-tuned using the QLoRA approach to enhance its predictive capabilities. Results: Among the CML models, XGBoost and RF achieved the highest accuracy, with F1 scores of 0.87 for internal validation and 0.83 for external validation. In the LLM category, GPT-4 was the top performer with an F1 score of 0.43. Fine-tuning Mistral-7b significantly improved its recall from 1% to 79%, resulting in an F1 score of 0.74, which was stable during external validation. Conclusion: While LLMs show moderate performance in zero-shot classification, fine-tuning can significantly enhance their effectiveness, potentially aligning them closer to CML models. However, CMLs still outperform LLMs in high-dimensional tabular data tasks., Comment: Code is available at: https://github.com/mohammad-gh009/Large-Language-Models-vs-Classical-Machine-learning and https://github.com/Sdamirsa/Tehran_COVID_Cohort. The datasets are available from the corresponding author on reasonable request (sdamirsa@ymail.com)

A distributed denial-of-service (DDoS) attack is an attempt to produce humongous traffic within a network by overwhelming a targeted server or its neighboring infrastructure with a flood of service requests ceaselessly coming from multiple remotely controlled malware-infected computers or network-connected devices. Thus, exploring DDoS attacks by recognizing their functionalities and differentiating them from normal traffic services are the primary concerns of network security issues particularly for online businesses. In modern networks, most DDoS attacks occur in the network and application layer including HTTP flood, UDP flood, SIDDOS, SMURF, SNMP flood, IP NULL, etc. The goal of this paper is to detect DDoS attacks from all service requests and classify them according to DDoS classes. In this regard, a standard dataset is collected from the internet which contains several network-related attributes and their corresponding DDoS attack class name. Two(2) different machine learning approaches, SVM and Logistic Regression, are implemented in the dataset for detecting and classifying DDoS attacks, and a comparative study is accomplished among them in terms of accuracy, precision, and recall rates. Logistic Regression and SVM both achieve 98.65% classification accuracy which is the highest achieved accuracy among other previous experiments with the same dataset.

We study procurement auctions, where an auctioneer seeks to acquire services from strategic sellers with private costs. The quality of services is measured by a submodular function known to the auctioneer. Our goal is to design computationally efficient procurement auctions that (approximately) maximize the difference between the quality of the acquired services and the total cost of the sellers, while ensuring incentive compatibility (IC), individual rationality (IR) for sellers, and non-negative surplus (NAS) for the auctioneer. Our contributions are twofold: (i) we provide an improved analysis of existing algorithms for non-positive submodular function maximization, and (ii) we design efficient frameworks that transform submodular optimization algorithms into mechanisms that are IC, IR, NAS, and approximation-preserving. These frameworks apply to both the offline setting, where all sellers' bids and services are available simultaneously, and the online setting, where sellers arrive in an adversarial order, requiring the auctioneer to make irrevocable decisions. We also explore whether state-of-the-art submodular optimization algorithms can be converted into descending auctions in adversarial settings, where the schedule of descending prices is determined by an adversary. We show that a submodular optimization algorithm satisfying bi-criteria $(1/2, 1)$-approximation in welfare can be effectively adapted to a descending auction. Additionally, we establish a connection between descending auctions and online submodular optimization. Finally, we demonstrate the practical applications of our frameworks by instantiating them with state-of-the-art submodular optimization algorithms and empirically comparing their welfare performance on publicly available datasets with thousands of sellers.

It is shown that the presence of an interaction between dark energy and dark matter can induce a transmutation of quintessence from thawing to scaling freezing. The analysis is done in a Lagrangian approach where dark energy and dark matter are treated as spin zero fields. Remarkably, with the cosmological constraints on energy densities of dark matter and dark energy, such a transmutation can occur in the late universe and searchable in accumulating cosmological data by DESI and other dark energy-related experiments. We give a fit to the dark energy equation of state which can be used by experimental collaborations to search for this transmutation that can occur at redshifts up to $z=13$ depending on the interaction strength. A transmutation in the range $z=1-3$ is already within reach of current and near future experiments., Comment: 6 pages, 4 figures (main) with supplemental material (7 pages, 5 figures)

This study focuses on recognizing Bangladeshi dialects and converting diverse Bengali accents into standardized formal Bengali speech. Dialects, often referred to as regional languages, are distinctive variations of a language spoken in a particular location and are identified by their phonetics, pronunciations, and lexicon. Subtle changes in pronunciation and intonation are also influenced by geographic location, educational attainment, and socioeconomic status. Dialect standardization is needed to ensure effective communication, educational consistency, access to technology, economic opportunities, and the preservation of linguistic resources while respecting cultural diversity. Being the fifth most spoken language with around 55 distinct dialects spoken by 160 million people, addressing Bangla dialects is crucial for developing inclusive communication tools. However, limited research exists due to a lack of comprehensive datasets and the challenges of handling diverse dialects. With the advancement in multilingual Large Language Models (mLLMs), emerging possibilities have been created to address the challenges of dialectal Automated Speech Recognition (ASR) and Machine Translation (MT). This study presents an end-to-end pipeline for converting dialectal Noakhali speech to standard Bangla speech. This investigation includes constructing a large-scale diverse dataset with dialectal speech signals that tailored the fine-tuning process in ASR and LLM for transcribing the dialect speech to dialect text and translating the dialect text to standard Bangla text. Our experiments demonstrated that fine-tuning the Whisper ASR model achieved a CER of 0.8% and WER of 1.5%, while the BanglaT5 model attained a BLEU score of 41.6% for dialect-to-standard text translation., Comment: Accepted in 2024 IEEE International Conference on Big Data (IEEE BigData)

