Search

This paper presents an analysis of biases in open-source Large Language Models (LLMs) across various genders, religions, and races. We introduce a methodology for generating a bias detection dataset using seven bias triggers: General Debate, Positioned Debate, Career Advice, Story Generation, Problem-Solving, Cover-Letter Writing, and CV Generation. We use GPT-4o to generate a diverse set of prompts for each trigger across various genders, religious and racial groups. We evaluate models from Llama and Gemma family on the generated dataset. We anonymise the LLM-generated text associated with each group using GPT-4o-mini and do a pairwise comparison using GPT-4o-as-a-Judge. To quantify bias in the LLM-generated text we use the number of wins and losses in the pairwise comparison. Our analysis spans three languages, English, German, and Arabic to explore how language influences bias manifestation. Our findings reveal that LLMs exhibit strong polarization toward certain groups across each category, with a notable consistency observed across models. However, when switching languages, variations and anomalies emerge, often attributable to cultural cues and contextual differences.

Understanding protein motion within the cell is crucial for predicting reaction rates and macromolecular transport in the cytoplasm. A key question is how crowded environments affect protein dynamics through hydrodynamic and direct interactions at molecular length scales. Using megahertz X-ray Photon Correlation Spectroscopy (MHz-XPCS) at the European X-ray Free Electron Laser (EuXFEL), we investigate ferritin diffusion at microsecond time scales. Our results reveal anomalous diffusion, indicated by the non-exponential decay of the intensity autocorrelation function $g_2(q,t)$ at high concentrations. This behavior is consistent with the presence of cage-trapping in between the short- and long-time protein diffusion regimes. Modeling with the $\delta\gamma$-theory of hydrodynamically interacting colloidal spheres successfully reproduces the experimental data by including a scaling factor linked to the protein direct interactions. These findings offer new insights into the complex molecular motion in crowded protein solutions, with potential applications for optimizing ferritin-based drug delivery, where protein diffusion is the rate-limiting step.

Active learning (AL) optimizes data labeling efficiency by selecting the most informative instances for annotation. A key component in this procedure is an acquisition function that guides the selection process and identifies the suitable instances for labeling from the unlabeled pool. However, these acquisition methods suffer from high computational costs with large unlabeled data pools, posing a roadblock to their applicability on large datasets. To address this challenge and bridge this gap, we introduce a novel plug-and-play unlabeled data pruning strategy, ActivePrune, which leverages language models to prune the unlabeled pool. ActivePrune implements a two-stage pruning process: an initial fast evaluation using perplexity scores from an n-gram language model, followed by a high-quality selection using metrics for data quality computed through a quantized LLM. Additionally, to enhance the diversity in the unlabeled pool, we propose a novel perplexity reweighting method that systematically brings forward underrepresented instances for selection in subsequent labeling iterations. Experiments on translation, sentiment analysis, topic classification, and summarization tasks on four diverse datasets and four active learning strategies demonstrate that ActivePrune outperforms existing data pruning methods. Finally, we compare the selection quality $\leftrightarrow$ efficiency tradeoff of the data pruning methods and demonstrate that ActivePrune is computationally more efficient than other LLM score-based pruning methods, and provides up to 74% reduction in the end-to-end time required for active learning., Comment: 20 pages, 4 figures

Advancements in generative modeling are pushing the state-of-the-art in synthetic medical image generation. These synthetic images can serve as an effective data augmentation method to aid the development of more accurate machine learning models for medical image analysis. While the fidelity of these synthetic images has progressively increased, the diversity of these images is an understudied phenomenon. In this work, we propose the SDICE index, which is based on the characterization of similarity distributions induced by a contrastive encoder. Given a synthetic dataset and a reference dataset of real images, the SDICE index measures the distance between the similarity score distributions of original and synthetic images, where the similarity scores are estimated using a pre-trained contrastive encoder. This distance is then normalized using an exponential function to provide a consistent metric that can be easily compared across domains. Experiments conducted on the MIMIC-chest X-ray and ImageNet datasets demonstrate the effectiveness of SDICE index in assessing synthetic medical dataset diversity., Comment: Accepted at BMVC 2024 - PFATCV

Lattice quantum chromodynamics (QCD) calculations share a defining challenge by requiring a small finite range of spatial separation $z$ between quark/gluon bilinears for controllable power corrections in the perturbative QCD factorization, and a large hadron boost $p_z$ for a successful determination of collinear parton distribution functions (PDFs). However, these two requirements make the determination of PDFs from lattice data very challenging. We present the application of generative machine learning algorithms to estimate the polarized and unpolarized gluon correlation functions utilizing short-distance data and extending the correlation up to $zp_z \lesssim 14$, surpassing the current capabilities of lattice QCD calculations. We train physics-informed machine learning algorithms to learn from the short-distance correlation at $z\lesssim 0.36$ fm and take the limit, $p_z \to \infty$, thereby minimizing possible contamination from the higher-twist effects for a successful reconstruction of the polarized gluon PDF. We also expose the bias and problems with underestimating uncertainties associated with the use of model-dependent and overly constrained functional forms, such as $x^\alpha(1-x)^\beta$ and its variants to extract PDFs from the lattice data. We propose the use of generative machine learning algorithms to mitigate these issues and present our determination of the polarized and unpolarized gluon PDFs in the nucleon., Comment: 24 pages, 18 figures

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

This paper presents a comprehensive evaluation of Urdu Automatic Speech Recognition (ASR) models. We analyze the performance of three ASR model families: Whisper, MMS, and Seamless-M4T using Word Error Rate (WER), along with a detailed examination of the most frequent wrong words and error types including insertions, deletions, and substitutions. Our analysis is conducted using two types of datasets, read speech and conversational speech. Notably, we present the first conversational speech dataset designed for benchmarking Urdu ASR models. We find that seamless-large outperforms other ASR models on the read speech dataset, while whisper-large performs best on the conversational speech dataset. Furthermore, this evaluation highlights the complexities of assessing ASR models for low-resource languages like Urdu using quantitative metrics alone and emphasizes the need for a robust Urdu text normalization system. Our findings contribute valuable insights for developing robust ASR systems for low-resource languages like Urdu.