This study investigates the automation of meta-analysis in scientific documents using large language models (LLMs). Meta-analysis is a robust statistical method that synthesizes the findings of multiple studies support articles to provide a comprehensive understanding. We know that a meta-article provides a structured analysis of several articles. However, conducting meta-analysis by hand is labor-intensive, time-consuming, and susceptible to human error, highlighting the need for automated pipelines to streamline the process. Our research introduces a novel approach that fine-tunes the LLM on extensive scientific datasets to address challenges in big data handling and structured data extraction. We automate and optimize the meta-analysis process by integrating Retrieval Augmented Generation (RAG). Tailored through prompt engineering and a new loss metric, Inverse Cosine Distance (ICD), designed for fine-tuning on large contextual datasets, LLMs efficiently generate structured meta-analysis content. Human evaluation then assesses relevance and provides information on model performance in key metrics. This research demonstrates that fine-tuned models outperform non-fine-tuned models, with fine-tuned LLMs generating 87.6% relevant meta-analysis abstracts. The relevance of the context, based on human evaluation, shows a reduction in irrelevancy from 4.56% to 1.9%. These experiments were conducted in a low-resource environment, highlighting the study's contribution to enhancing the efficiency and reliability of meta-analysis automation., Comment: Accepted in 2024 IEEE International Conference on Big Data (IEEE BigData)

Understanding the interplay of non-Hermiticity and topology is crucial given the intrinsic openness of most natural and engineered systems and it has important ramifications in topological lasers and sensors. Intense efforts have been devoted to unveiling how non-Hermiticity may impact the most significant features of topological systems, but only recently it has been theoretically proposed that topological features could originate solely from the system's non-Hermiticity in photonic systems. In this work, we experimentally demonstrate the appearance of non-Hermitian topology exclusively from loss modulation in a photonic system that is topologically trivial in the absence of loss. We do this by implementing a non-Hermitian generalization of an Aubry-Andre-Harper model with purely imaginary potential in a programmable integrated photonics platform, which allows us to investigate different periodic and quasi-periodic configurations of the model. In both cases, we show the emergence of topological edge modes and explore their resilience to different kinds of disorder. Our work highlights loss engineering as a mechanism to generate topological properties.

As AI systems become integral to critical operations across industries and services, ensuring their reliability and safety is essential. We offer a framework that integrates established reliability and resilience engineering principles into AI systems. By applying traditional metrics such as failure rate and Mean Time Between Failures (MTBF) along with resilience engineering and human reliability analysis, we propose an integrate framework to manage AI system performance, and prevent or efficiently recover from failures. Our work adapts classical engineering methods to AI systems and outlines a research agenda for future technical studies. We apply our framework to a real-world AI system, using system status data from platforms such as openAI, to demonstrate its practical applicability. This framework aligns with emerging global standards and regulatory frameworks, providing a methodology to enhance the trustworthiness of AI systems. Our aim is to guide policy, regulation, and the development of reliable, safe, and adaptable AI technologies capable of consistent performance in real-world environments.

Network function virtualization enables network operators to implement new services through a process called service function chain mapping. The concept of Service Function Chain (SFC) is introduced to provide complex services, which is an ordered set of Network Functions (NF). The network functions of an SFC can be decomposed in several ways into some Virtual Network Functions (VNF). Additionally, the decomposed NFs can be placed (mapped) as VNFs on different machines on the underlying physical infrastructure. Selecting good decompositions and good placements among the possible options greatly affects both costs and service quality metrics. Previous research has addressed NF decomposition and VNF placement as separate problems. However, in this paper, we address both NF decomposition and VNF placement simultaneously as a single problem. Since finding an optimal solution is NP-hard, we have employed heuristic algorithms to solve the problem. Specifically, we have introduced a multiobjective decomposition and mapping VNFs (MODMVNF) method based on the non-dominated sorting genetic multi-objective algorithm (NSGAII) to solve the problem. The goal is to find near-optimal decomposition and mapping on the physical network at the same time to minimize the mapping cost and communication latency of SFC. The comparison of the results of the proposed method with the results obtained by solving ILP formulation of the problem as well as the results obtained from the multi-objective particle swarm algorithm shows the efficiency and effectiveness of the proposed method in terms of cost and communication latency.

We study stationary online bipartite matching, where both types of nodes--offline and online--arrive according to Poisson processes. Offline nodes wait to be matched for some random time, determined by an exponential distribution, while online nodes need to be matched immediately. This model captures scenarios such as deceased organ donation and time-sensitive task assignments, where there is an inflow of patients and workers (offline nodes) with limited patience, while organs and tasks (online nodes) must be assigned upon arrival. We present an efficient online algorithm that achieves a $(1-1/e+\delta)$-approximation to the optimal online policy's reward for a constant $\delta > 0$, simplifying and improving previous work by Aouad and Sarita\c{c} (2022). Our solution combines recent online matching techniques, particularly pivotal sampling, which enables correlated rounding of tighter linear programming approximations, and a greedy-like algorithm. A key technical component is the analysis of a stochastic process that exploits subtle correlations between offline nodes, using renewal theory. A byproduct of our result is an improvement to the best-known competitive ratio--that compares an algorithm's performance to the optimal offline policy--via a $(1-1/\sqrt{e} + \eta)$-competitive algorithm for a universal constant $\eta > 0$, advancing the results of Patel and Wajc (2024).

Over the past decade, Deep Learning (DL) has become an integral part of our daily lives. This surge in DL usage has heightened the need for developing reliable DL software systems. Given that fault localization is a critical task in reliability assessment, researchers have proposed several fault localization techniques for DL-based software, primarily focusing on faults within the DL model. While the DL model is central to DL components, there are other elements that significantly impact the performance of DL components. As a result, fault localization methods that concentrate solely on the DL model overlook a large portion of the system. To address this, we introduce FL4Deep, a system-level fault localization approach considering the entire DL development pipeline to effectively localize faults across the DL-based systems. In an evaluation using 100 faulty DL scripts, FL4Deep outperformed four previous approaches in terms of accuracy for three out of six DL-related faults, including issues related to data (84%), mismatched libraries between training and deployment (100%), and loss function (69%). Additionally, FL4Deep demonstrated superior precision and recall in fault localization for five categories of faults including three mentioned fault types in terms of accuracy, plus insufficient training iteration and activation function.