We investigate the role of non-selective measurement on the estimation of system-environment parameters. Projective measurement is the popular method of initial state preparation which always prepares a pure state. However, in various physical situations of physical interest, this selective measurement becomes unrealistic. In this paper, we compare the estimation results obtained via projective measurement with the results obtained via unitary operation. We argue that in typical situations, parameters can be estimated with higher accuracy if the initial state is prepared with the unitary operator (a pulse). We consider the spin-spin model where a central two-level system (probe) interacts with the collections of two-level systems (bath). A probe interacts with a bath and attains a thermal equilibrium state, then via unitary operation, the initial state is prepared which evolves unitarily. The properties of the bath are imprinted on the reduced dynamics. Due to the initial probe-bath correlations present in the thermal equilibrium state, an additional factor arises in the dynamics which has a phenomenal role in the parameter estimation. In this paper, we study the estimation of bath temperature and probe-bath coupling strength which is quantified by the quantum Fisher information. Our results are promising as one can improve the precision of the estimates by orders of magnitude via non-selective measurement and by incorporating the effect of initial correlations., Comment: 10 Pages, 7 figures

In medical image classification, supervised learning is challenging due to the lack of labeled medical images. Contrary to the traditional \textit{modus operandi} of pre-training followed by fine-tuning, this work leverages the visual-textual alignment within Vision-Language models (\texttt{VLMs}) to facilitate the unsupervised learning. Specifically, we propose \underline{Med}ical \underline{Un}supervised \underline{A}daptation (\texttt{MedUnA}), constituting two-stage training: Adapter Pre-training, and Unsupervised Learning. In the first stage, we use descriptions generated by a Large Language Model (\texttt{LLM}) corresponding to class labels, which are passed through the text encoder \texttt{BioBERT}. The resulting text embeddings are then aligned with the class labels by training a lightweight \texttt{adapter}. We choose \texttt{\texttt{LLMs}} because of their capability to generate detailed, contextually relevant descriptions to obtain enhanced text embeddings. In the second stage, the trained \texttt{adapter} is integrated with the visual encoder of \texttt{MedCLIP}. This stage employs a contrastive entropy-based loss and prompt tuning to align visual embeddings. We incorporate self-entropy minimization into the overall training objective to ensure more confident embeddings, which are crucial for effective unsupervised learning and alignment. We evaluate the performance of \texttt{MedUnA} on three different kinds of data modalities - chest X-rays, eye fundus and skin lesion images. The results demonstrate significant accuracy gain on average compared to the baselines across different datasets, highlighting the efficacy of our approach.

We introduce Spurfies, a novel method for sparse-view surface reconstruction that disentangles appearance and geometry information to utilize local geometry priors trained on synthetic data. Recent research heavily focuses on 3D reconstruction using dense multi-view setups, typically requiring hundreds of images. However, these methods often struggle with few-view scenarios. Existing sparse-view reconstruction techniques often rely on multi-view stereo networks that need to learn joint priors for geometry and appearance from a large amount of data. In contrast, we introduce a neural point representation that disentangles geometry and appearance to train a local geometry prior using a subset of the synthetic ShapeNet dataset only. During inference, we utilize this surface prior as additional constraint for surface and appearance reconstruction from sparse input views via differentiable volume rendering, restricting the space of possible solutions. We validate the effectiveness of our method on the DTU dataset and demonstrate that it outperforms previous state of the art by 35% in surface quality while achieving competitive novel view synthesis quality. Moreover, in contrast to previous works, our method can be applied to larger, unbounded scenes, such as Mip-NeRF 360., Comment: https://geometric-rl.mpi-inf.mpg.de/spurfies/

In this paper, a compact multi-service antenna (MSA) is presented for sensing and communication using a reconfigurable complementary spiral resonator. A three turns complementary spiral resonator (3-CSR) is inserted in the ground plane of a modified patch antenna to create a miniaturized structure. Two Positive-Intrinsic-Negative (PIN) diodes (D1, D2) are also integrated with the 3-CSR to achieve frequency reconfiguration. The proposed structure operates in three different modes i.e., dual-band joint communication and sensing antenna (JCASA), dual-band antenna, and single-band antenna. The required mode can be selected by changing the state of the PIN diodes. In mode-1, the first band (0.95-0.97 GHz) of the antenna is dedicated to sensing by using frequency domain reflectometry (FDR), while the second band (1.53-1.56 GHz) is allocated to communication. The sensing ability of the proposed structure is utilized to measure soil moisture using FDR. Based on the frequency shift, permittivity of the soil is observed to measure soil moisture. In mode-2 and mode-3, the structure operates as a standard dual and single band antenna, respectively, with a maximum gain of 1.5 dBi at 1.55 GHz. The proposed planar structure, with its simple geometry and a high sensitivity of 1.7%, is a suitable candidate for precision farming. The proposed structure is versatile and capable of being utilized as a single or dual-band antenna and also measuring permittivity of materials within the range of 1-20. Hence, it is adaptable to a range of applications.

This study establishes bounds on the leading-order (LO) hadronic vacuum polarization (HVP) contribution to the anomalous magnetic moment of the muon ($a_\mu^{\mathrm{HVP,LO}}$, $a_\mu = (g-2)_\mu/2$) by using H\"older's inequality and related inequalities in Finite-Energy QCD sum rules. Considering contributions from light quarks ($u,d,s$) up to five-loop order in perturbation theory within the chiral limit, leading-order light-quark mass corrections, next-to-leading order for dimension-four QCD condensates, and leading-order for dimension-six QCD condensates, the study finds QCD lower and upper bounds as $\left(657.0\pm 34.8\right)\times 10^{-10}\leq a_\mu^{\mathrm{HVP,LO}} \leq \left(788.4\pm 41.8\right)\times10^{-10}\,$., Comment: 7 pages, 2 figures, 3 tables. Proceedings article for QCD24: 27th High-Energy Physics International Conference in Quantum Chromodynamis

A miniaturized linearly-polarized patch antenna is presented for Wireless Power Transfer (WPT) at 1. 8 GHz. The proposed antenna consists of a patch element and a 3-turn Complementary Spiral Resonator (3-CSR) with antenna dimension of 50 mm x 50 mm. 3-CSR is inserted in the ground plane to reduce the antenna size. This modification also increased the impedance bandwidth from 43 MHz (1.78-1.83 GHz) to 310 MHz (1.69-2.0 GHz) . Moreover, antenna is fabricated and simulated and measured results are in good agreement. Additionally, a rectifier and matching circuits are designed at -10 dBm to realize a rectenna (rectifying antenna) for WPT application. Rectenna efficiency of 53.6 % is achieved at a low input power of -10 dBm.

Background: The rapid advancement of Machine Learning (ML) represents novel opportunities to enhance public health research, surveillance, and decision-making. However, there is a lack of comprehensive understanding of algorithmic bias, systematic errors in predicted population health outcomes, resulting from the public health application of ML. The objective of this narrative review is to explore the types of bias generated by ML and quantitative metrics to assess these biases. Methods : We performed search on PubMed, MEDLINE, IEEE (Institute of Electrical and Electronics Engineers), ACM (Association for Computing Machinery) Digital Library, Science Direct, and Springer Nature. We used keywords to identify studies describing types of bias and metrics to measure these in the domain of ML and public and population health published in English between 2008 and 2023, inclusive. Results: A total of 72 articles met the inclusion criteria. Our review identified the commonly described types of bias and quantitative metrics to assess these biases from an equity perspective. Conclusion : The review will help formalize the evaluation framework for ML on public health from an equity perspective., Comment: under review

Large language models (LLMs) have become increasingly prevalent in our daily lives, leading to an expectation for LLMs to be trustworthy -- - both accurate and well-calibrated (the prediction confidence should align with its ground truth correctness likelihood). Nowadays, fine-tuning has become the most popular method for adapting a model to practical usage by significantly increasing accuracy on downstream tasks. Despite the great accuracy it achieves, we found fine-tuning is still far away from satisfactory trustworthiness due to "tuning-induced mis-calibration". In this paper, we delve deeply into why and how mis-calibration exists in fine-tuned models, and how distillation can alleviate the issue. Then we further propose a brand new method named Efficient Trustworthy Distillation (FIRST), which utilizes a small portion of teacher's knowledge to obtain a reliable language model in a cost-efficient way. Specifically, we identify the "concentrated knowledge" phenomenon during distillation, which can significantly reduce the computational burden. Then we apply a "trustworthy maximization" process to optimize the utilization of this small portion of concentrated knowledge before transferring it to the student. Experimental results demonstrate the effectiveness of our method, where better accuracy (+2.3%) and less mis-calibration (-10%) are achieved on average across both in-domain and out-of-domain scenarios, indicating better trustworthiness., Comment: EMNLP 2024

Capacitive coupling wireless power transfer (CCWPT) is one of the pervasive methods to transfer power in the reactive near-field zone. In this paper, a flexible design methodology based on Binary Particle Swarm Optimization (BPSO) algorithm is proposed for a pixelated microstrip structure. The pixel configuration of each parallel plate (43x43 pixels) determines the frequency response of the system (S-parameters) and by changing this configuration, we can achieve the dedicated operating frequency (resonance frequency) and its related |S21| value. Due to the large number of pixels, iterative optimization algorithm (BPSO) is the solution for designing a CCWPT system. However, the output of each iteration should be simulated in electromagnetic simulators (e.g., CST, HFSS, etc.), hence, the whole optimization process is time-consuming. This paper develops a rapid, agile and efficient method for designing two parallel pixelated microstrip plates of a CCWPT system based on deep neural networks. In the proposed method, CST-based BPSO algorithm is replaced with an AI-based method using ResNet-18. Advantages of the AI-based iterative method are automatic design process, more efficient, less time-consuming, less computational resource-consuming and less background EM knowledge requirements compared to the conventional techniques. Finally, the prototype of the proposed simulated structure is fabricated and measured. The simulation and measurement results validate the design procedure accuracy, using AI-based BPSO algorithm. The MAE (Mean Absolute Error) of prediction for the main resonance frequency and related |S21| are 110 MHz and 0.18 dB, respectively and according to the simulation results, the whole design process is 3629 times faster than the CST-based BPSO algorithm.