We introduce a continuous domain for function spaces over topological spaces which are not core-compact. Notable examples of such topological spaces include the real line with the upper limit topology, which is used in solution of initial value problems with temporal discretization, and various infinite dimensional Banach spaces which are ubiquitous in functional analysis and solution of partial differential equations. If a topological space $\mathbb{X}$ is not core-compact and $\mathbb{D}$ is a non-singleton bounded-complete domain, the function space $[\mathbb{X} \to \mathbb{D}]$ is not a continuous domain. To construct a continuous domain, we consider a spectral compactification $\mathbb{Y}$ of $\mathbb{X}$ and relate $[\mathbb{X} \to \mathbb{D}]$ with the continuous domain $[\mathbb{Y} \to \mathbb{D}]$ via a Galois connection. This allows us to perform computations in the native structure $[\mathbb{X} \to \mathbb{D}]$ while computable analysis is performed in the continuous domain $[\mathbb{Y} \to \mathbb{D}]$, with the left and right adjoints used for moving between the two function spaces., Comment: 19 pages, 1 figure, Mathematical Foundations of Programming Semantics (MFPS) 2024

The development of the sixth generation of communication networks (6G) has been gaining momentum over the past years, with a target of being introduced by 2030. Several initiatives worldwide are developing innovative solutions and setting the direction for the key features of these networks. Some common emerging themes are the tight integration of AI, the convergence of multiple access technologies and sustainable operation, aiming to meet stringent performance and societal requirements. To that end, we are introducing REASON - Realising Enabling Architectures and Solutions for Open Networks. The REASON project aims to address technical challenges in future network deployments, such as E2E service orchestration, sustainability, security and trust management, and policy management, utilising AI-native principles, considering multiple access technologies and cloud-native solutions. This paper presents REASON's architecture and the identified requirements for future networks. The architecture is meticulously designed for modularity, interoperability, scalability, simplified troubleshooting, flexibility, and enhanced security, taking into consideration current and future standardisation efforts, and the ease of implementation and training. It is structured into four horizontal layers: Physical Infrastructure, Network Service, Knowledge, and End-User Application, complemented by two vertical layers: Management and Orchestration, and E2E Security. This layered approach ensures a robust, adaptable framework to support the diverse and evolving requirements of 6G networks, fostering innovation and facilitating seamless integration of advanced technologies., Comment: Accepted for publication at IEEE Access

Privacy-preserving and secure data sharing are critical for medical image analysis while maintaining accuracy and minimizing computational overhead are also crucial. Applying existing deep neural networks (DNNs) to encrypted medical data is not always easy and often compromises performance and security. To address these limitations, this research introduces a secure framework consisting of a learnable encryption method based on the block-pixel operation to encrypt the data and subsequently integrate it with the Vision Transformer (ViT). The proposed framework ensures data privacy and security by creating unique scrambling patterns per key, providing robust performance against leading bit attacks and minimum difference attacks., Comment: 2 pages, 2 figures

Automatic brain tissue segmentation from Magnetic Resonance Imaging (MRI) images is vital for accurate diagnosis and further analysis in medical imaging. Despite advancements in segmentation techniques, a comprehensive comparison between traditional statistical methods and modern deep learning approaches using pre-processing techniques like N4 Bias Field Correction and Anisotropic Diffusion remains underexplored. This study provides a comparative analysis of various segmentation models, including Probabilistic ATLAS, U-Net, nnU-Net, and LinkNet, enhanced with these pre-processing techniques to segment brain tissues (white matter (WM), grey matter (GM) and cerebrospinal fluid (CSF)) on the Internet Brain Segmentation Repository (IBSR18) dataset. Our results demonstrate that the 3D nnU-Net model outperforms others, achieving the highest mean Dice Coefficient score (0.937 +- 0.012), while the 2D nnU-Net model recorded the lowest mean Hausdorff Distance (5.005 +- 0.343 mm) and the lowest mean Absolute Volumetric Difference (3.695 +- 2.931 mm) across five unseen test samples. The findings highlight the superiority of nnU-Net models in brain tissue segmentation, particularly when combined with N4 Bias Field Correction and Anisotropic Diffusion pre-processing techniques. Our implemented code can be accessed via GitHub.

In the context of low-resource languages, the Algerian dialect (AD) faces challenges due to the absence of annotated corpora, hindering its effective processing, notably in Machine Learning (ML) applications reliant on corpora for training and assessment. This study outlines the development process of a specialized corpus for Fake News (FN) detection and sentiment analysis (SA) in AD called FASSILA. This corpus comprises 10,087 sentences, encompassing over 19,497 unique words in AD, and addresses the significant lack of linguistic resources in the language and covers seven distinct domains. We propose an annotation scheme for FN detection and SA, detailing the data collection, cleaning, and labelling process. Remarkable Inter-Annotator Agreement indicates that the annotation scheme produces consistent annotations of high quality. Subsequent classification experiments using BERT-based models and ML models are presented, demonstrate promising results and highlight avenues for further research. The dataset is made freely available on GitHub (https://github.com/amincoding/FASSILA) to facilitate future advancements in the field., Comment: 16 pages, 6 Figuers

Axion-like particles may form a network of cosmic strings in the Universe today that can rotate the plane of polarization of cosmic microwave background (CMB) photons. Future CMB observations with improved sensitivity might detect this axion-string-induced birefringence effect, thereby revealing an as-yet unseen constituent of the Universe and offering a new probe of particles and forces that are beyond the Standard Model of Elementary Particle Physics. In this work, we explore how spherical convolutional neural networks (SCNNs) may be used to extract information about the axion string network from simulated birefringence maps. We construct a pipeline to simulate the anisotropic birefringence that would arise from an axion string network, and we train SCNNs to estimate three parameters related to the cosmic string length, the cosmic string abundance, and the axion-photon coupling. Our results demonstrate that neural networks are able to extract information from a birefringence map that is inaccessible with two-point statistics alone (i.e., the angular power spectrum). We also assess the impact of noise on the accuracy of our SCNN estimators, demonstrating that noise at the level anticipated for Stage IV (CMB-S4) measurements would significantly bias parameter estimation for SCNNs trained on noiseless simulated data, and necessitate modeling the noise in the training data., Comment: 23 pages, 7 figures, 4 tables

This document summarizes the discussions which took place during the PITT-PACC Workshop entitled "Non-Standard Cosmological Epochs and Expansion Histories," held in Pittsburgh, Pennsylvania, Sept. 5-7, 2024. Much like the non-standard cosmological epochs that were the subject of these discussions, the format of this workshop was also non-standard. Rather than consisting of a series of talks from participants, with each person presenting their own work, this workshop was instead organized around free-form discussion blocks, with each centered on a different overall theme and guided by a different set of Discussion Leaders. This document is not intended to serve as a comprehensive review of these topics, but rather as an informal record of the discussions that took place during the workshop, in the hope that the content and free-flowing spirit of these discussions may inspire new ideas and research directions., Comment: 33 pages, LaTeX, 2 figures

Advancing the design of robust hearing aid (HA) voice control is crucial to increase the HA use rate among hard of hearing people as well as to improve HA users' experience. In this work, we contribute towards this goal by, first, presenting a novel HA speech dataset consisting of noisy own voice captured by 2 behind-the-ear (BTE) and 1 in-ear-canal (IEC) microphones. Second, we provide baseline HA voice control results from the evaluation of light, state-of-the-art keyword spotting models utilizing different combinations of HA microphone signals. Experimental results show the benefits of exploiting bandwidth-limited bone-conducted speech (BCS) from the IEC microphone to achieve noise-robust HA voice control. Furthermore, results also demonstrate that voice control performance can be boosted by assisting BCS by the broader-bandwidth BTE microphone signals. Aiming at setting a baseline upon which the scientific community can continue to progress, the HA noisy speech dataset has been made publicly available., Comment: Submitted to IEEE Signal Processing Letters

Text-conditioned generation models are commonly evaluated based on the quality of the generated data and its alignment with the input text prompt. On the other hand, several applications of prompt-based generative models require sufficient diversity in the generated data to ensure the models' capability of generating image and video samples possessing a variety of features. However, most existing diversity metrics are designed for unconditional generative models, and thus cannot distinguish the diversity arising from variations in text prompts and that contributed by the generative model itself. In this work, our goal is to quantify the prompt-induced and model-induced diversity in samples generated by prompt-based models. We propose an information-theoretic approach for internal diversity quantification, where we decompose the kernel-based entropy $H(X)$ of the generated data $X$ into the sum of the conditional entropy $H(X|T)$, given text variable $T$, and the mutual information $I(X; T)$ between the text and data variables. We introduce the \emph{Conditional-Vendi} score based on $H(X|T)$ to quantify the internal diversity of the model and the \emph{Information-Vendi} score based on $I(X; T)$ to measure the statistical relevance between the generated data and text prompts. We provide theoretical results to statistically interpret these scores and relate them to the unconditional Vendi score. We conduct several numerical experiments to show the correlation between the Conditional-Vendi score and the internal diversity of text-conditioned generative models. The codebase is available at \href{https://github.com/mjalali/conditional-vendi}{https://github.com/mjalali/conditional-vendi}.

The Spatial Knowledge Graphs (SKG) are experiencing growing adoption as a means to model real-world entities, proving especially invaluable in domains like crisis management and urban planning. Considering that RDF specifications offer limited support for effectively managing spatial information, it's common practice to include text-based serializations of geometrical features, such as polygons and lines, as string literals in knowledge graphs. Consequently, Spatial Knowledge Graphs (SKGs) often rely on geo-enabled RDF Stores capable of parsing, interpreting, and indexing such serializations. In this paper, we leverage grid cells as the foundational element of SKGs and demonstrate how efficiently the spatial characteristics of real-world entities and their attributes can be encoded within knowledge graphs. Furthermore, we introduce a novel methodology for representing street networks in knowledge graphs, diverging from the conventional practice of individually capturing each street segment. Instead, our approach is based on tessellating the street network using grid cells and creating a simplified representation that could be utilized for various routing and navigation tasks, solely relying on RDF specifications.

Using polarization spinor methods in conjunction with the superspace formalism, we construct 3-point superconformal invariants that are used to determine the form of 3-point correlators of spinning superfield operators in $\mathcal{N}=1$ superconformal field theories (SCFTs) in 4-dimensions. We enumerate the structural form of various spinning 3-point correlators using these invariants and find additional constraints on their form when the operators are conserved supercurrents. For these purposes, we first construct the invariants and 3-point correlators in non-supersymmetric $4d$ CFTs which are then extended using superspace methods to $4d$ SCFTs., Comment: 38 pages

Continual learning (CL) in medical imaging presents a unique challenge, requiring models to adapt to new domains while retaining previously acquired knowledge. We introduce NCAdapt, a Neural Cellular Automata (NCA) based method designed to address this challenge. NCAdapt features a domain-specific multi-head structure, integrating adaptable convolutional layers into the NCA backbone for each new domain encountered. After initial training, the NCA backbone is frozen, and only the newly added adaptable convolutional layers, consisting of 384 parameters, are trained along with domain-specific NCA convolutions. We evaluate NCAdapt on hippocampus segmentation tasks, benchmarking its performance against Lifelong nnU-Net and U-Net models with state-of-the-art (SOTA) CL methods. Our lightweight approach achieves SOTA performance, underscoring its effectiveness in addressing CL challenges in medical imaging. Upon acceptance, we will make our code base publicly accessible to support reproducibility and foster further advancements in medical CL.