To date, most dendritic studies have predominantly focused on the apical zone of pyramidal two-point neurons (TPNs) receiving only feedback (FB) connections from higher perceptual layers and using them for learning. Recent cellular neurophysiology and computational neuroscience studies suggests that the apical input (context), coming from feedback and lateral connections, is multifaceted and far more diverse, with greater implications for ongoing learning and processing in the brain than previously realized. In addition to the FB, the apical tuft receives signals from neighboring cells of the same network as proximal (P) context, other parts of the brain as distal (D) context, and overall coherent information across the network as universal (U) context. The integrated context (C) amplifies and suppresses the transmission of coherent and conflicting feedforward (FF) signals, respectively. Specifically, we show that complex context-sensitive (CS)-TPNs flexibly integrate C moment-by-moment with the FF somatic current at the soma such that the somatic current is amplified when both feedforward (FF) and C are coherent; otherwise, it is attenuated. This generates the event only when the FF and C currents are coherent, which is then translated into a singlet or a burst based on the FB information. Spiking simulation results show that this flexible integration of somatic and contextual currents enables the propagation of more coherent signals (bursts), making learning faster with fewer neurons. Similar behavior is observed when this functioning is used in conventional artificial networks, where orders of magnitude fewer neurons are required to process vast amounts of heterogeneous real-world audio-visual (AV) data trained using backpropagation (BP). The computational findings presented here demonstrate the universality of CS-TPNs, suggesting a dendritic narrative that was previously overlooked.

This paper presents a novel methodology for generating synthetic Preference Optimization (PO) datasets using multi-agent workflows. We evaluate the effectiveness and potential of these workflows in automating and enhancing the dataset generation process. PO dataset generation requires two modules: (1) response evaluation, and (2) response generation. In the response evaluation module, the responses from Large Language Models (LLMs) are evaluated and ranked - a task typically carried out by human annotators that we automate using LLMs. We assess the response evaluation module in a 2 step process. In step 1, we assess LLMs as evaluators using three distinct prompting strategies. In step 2, we apply the winning prompting strategy to compare the performance of LLM-as-a-Judge, LLMs-as-a-Jury, and LLM Debate. Our evaluation shows that GPT-4o-as-a-Judge is more consistent across all datasets. For the response generation module, we use the identified LLM evaluator configuration and compare different configurations of the LLM Feedback Loop. We use the win rate to determine the best multi-agent configuration for generation. Experimenting with various configurations, we find that the LLM Feedback Loop, with Llama as the generator and Gemma as the reviewer, achieves a notable 71.8% and 73.8% win rate over single-agent Llama and Gemma, respectively. After identifying the best configurations for both modules, we generate our PO datasets using the above pipeline.

The Transformer architecture has revolutionized deep learning through its Self-Attention mechanism, which effectively captures contextual information. However, the memory footprint of Self-Attention presents significant challenges for long-sequence tasks. Grouped Query Attention (GQA) addresses this issue by grouping queries and mean-pooling the corresponding key-value heads - reducing the number of overall parameters and memory requirements in a flexible manner without adversely compromising model accuracy. In this work, we introduce enhancements to GQA, focusing on two novel approaches that deviate from the static nature of grouping: Key-Distributed GQA (KDGQA) and Dynamic Key-Distributed GQA (DGQA), which leverage information from the norms of the key heads to inform query allocation. Specifically, KDGQA looks at the ratios of the norms of the key heads during each forward pass, while DGQA examines the ratios of the norms as they evolve through training. Additionally, we present Perturbed GQA (PGQA) as a case-study, which introduces variability in (static) group formation via subtracting noise from the attention maps. Our experiments with up-trained Vision Transformers, for Image Classification on datasets such as CIFAR-10, CIFAR-100, Food101, and Tiny ImageNet, demonstrate the promise of these variants in improving upon the original GQA through more informed and adaptive grouping mechanisms: specifically ViT-L experiences accuracy gains of up to 8% when utilizing DGQA in comparison to GQA and other variants. We further analyze the impact of the number of Key-Value Heads on performance, underscoring the importance of utilizing query-key affinities. Code is available on GitHub., Comment: 11 pages, 9 figures

Electromagnetic follow-up observations of gravitational wave events offer critical insights and provide significant scientific gain from this new class of astrophysical transients. Accurate identification of gravitational wave candidates and rapid release of sky localization information are crucial for the success of these electromagnetic follow-up observations. However, searches for gravitational wave candidates in real time suffer a non-negligible false alarm rate. By leveraging the sky localization information and other metadata associated with gravitational wave candidates, GWSkyNet, a machine learning classifier developed by Cabero et al. (2020), demonstrated promising accuracy for the identification of the origin of event candidates. We improve the performance of the classifier for LIGO-Virgo-KAGRA's fourth observing run by reviewing and updating the architecture and features used as inputs by the algorithm. We also retrain and fine-tune the classifier with data from the third observing run. To improve the prospect of electromagnetic follow-up observations, we incorporate GWSkyNet into LIGO-Virgo-KAGRA's low-latency infrastructure as an automatic pipeline for the evaluation of gravitational wave alerts in real time. We test the readiness of the algorithm on a LIGO-Virgo-KAGRA mock data challenge campaign. The results show that by thresholding on the GWSkyNet score, noise masquerading as astrophysical sources can be rejected efficiently and the majority of true astrophysical signals correctly identified.