This paper presents the multiplicity of neutral and charged hadrons produced in 90 GeV$/c$ proton-carbon interactions from a dataset taken by the NA61/SHINE experiment in 2017. Particle identification via dE/dx was performed for the charged hadrons $\pi^\pm$, $K^\pm$, and $p / \bar{p}$; the neutral hadrons $K^0_S$, $\Lambda$, and $\bar{\Lambda}$ were identified via an invariant mass analysis of their decays to charged hadrons. Double-differential multiplicity results as a function of laboratory momentum and polar angle are presented for each particle species; these results provide vital constraints on the predicted neutrino beam flux for current and future long-baseline neutrino oscillation experiments.

For modern industrial applications, accurately detecting and diagnosing anomalies in multivariate time series data is essential. Despite such need, most state-of-the-art methods often prioritize detection performance over model interpretability. Addressing this gap, we introduce MIXAD (Memory-Induced Explainable Time Series Anomaly Detection), a model designed for interpretable anomaly detection. MIXAD leverages a memory network alongside spatiotemporal processing units to understand the intricate dynamics and topological structures inherent in sensor relationships. We also introduce a novel anomaly scoring method that detects significant shifts in memory activation patterns during anomalies. Our approach not only ensures decent detection performance but also outperforms state-of-the-art baselines by 34.30% and 34.51% in interpretability metrics., Comment: ICPR 2024 (oral paper)

Medical image registration is a critical process that aligns various patient scans, facilitating tasks like diagnosis, surgical planning, and tracking. Traditional optimization based methods are slow, prompting the use of Deep Learning (DL) techniques, such as VoxelMorph and Transformer-based strategies, for faster results. However, these DL methods often impose significant resource demands. In response to these challenges, we present NCA-Morph, an innovative approach that seamlessly blends DL with a bio-inspired communication and networking approach, enabled by Neural Cellular Automata (NCAs). NCA-Morph not only harnesses the power of DL for efficient image registration but also builds a network of local communications between cells and respective voxels over time, mimicking the interaction observed in living systems. In our extensive experiments, we subject NCA-Morph to evaluations across three distinct 3D registration tasks, encompassing Brain, Prostate and Hippocampus images from both healthy and diseased patients. The results showcase NCA-Morph's ability to achieve state-of-the-art performance. Notably, NCA-Morph distinguishes itself as a lightweight architecture with significantly fewer parameters; 60% and 99.7% less than VoxelMorph and TransMorph. This characteristic positions NCA-Morph as an ideal solution for resource-constrained medical applications, such as primary care settings and operating rooms.

Given the growing focus on memristive crossbar-based in-memory computing (IMC) architectures as a potential alternative to current energy-hungry machine learning hardware, the availability of a fast and accurate circuit-level simulation framework could greatly enhance research and development efforts in this field. This paper introduces XbarSim, a domain-specific circuit-level simulator designed to analyze the nodal equations of memristive crossbars. The first version of XbarSim, proposed herein, leverages the lower-upper (LU) decomposition approach to solve the nodal equations for the matrices associated with crossbars. The XbarSim is capable of simulating interconnect parasitics within crossbars and supports batch processing of the inputs. Through comprehensive experiments, we demonstrate that the XbarSim can achieve orders of magnitude speedup compared to HSPICE across various sizes of memristive crossbars. The XbarSim's full suite of features is accessible to researchers as an open-source tool.

Outflows are critical components of many astrophysical systems, including accreting compact binaries and active galactic nuclei (AGN). These outflows can significantly affect a system's evolution and alter its observational appearance by reprocessing the radiation produced by the central engine. Sirocco (Simulating Ionization and Radiation in Outflows Created by Compact Objects - or "the code formerly known as Python") is a Sobolev-based Monte Carlo ionization and radiative transfer code. It is designed to simulate the spectra produced by any system with an azimuthally-symmetric outflow, from spherical stellar winds to rotating, biconical accretion disc winds. Wind models can either be parametrized or imported, e.g. from hydrodynamical simulations. The radiation sources include an optically thick accretion disc and various central sources with flexible spectra and geometries. The code tracks the "photon packets" produced by the sources in any given simulation as they traverse and interact with the wind. The code assumes radiative near-equilibrium, so the thermal and ionization state can be determined iteratively from these interactions. Once the physical properties in the wind have converged, Sirocco can be used to generate synthetic spectra at a series of observer sightlines. Here, we describe the physical assumptions, operation, performance and limitations of the code. We validate it against tardis, cmfgen and cloudy, finding good agreement, and present illustrative synthetic spectra from disc winds in cataclysmic variables, tidal disruption events, AGN and X-ray binaries. Sirocco is publicly available on GitHub, alongside its associated data, documentation and sample input files covering a wide range of astrophysical applications., Comment: 26 pages, 16 figures. Submitted to MNRAS. This is the release paper for the SIROCCO code, which can be found at https://github.com/sirocco-rt/sirocco with links to documentation. Underlying data and figure scripts available at https://github.com/sirocco-rt/release-models -- comments on the code, paper or documentation are welcomed

Magnetic Resonance Spectroscopic Imaging (MRSI) is a non-invasive imaging technique for studying metabolism and has become a crucial tool for understanding neurological diseases, cancers and diabetes. High spatial resolution MRSI is needed to characterize lesions, but in practice MRSI is acquired at low resolution due to time and sensitivity restrictions caused by the low metabolite concentrations. Therefore, there is an imperative need for a post-processing approach to generate high-resolution MRSI from low-resolution data that can be acquired fast and with high sensitivity. Deep learning-based super-resolution methods provided promising results for improving the spatial resolution of MRSI, but they still have limited capability to generate accurate and high-quality images. Recently, diffusion models have demonstrated superior learning capability than other generative models in various tasks, but sampling from diffusion models requires iterating through a large number of diffusion steps, which is time-consuming. This work introduces a Flow-based Truncated Denoising Diffusion Model (FTDDM) for super-resolution MRSI, which shortens the diffusion process by truncating the diffusion chain, and the truncated steps are estimated using a normalizing flow-based network. The network is conditioned on upscaling factors to enable multi-scale super-resolution. To train and evaluate the deep learning models, we developed a 1H-MRSI dataset acquired from 25 high-grade glioma patients. We demonstrate that FTDDM outperforms existing generative models while speeding up the sampling process by over 9-fold compared to the baseline diffusion model. Neuroradiologists' evaluations confirmed the clinical advantages of our method, which also supports uncertainty estimation and sharpness adjustment, extending its potential clinical applications., Comment: Accepted by Medical Image Analysis (MedIA)

The present paper investigated automatic melody construction for Persian lyrics as an input. It was assumed that there is a phonological correlation between the lyric syllables and the melody in a song. A seq2seq neural network was developed to investigate this assumption, trained on parallel syllable and note sequences in Persian songs to suggest a pleasant melody for a new sequence of syllables. More than 100 pieces of Persian music were collected and converted from the printed version to the digital format due to the lack of a dataset on Persian digital music. Finally, 14 new lyrics were given to the model as input, and the suggested melodies were performed and recorded by music experts to evaluate the trained model. The evaluation was conducted using an audio questionnaire, which more than 170 persons answered. According to the answers about the pleasantness of melody, the system outputs scored an average of 3.005 from 5, while the human-made melodies for the same lyrics obtained an average score of 4.078.

By introducing the small noise expansion techniques, we show that the fully nonlinear (non-Markovian) stochastic inflationary system, may be re-cast in terms of an infinite set of Wiener processes (stochastic equations with white noises). As a byproduct, we show that the Starobinsky test field approximation might only provide information about the linear regime of cosmological perturbations and scalar-feld non-Gaussianities might only appear at next to leading order in slow-roll parameters., Comment: 24 pages 2 appendices

We explore the cosmological implications of the quasi-dilaton massive gravity theory with a non-minimal kinetic coupling. We investigate the resulting gravitational dynamics by introducing a nonzero mass for the graviton. We derive the complete set of background equations of motion and obtain self-accelerating background solutions to explain the late-time accelerated expansion of the Universe using the effective cosmological constant originating from massive gravity. Furthermore, we derive the dispersion relation of gravitational waves to obtain the stability condition of the system. This allows us to use the propagation of gravitational perturbations in a Friedman-Lemaitre-Robertson-Walker (FLRW) cosmology to constrain the quasi-dilaton massive gravity with a non-minimal kinetic coupling theory.

The US opioid overdose epidemic has been a major public health concern in recent decades. There has been increasing recognition that its etiology is rooted in part in the social contexts that mediate substance use and access; however, reliable statistical measures of social influence are lacking in the literature. We use Facebook's social connectedness index (SCI) as a proxy for real-life social networks across diverse spatial regions that help quantify social connectivity across different spatial units. This is a measure of the relative probability of connections between localities that offers a unique lens to understand the effects of social networks on health outcomes. We use SCI to develop a variable, called "deaths in social proximity", to measure the influence of social networks on opioid overdose deaths (OODs) in US counties. Our results show a statistically significant effect size for deaths in social proximity on OODs in counties in the United States, controlling for spatial proximity, as well as demographic and clinical covariates. The effect size of standardized deaths in social proximity in our cluster-robust linear regression model indicates that a one-standard-deviation increase, equal to 11.70 more deaths per 100,000 population in the social proximity of ego counties in the contiguous United States, is associated with thirteen more deaths per 100,000 population in ego counties. To further validate our findings, we performed a series of robustness checks using a network autocorrelation model to account for social network effects, a spatial autocorrelation model to capture spatial dependencies, and a two-way fixed-effect model to control for unobserved spatial and time-invariant characteristics. These checks consistently provide statistically robust evidence of positive social influence on OODs in US counties.

Function-space priors in Bayesian Neural Networks provide a more intuitive approach to embedding beliefs directly into the model's output, thereby enhancing regularization, uncertainty quantification, and risk-aware decision-making. However, imposing function-space priors on BNNs is challenging. We address this task through optimization techniques that explore how trainable activations can accommodate complex priors and match intricate target function distributions. We discuss critical learning challenges, including identifiability, loss construction, and symmetries that arise in this context. Furthermore, we enable evidence maximization to facilitate model selection by conditioning the functional priors on additional hyperparameters. Our empirical findings demonstrate that even BNNs with a single wide hidden layer, when equipped with these adaptive trainable activations and conditioning strategies, can effectively achieve high-fidelity function-space priors, providing a robust and flexible framework for enhancing Bayesian neural network performance.

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

This work focuses on advancing security research in the hardware design space by formally defining the realistic problem of Hardware Trojan (HT) detection. The goal is to model HT detection more closely to the real world, i.e., describing the problem as The Seeker's Dilemma where a detecting agent is unaware of whether circuits are infected by HTs or not. Using this theoretical problem formulation, we create a benchmark that consists of a mixture of HT-free and HT-infected restructured circuits while preserving their original functionalities. The restructured circuits are randomly infected by HTs, causing a situation where the defender is uncertain if a circuit is infected or not. We believe that our innovative benchmark and methodology of creating benchmarks will help the community judge the detection quality of different methods by comparing their success rates in circuit classification. We use our developed benchmark to evaluate three state-of-the-art HT detection tools to show baseline results for this approach. We use Principal Component Analysis to assess the strength of our benchmark, where we observe that some restructured HT-infected circuits are mapped closely to HT-free circuits, leading to significant label misclassification by detectors., Comment: arXiv admin note: substantial text overlap with arXiv:2402.17918