Precise identification and detection of the Mental Foramen are crucial in dentistry, impacting procedures such as impacted tooth removal, cyst surgeries, and implants. Accurately identifying this anatomical feature facilitates post-surgery issues and improves patient outcomes. Moreover, this study aims to accelerate dental procedures, elevating patient care and healthcare efficiency in dentistry. This research used Deep Learning methods to accurately detect and segment the Mental Foramen from panoramic radiograph images. Two mask types, circular and square, were used during model training. Multiple segmentation models were employed to identify and segment the Mental Foramen, and their effectiveness was evaluated using diverse metrics. An in-house dataset comprising 1000 panoramic radiographs was created for this study. Our experiments demonstrated that the Classical UNet model performed exceptionally well on the test data, achieving a Dice Coefficient of 0.79 and an Intersection over Union (IoU) of 0.67. Moreover, ResUNet++ and UNet Attention models showed competitive performance, with Dice scores of 0.675 and 0.676, and IoU values of 0.683 and 0.671, respectively. We also investigated transfer learning models with varied backbone architectures, finding LinkNet to produce the best outcomes. In conclusion, our research highlights the efficacy of the classical Unet model in accurately identifying and outlining the Mental Foramen in panoramic radiographs. While vital, this task is comparatively simpler than segmenting complex medical datasets such as brain tumours or skin cancer, given their diverse sizes and shapes. This research also holds value in optimizing dental practice, benefiting practitioners and patients., Comment: 9 pages

A digital currency is money in a digital form. In this model, maintaining integrity of the supply is a core concern, therefore protections against double-spending are often at the heart of a secure digital money scheme. Quantum money exploits the quantum mechanical principle of no-cloning to enable a currency that is immune to double spending. One of the challenges of the scheme is that users require technology that is currently out of reach. Here, we propose a model for quantum currency, which alleviates the need for quantum wallets by delegating quantum storage and processing to an intermediary that we call a "quantum vault". We develop the basic building blocks of this quantum-enabled digital currency and discuss its benefits and challenges., Comment: 11 pages, 4 figures

In a controlled cyber-physical network, such as a power grid, any malicious data injection in the sensor measurements can lead to widespread impact due to the actions of the closed-loop controllers. While fast identification of the attack signatures is imperative for reliable operations, it is challenging to do so in a large dynamical network with tightly coupled nodes. A particularly challenging scenario arises when the cyberattacks are strategically launched during a grid stress condition, caused by non-malicious physical disturbances. In this work, we propose an algorithmic framework -- based on Koopman mode (KM) decomposition -- for online identification and visualization of the cyberattack signatures in streaming time-series measurements from a power network. The KMs are capable of capturing the spatial embedding of both natural and anomalous modes of oscillations in the sensor measurements and thus revealing the specific influences of cyberattacks, even under existing non-malicious grid stress events. Most importantly, it enables us to quantitatively compare the outcomes of different potential cyberattacks injected by an attacker. The performance of the proposed algorithmic framework is illustrated on the IEEE 68-bus test system using synthetic attack scenarios. Such knowledge regarding the detection of various cyberattacks will enable us to devise appropriate diagnostic scheme while considering varied constraints arising from different attacks., Comment: accepted as a work-in-progress paper at the 2024 Annual Conference of the IEEE Industrial Electronics Society (IECON)

Accurate segmentation of the Left Ventricle (LV) holds substantial importance due to its implications in disease detection, regional analysis, and the development of complex models for cardiac surgical planning. CMR is a golden standard for diagnosis of serveral cardiac diseases. LV in CMR comprises of three distinct sections: Basal, Mid-Ventricle, and Apical. This research focuses on the precise segmentation of the LV from Cardiac MRI (CMR) scans, joining with the capabilities of Machine Learning (ML). The central challenge in this research revolves around the absence of a set of parameters applicable to all three types of LV slices. Parameters optimized for basal slices often fall short when applied to mid-ventricular and apical slices, and vice versa. To handle this issue, a new method is proposed to enhance LV segmentation. The proposed method involves using distinct sets of parameters for each type of slice, resulting in a two-phase segmentation approach. The initial phase categorizes images into three groups based on the type of LV slice, while the second phase aims to segment CMR images using parameters derived from the preceding phase. A publicly available dataset (Automated Cardiac Diagnosis Challenge (ACDC)) is used. 10-Fold Cross Validation is used and it achieved a mean score of 0.9228. Comprehensive testing indicates that the best parameter set for a particular type of slice does not perform adequately for the other slice types. All results show that the proposed approach fills a critical void in parameter standardization through a two-phase segmentation model for the LV, aiming to not only improve the accuracy of cardiac image analysis but also contribute advancements to the field of LV segmentation.

The QMA-completeness of the local Hamiltonian problem is a landmark result of the field of Hamiltonian complexity that studies the computational complexity of problems in quantum many-body physics. Since its proposal, substantial effort has been invested in better understanding the problem for physically motivated important families of Hamiltonians. In particular, the QMA-completeness of approximating the ground state energy of local Hamiltonians has been extended to the case where the Hamiltonians are geometrically local in one and two spatial dimensions. Among those physically motivated Hamiltonians, stoquastic Hamiltonians play a particularly crucial role, as they constitute the manifestly sign-free Hamiltonians in Monte Carlo approaches. Interestingly, for such Hamiltonians, the problem at hand becomes more ''classical'', being hard for the class MA (the randomized version of NP) and its complexity has tight connections with derandomization. In this work, we prove that both the two- and one-dimensional geometrically local analogues remain MA-hard with high enough qudit dimension. Moreover, we show that related problems are StoqMA-complete.