The aggressiveness of prostate cancer, the most common cancer in men worldwide, is primarily assessed based on histopathological data using the Gleason scoring system. While artificial intelligence (AI) has shown promise in accurately predicting Gleason scores, these predictions often lack inherent explainability, potentially leading to distrust in human-machine interactions. To address this issue, we introduce a novel dataset of 1,015 tissue microarray core images, annotated by an international group of 54 pathologists. The annotations provide detailed localized pattern descriptions for Gleason grading in line with international guidelines. Utilizing this dataset, we develop an inherently explainable AI system based on a U-Net architecture that provides predictions leveraging pathologists' terminology. This approach circumvents post-hoc explainability methods while maintaining or exceeding the performance of methods trained directly for Gleason pattern segmentation (Dice score: 0.713 $\pm$ 0.003 trained on explanations vs. 0.691 $\pm$ 0.010 trained on Gleason patterns). By employing soft labels during training, we capture the intrinsic uncertainty in the data, yielding strong results in Gleason pattern segmentation even in the context of high interobserver variability. With the release of this dataset, we aim to encourage further research into segmentation in medical tasks with high levels of subjectivity and to advance the understanding of pathologists' reasoning processes., Comment: 58 pages, 15 figures (incl. supplementary)

The aim of this paper is to address the challenge of gradual domain adaptation within a class of manifold-constrained data distributions. In particular, we consider a sequence of $T\ge2$ data distributions $P_1,\ldots,P_T$ undergoing a gradual shift, where each pair of consecutive measures $P_i,P_{i+1}$ are close to each other in Wasserstein distance. We have a supervised dataset of size $n$ sampled from $P_0$, while for the subsequent distributions in the sequence, only unlabeled i.i.d. samples are available. Moreover, we assume that all distributions exhibit a known favorable attribute, such as (but not limited to) having intra-class soft/hard margins. In this context, we propose a methodology rooted in Distributionally Robust Optimization (DRO) with an adaptive Wasserstein radius. We theoretically show that this method guarantees the classification error across all $P_i$s can be suitably bounded. Our bounds rely on a newly introduced {\it {compatibility}} measure, which fully characterizes the error propagation dynamics along the sequence. Specifically, for inadequately constrained distributions, the error can exponentially escalate as we progress through the gradual shifts. Conversely, for appropriately constrained distributions, the error can be demonstrated to be linear or even entirely eradicated. We have substantiated our theoretical findings through several experimental results., Comment: Published at Proceedings of Neural Information Processing Systems (NeurIPS) 2024

Traditional Retrieval-Augmented Generation (RAG) benchmarks rely on different heuristic-based metrics for evaluation, but these require human preferences as ground truth for reference. In contrast, arena-based benchmarks, where two models compete each other, require an expensive Large Language Model (LLM) as a judge for a reliable evaluation. We present an easy and efficient technique to get the best of both worlds. The idea is to train a learning to rank model as a "surrogate" judge using RAG-based evaluation heuristics as input, to produce a synthetic arena-based leaderboard. Using this idea, We develop MIRAGE-Bench, a standardized arena-based multilingual RAG benchmark for 18 diverse languages on Wikipedia. The benchmark is constructed using MIRACL, a retrieval dataset, and extended for multilingual generation evaluation. MIRAGE-Bench evaluates RAG extensively coupling both heuristic features and LLM as a judge evaluator. In our work, we benchmark 19 diverse multilingual-focused LLMs, and achieve a high correlation (Kendall Tau ($\tau$) = 0.909) using our surrogate judge learned using heuristic features with pairwise evaluations and between GPT-4o as a teacher on the MIRAGE-Bench leaderboard using the Bradley-Terry framework. We observe proprietary and large open-source LLMs currently dominate in multilingual RAG. MIRAGE-Bench is available at: https://github.com/vectara/mirage-bench.

Summarization is one of the most common tasks performed by large language models (LLMs), especially in applications like Retrieval-Augmented Generation (RAG). However, existing evaluations of hallucinations in LLM-generated summaries, and evaluations of hallucination detection models both suffer from a lack of diversity and recency in the LLM and LLM families considered. This paper introduces FaithBench, a summarization hallucination benchmark comprising challenging hallucinations made by 10 modern LLMs from 8 different families, with ground truth annotations by human experts. ``Challenging'' here means summaries on which popular, state-of-the-art hallucination detection models, including GPT-4o-as-a-judge, disagreed on. Our results show GPT-4o and GPT-3.5-Turbo produce the least hallucinations. However, even the best hallucination detection models have near 50\% accuracies on FaithBench, indicating lots of room for future improvement. The repo is https://github.com/vectara/FaithBench

Attributing answers to source documents is an approach used to enhance the verifiability of a model's output in retrieval augmented generation (RAG). Prior work has mainly focused on improving and evaluating the attribution quality of large language models (LLMs) in RAG, but this may come at the expense of inducing biases in the attribution of answers. We define and examine two aspects in the evaluation of LLMs in RAG pipelines, namely attribution sensitivity and bias with respect to authorship information. We explicitly inform an LLM about the authors of source documents, instruct it to attribute its answers, and analyze (i) how sensitive the LLM's output is to the author of source documents, and (ii) whether the LLM exhibits a bias towards human-written or AI-generated source documents. We design an experimental setup in which we use counterfactual evaluation to study three LLMs in terms of their attribution sensitivity and bias in RAG pipelines. Our results show that adding authorship information to source documents can significantly change the attribution quality of LLMs by 3% to 18%. Moreover, we show that LLMs can have an attribution bias towards explicit human authorship, which can serve as a competing hypothesis for findings of prior work that shows that LLM-generated content may be preferred over human-written contents. Our findings indicate that metadata of source documents can influence LLMs' trust, and how they attribute their answers. Furthermore, our research highlights attribution bias and sensitivity as a novel aspect of brittleness in LLMs.