The conventional modus operandi for adapting pre-trained vision-language models (VLMs) during test-time involves tuning learnable prompts, ie, test-time prompt tuning. This paper introduces Test-Time Low-rank adaptation (TTL) as an alternative to prompt tuning for zero-shot generalization of large-scale VLMs. Taking inspiration from recent advancements in efficiently fine-tuning large language models, TTL offers a test-time parameter-efficient adaptation approach that updates the attention weights of the transformer encoder by maximizing prediction confidence. The self-supervised confidence maximization objective is specified using a weighted entropy loss that enforces consistency among predictions of augmented samples. TTL introduces only a small amount of trainable parameters for low-rank adapters in the model space while keeping the prompts and backbone frozen. Extensive experiments on a variety of natural distribution and cross-domain tasks show that TTL can outperform other techniques for test-time optimization of VLMs in strict zero-shot settings. Specifically, TTL outperforms test-time prompt tuning baselines with a significant improvement on average. Our code is available at at https://github.com/Razaimam45/TTL-Test-Time-Low-Rank-Adaptation., Comment: Main paper: 11 pages, Supplementary material: 5 pages

Accurate assessment of myocardial tissue stiffness is pivotal for the diagnosis and prognosis of heart diseases. Left ventricular diastolic stiffness ($\beta$) obtained from the end-diastolic pressure-volume relationship (EDPVR) has conventionally been utilized as a representative metric of myocardial stiffness. The EDPVR can be employed to estimate the intrinsic stiffness of myocardial tissues through image-based in-silico inverse optimization. However, whether $\beta$, as an organ-level metric, accurately represents the tissue-level myocardial tissue stiffness in healthy and diseased myocardium remains elusive. We developed a modeling-based approach utilizing a two-parameter material model for the myocardium (denoted by $a_f$ and $b_f$) in image-based in-silico biventricular heart models to generate EDPVRs for different material parameters. Our results indicated a variable relationship between $\beta$ and the material parameters depending on the range of the parameters. Interestingly, $\beta$ showed a very low sensitivity to $a_f$, once averaged across several LV geometries, and even a negative correlation with $a_f$ for small values of $a_f$. These findings call for a critical assessment of the reliability and confoundedness of EDPVR-derived metrics to represent tissue-level myocardial stiffness. Our results also underscore the necessity to explore image-based in-silico frameworks, promising to provide a high-fidelity and potentially non-invasive assessment of myocardial stiffness., Comment: 4 pages, 5 figures, accepted in the IEEE EMBC 2024 conference

Often regarded as the study of how order emerges from randomness, Ramsey theory has played an important role in mathematics and computer science, giving rise to applications in numerous domains such as logic, parallel processing, and number theory. The core of graph Ramsey theory is arrowing: For fixed graphs $F$ and $H$, the $(F, H)$-Arrowing problem asks whether a given graph, $G$, has a red/blue coloring of the edges of $G$ such that there are no red copies of $F$ and no blue copies of $H$. For some cases, the problem has been shown to be coNP-complete, or solvable in polynomial time. However, a more systematic approach is needed to categorize the complexity of all cases. We focus on $(P_3, H)$-Arrowing as $F = P_3$ is the simplest meaningful case for which the complexity question remains open, and the hardness for this case likely extends to general $(F, H)$-Arrowing for nontrivial $F$. In this pursuit, we also gain insight into the complexity of a class of matching removal problems, since $(P_3, H)$-Arrowing is equivalent to $H$-free Matching Removal. We show that $(P_3, H)$-Arrowing is coNP-complete for all $2$-connected $H$ except when $H = K_3$, in which case the problem is in P. We introduce a new graph invariant to help us carefully combine graphs when constructing the gadgets for our reductions. Moreover, we show how $(P_3,H)$-Arrowing hardness results can be extended to other $(F,H)$-Arrowing problems. This allows for more intuitive and palatable hardness proofs instead of ad-hoc constructions of SAT gadgets, bringing us closer to categorizing the complexity of all $(F, H)$-Arrowing problems., Comment: To appear in MFCS 2024

This paper introduces a highly efficient rectenna (rectifying antenna) using a binary optimization algorithm. A novel pixelated receiving antenna has been developed to match the diode impedance of a rectifier, eliminating the need for a separate matching circuit in the rectenna's rectifier. The receiving antenna configuration is fine-tuned via a binary optimization algorithm. A rectenna is designed using optimization algorithm at 2.5 GHz with 38% RF-DC conversion efficiency when subjected to 0 dBm incident power, with an output voltage of 815mV. The proposed rectenna demonstrates versatility across various low-power WPT (wireless power transfer) applications.

Recommender Systems (RS) play an integral role in enhancing user experiences by providing personalized item suggestions. This survey reviews the progress in RS inclusively from 2017 to 2024, effectively connecting theoretical advances with practical applications. We explore the development from traditional RS techniques like content-based and collaborative filtering to advanced methods involving deep learning, graph-based models, reinforcement learning, and large language models. We also discuss specialized systems such as context-aware, review-based, and fairness-aware RS. The primary goal of this survey is to bridge theory with practice. It addresses challenges across various sectors, including e-commerce, healthcare, and finance, emphasizing the need for scalable, real-time, and trustworthy solutions. Through this survey, we promote stronger partnerships between academic research and industry practices. The insights offered by this survey aim to guide industry professionals in optimizing RS deployment and to inspire future research directions, especially in addressing emerging technological and societal trends, Comment: we quarterly update of this literature

In this paper, a joint sensing and communication system is presented for smart agriculture. The system integrates an Ultra-compact Soil Moisture Sensor (UCSMS) for precise sensing, along with a Pattern Reconfigurable Antenna (PRA) for efficient transmission of information to the base station. A multiturn complementary spiral resonator (MCSR) is etched onto the ground plane of a microstrip transmission line to achieve miniaturization. The UCSMS operates at 180 MHz with a 3-turn complementary spiral resonator (3-CSR), at 102 MHz with a 4- turn complementary spiral resonator (4-CSR), and at 86 MHz with a 5-turn complementary spiral resonator (5-CSR). Due to its low resonance frequency, the proposed UCSMS is insensitive to variations in the Volume Under Test (VUT) of soil. A probe-fed circular patch antenna is designed in the Wireless Local Area Network (WLAN) band (2.45 GHz) with a maximum measured gain of 5.63 dBi. Additionally, four varactor diodes are integrated across the slots on the bottom side of the substrate to achieve pattern reconfiguration. Six different radiation patterns have been achieved by using different bias conditions of the diodes. In standby mode, PRA can serve as a means for Wireless Power Transfer (WPT) or Energy Harvesting (EH) to store power in a battery. This stored power can then be utilized to bias the varactor diodes. The combination of UCSMS and PRA enables the realization of a joint sensing and communication system. The proposed system's planar and simple geometry, along with its high sensitivity of 2.05 %, makes it suitable for smart agriculture applications. Moreover, the sensor is adaptive and capable of measuring the permittivity of various Material Under Test (MUT) within the range of 1 to 23.

Modern applications are end-to-end encrypted to prevent data from being read or secretly modified. 5G tech nology provides ubiquitous access to these applications without compromising the application-specific performance and latency goals. In this paper, we empirically demonstrate that 5G radio communication becomes the side channel to precisely infer the user's applications in real-time. The key idea lies in observing the 5G physical and MAC layer interactions over time that reveal the application's behavior. The MAC layer receives the data from the application and requests the network to assign the radio resource blocks. The network assigns the radio resources as per application requirements, such as priority, Quality of Service (QoS) needs, amount of data to be transmitted, and buffer size. The adversary can passively observe the radio resources to fingerprint the applications. We empirically demonstrate this attack by considering four different categories of applications: online shopping, voice/video conferencing, video streaming, and Over-The-Top (OTT) media platforms. Finally, we have also demonstrated that an attacker can differentiate various types of applications in real-time within each category., Comment: 9 pages, 8 figures, 2 tables. This paper has been accepted for publication by the 21st IEEE International Conference on Mobile Ad-Hoc and Smart Systems (MASS 2024)

The prevalence of AI-generated imagery has raised concerns about the authenticity of astronomical images, especially with advanced text-to-image models like Stable Diffusion producing highly realistic synthetic samples. Existing detection methods, primarily based on convolutional neural networks (CNNs) or spectral analysis, have limitations when used independently. We present AstroSpy, a hybrid model that integrates both spectral and image features to distinguish real from synthetic astronomical images. Trained on a unique dataset of real NASA images and AI-generated fakes (approximately 18k samples), AstroSpy utilizes a dual-pathway architecture to fuse spatial and spectral information. This approach enables AstroSpy to achieve superior performance in identifying authentic astronomical images. Extensive evaluations demonstrate AstroSpy's effectiveness and robustness, significantly outperforming baseline models in both in-domain and cross-domain tasks, highlighting its potential to combat misinformation in astronomy.

Existing vision-text contrastive learning models enhance representation transferability and support zero-shot prediction by matching paired image and caption embeddings while pushing unrelated pairs apart. However, astronomical image-label datasets are significantly smaller compared to general image and label datasets available from the internet. We introduce CosmoCLIP, an astronomical image-text contrastive learning framework precisely fine-tuned on the pre-trained CLIP model using SpaceNet and BLIP-based captions. SpaceNet, attained via FLARE, constitutes ~13k optimally distributed images, while BLIP acts as a rich knowledge extractor. The rich semantics derived from this SpaceNet and BLIP descriptions, when learned contrastively, enable CosmoCLIP to achieve superior generalization across various in-domain and out-of-domain tasks. Our results demonstrate that CosmoCLIP is a straightforward yet powerful framework, significantly outperforming CLIP in zero-shot classification and image-text retrieval tasks., Comment: Accepted at SPAICE Conference, ECSAT, UK, 2024

In this paper, we compare general-purpose models, GPT-4-Turbo and Llama-3-8b, with special-purpose models--XLM-Roberta-large, mT5-large, and Llama-3-8b--that have been fine-tuned on specific tasks. We focus on seven classification and seven generation tasks to evaluate the performance of these models on Urdu language. Urdu has 70 million native speakers, yet it remains underrepresented in Natural Language Processing (NLP). Despite the frequent advancements in Large Language Models (LLMs), their performance in low-resource languages, including Urdu, still needs to be explored. We also conduct a human evaluation for the generation tasks and compare the results with the evaluations performed by GPT-4-Turbo, Llama-3-8b and Claude 3.5 Sonnet. We find that special-purpose models consistently outperform general-purpose models across various tasks. We also find that the evaluation done by GPT-4-Turbo for generation tasks aligns more closely with human evaluation compared to the evaluation the evaluation done by Llama-3-8b. This paper contributes to the NLP community by providing insights into the effectiveness of general and specific-purpose LLMs for low-resource languages.