Healthcare industries face challenges when experiencing rare diseases due to limited samples. Artificial Intelligence (AI) communities overcome this situation to create synthetic data which is an ethical and privacy issue in the medical domain. This research introduces the CAT-U-Net framework as a new approach to overcome these limitations, which enhances feature extraction from medical images without the need for large datasets. The proposed framework adds an extra concatenation layer with downsampling parts, thereby improving its ability to learn from limited data while maintaining patient privacy. To validate, the proposed framework's robustness, different medical conditioning datasets were utilized including COVID-19, brain tumors, and wrist fractures. The framework achieved nearly 98% reconstruction accuracy, with a Dice coefficient close to 0.946. The proposed CAT-U-Net has the potential to make a big difference in medical image diagnostics in settings with limited data., Comment: 7 pages, 7 figures

Finetuning foundation models for specific tasks is an emerging paradigm in modern machine learning. The efficacy of task-specific finetuning largely depends on the selection of appropriate training data. We present TSDS (Task-Specific Data Selection), a framework to select data for task-specific model finetuning, guided by a small but representative set of examples from the target task. To do so, we formulate data selection for task-specific finetuning as an optimization problem with a distribution alignment loss based on optimal transport to capture the discrepancy between the selected data and the target distribution. In addition, we add a regularizer to encourage the diversity of the selected data and incorporate kernel density estimation into the regularizer to reduce the negative effects of near-duplicates among the candidate data. We connect our optimization problem to nearest neighbor search and design efficient algorithms to compute the optimal solution based on approximate nearest neighbor search techniques. We evaluate our method on data selection for both continued pretraining and instruction tuning of language models. We show that instruction tuning using data selected by our method with a 1% selection ratio often outperforms using the full dataset and beats the baseline selection methods by 1.5 points in F1 score on average., Comment: 31 pages, 1 figure

Large Language Models (LLMs) have demonstrated remarkable capabilities in performing tasks across various domains without needing explicit retraining. This capability, known as In-Context Learning (ICL), while impressive, exposes LLMs to a variety of adversarial prompts and jailbreaks that manipulate safety-trained LLMs into generating undesired or harmful output. In this paper, we propose a novel interpretation of ICL in LLMs through the lens of cognitive neuroscience, by drawing parallels between learning in human cognition with ICL. We applied the principles of Cognitive Load Theory in LLMs and empirically validate that similar to human cognition, LLMs also suffer from cognitive overload a state where the demand on cognitive processing exceeds the available capacity of the model, leading to potential errors. Furthermore, we demonstrated how an attacker can exploit ICL to jailbreak LLMs through deliberately designed prompts that induce cognitive overload on LLMs, thereby compromising the safety mechanisms of LLMs. We empirically validate this threat model by crafting various cognitive overload prompts and show that advanced models such as GPT-4, Claude-3.5 Sonnet, Claude-3 OPUS, Llama-3-70B-Instruct, Gemini-1.0-Pro, and Gemini-1.5-Pro can be successfully jailbroken, with attack success rates of up to 99.99%. Our findings highlight critical vulnerabilities in LLMs and underscore the urgency of developing robust safeguards. We propose integrating insights from cognitive load theory into the design and evaluation of LLMs to better anticipate and mitigate the risks of adversarial attacks. By expanding our experiments to encompass a broader range of models and by highlighting vulnerabilities in LLMs' ICL, we aim to ensure the development of safer and more reliable AI systems., Comment: 40 pages, 31 Figures

Healthcare technology has advanced with the development of an IoT-based real-time monitoring system for post-operative heart disease patients. This system integrates advanced sensor technologies with cloud-connected platforms to continuously track vital signs, enabling more effective post-operative care. The system employs the HW-827 Heart Rate Sensor and DS18B20 Temperature Sensor, integrated with an ESP32 for wireless connectivity, to collect real-time data. Data was collected from 20 participants in each of the following categories: healthy individuals, pre-operative heart disease patients, and post-operative heart disease patients. The results revealed distinct physiological variations across the groups. Healthy individuals maintained stable temperatures (mean 36.3{\deg}C), while post-operative patients exhibited slightly elevated temperatures (mean 36.6{\deg}C). Heart rate analysis indicated higher rates in pre-operative (mean 87.8 bpm) and post-operative patients (mean 82.0 bpm) compared to healthy individuals (mean 72.4 bpm). ECG analysis showed elevated readings in pre-operative patients (mean 0.16 mV) compared to healthy individuals (mean 0.10 mV) and post-operative patients (mean 0.12 mV). Statistical analyses, including ANOVA tests, confirmed significant differences among the groups in terms of temperature (F(2, 12) = 4.57, p < 0.05), heart rate (F(2, 12) = 5.91, p < 0.01), and ECG readings (F(2, 12) = 3.24, p < 0.05). This study demonstrates the potential of IoT technologies in enhancing post-operative care through continuous monitoring and data-driven insights. It emphasizes the role of real-time patient data in improving personalized healthcare management., Comment: 17 Pages, 7 figures, 7 Tables, Journal

This paper presents the findings from the third edition of the Chat Translation Shared Task. As with previous editions, the task involved translating bilingual customer support conversations, specifically focusing on the impact of conversation context in translation quality and evaluation. We also include two new language pairs: English-Korean and English-Dutch, in addition to the set of language pairs from previous editions: English-German, English-French, and English-Brazilian Portuguese. We received 22 primary submissions and 32 contrastive submissions from eight teams, with each language pair having participation from at least three teams. We evaluated the systems comprehensively using both automatic metrics and human judgments via a direct assessment framework. The official rankings for each language pair were determined based on human evaluation scores, considering performance in both translation directions--agent and customer. Our analysis shows that while the systems excelled at translating individual turns, there is room for improvement in overall conversation-level translation quality., Comment: 12 pages, 5 figures, 13 tables

This paper introduces a robust stress detection system utilizing a Convolutional Neural Network (CNN) designed for the analysis of Photoplethysmogram (PPG) signals. Employing the WESAD dataset, we applied Continuous Wavelet Transform (CWT) to extract informative features from wrist PPG signals, demonstrating enhanced stress detection and learning compared to conventional techniques. Notably, the CNN achieved an impressive accuracy of 93.7% after five epochs, post-implementation on a resource-constrained microcontroller. The optimization process, including pruning and Post-Train Quantization, was crucial to reduce the model size to 1.6 megabytes, overcoming the microcontroller's limited resources of 2 megabytes of Flash memory and 512 kilobytes of RAM. This optimized model not only addresses resource constraints but also outperforms traditional signal processing methods, positioning it as a promising solution for real-time stress monitoring on wearable devices., Comment: 5 figures, implemented on Microcontroller