We investigate the effect of initial system-environment correlations to improve the estimation of environment parameters. By employing various physical situations of interest, we present results for the environment temperature and system-environment coupling strength. We consider the spin-spin model whereby a probe (a small controllable quantum system) interacts with a bath of quantum spins and attains a thermal equilibrium state. A projective measurement is then performed to prepare the initial state and allow it to evolve unitarily. The properties of the environment are imprinted upon the dynamics of the probe. The reduced density matrix of the probe state contains a modified decoherence factor and dissipation. This additional factor acts in such a way to improve the estimation of the environment parameters, as quantified by the quantum Fisher information (QFI). In the temperature estimation case, our results are promising as one can improve the precision of the estimates by orders of magnitude by incorporating the effect of initial correlations. The precision increases in the strong coupling regime even if the nearest neighbours' interaction is taken into account. In the case of coupling strength, interestingly the accuracy was found to be continuously increasing in both with and without correlations cases. More importantly, one can see the noticeable role of correlations in improving precision, especially at low temperatures., Comment: Comments Welcome. arXiv admin note: text overlap with arXiv:1808.04988 by other authors

In this study, we introduce the application of causal disparity analysis to unveil intricate relationships and causal pathways between sensitive attributes and the targeted outcomes within real-world observational data. Our methodology involves employing causal decomposition analysis to quantify and examine the causal interplay between sensitive attributes and outcomes. We also emphasize the significance of integrating heterogeneity assessment in causal disparity analysis to gain deeper insights into the impact of sensitive attributes within specific sub-groups on outcomes. Our two-step investigation focuses on datasets where race serves as the sensitive attribute. The results on two datasets indicate the benefit of leveraging causal analysis and heterogeneity assessment not only for quantifying biases in the data but also for disentangling their influences on outcomes. We demonstrate that the sub-groups identified by our approach to be affected the most by disparities are the ones with the largest ML classification errors. We also show that grouping the data only based on a sensitive attribute is not enough, and through these analyses, we can find sub-groups that are directly affected by disparities. We hope that our findings will encourage the adoption of such methodologies in future ethical AI practices and bias audits, fostering a more equitable and fair technological landscape.

Voice signal classification based on human behaviours involves analyzing various aspects of speech patterns and delivery styles. In this study, a real-time dataset collection is performed where participants are instructed to speak twelve psychology questions in two distinct manners: first, in a harsh voice, which is categorized as "misbehaved"; and second, in a polite manner, categorized as "normal". These classifications are crucial in understanding how different vocal behaviours affect the interpretation and classification of voice signals. This research highlights the significance of voice tone and delivery in automated machine-learning systems for voice analysis and recognition. This research contributes to the broader field of voice signal analysis by elucidating the impact of human behaviour on the perception and categorization of voice signals, thereby enhancing the development of more accurate and context-aware voice recognition technologies.

Unmanned Ariel Vehicle (UAV) services with 5G connectivity is an emerging field with numerous applications. Operator-controlled UAV flights and manual static flight configurations are major limitations for the wide adoption of scalability of UAV services. Several services depend on excellent UAV connectivity with a cellular network and maintaining it is challenging in predetermined flight paths. This paper addresses these limitations by proposing a Deep Reinforcement Learning (DRL) framework for UAV path planning with assured connectivity (DUPAC). During UAV flight, DUPAC determines the best route from a defined source to the destination in terms of distance and signal quality. The viability and performance of DUPAC are evaluated under simulated real-world urban scenarios using the Unity framework. The results confirm that DUPAC achieves an autonomous UAV flight path similar to base method with only 2% increment while maintaining an average 9% better connection quality throughout the flight., Comment: 5 pages, 4 figures, Published in the 2024 IEEE Network Operations and Management Symposium (NOMS 2024)

With the proliferation of digitization and its usage in critical sectors, it is necessary to include information about the occurrence and assessment of cyber threats in an organization's threat mitigation strategy. This Cyber Threat Intelligence (CTI) is becoming increasingly important, or rather necessary, for critical national and industrial infrastructures. Current CTI solutions are rather federated and unsuitable for sharing threat information from low-power IoT devices. This paper presents a taxonomy and analysis of the CTI frameworks and CTI exchange platforms available today. It proposes a new CTI architecture relying on the MISP Threat Intelligence Sharing Platform customized and focusing on IoT environment. The paper also introduces a tailored version of STIX (which we call tinySTIX), one of the most prominent standards adopted for CTI data modeling, optimized for low-power IoT devices using the new lightweight encoding and cryptography solutions. The proposed CTI architecture will be very beneficial for securing IoT networks, especially the ones working in harsh and adversarial environments.

Accurate volumetric image registration is highly relevant for clinical routines and computer-aided medical diagnosis. Recently, researchers have begun to use transformers in learning-based methods for medical image registration, and have achieved remarkable success. Due to the strong global modeling capability, Transformers are considered a better option than convolutional neural networks (CNNs) for registration. However, they use bulky models with huge parameter sets, which require high computation edge devices for deployment as portable devices or in hospitals. Transformers also need a large amount of training data to produce significant results, and it is often challenging to collect suitable annotated data. Although existing CNN-based image registration can offer rich local information, their global modeling capability is poor for handling long-distance information interaction and limits registration performance. In this work, we propose a CNN-based registration method with an enhanced receptive field, a low number of parameters, and significant results on a limited training dataset. For this, we propose a residual U-Net with embedded parallel dilated-convolutional blocks to enhance the receptive field. The proposed method is evaluated on inter-patient and atlas-based datasets. We show that the performance of the proposed method is comparable and slightly better than transformer-based methods by using only $\SI{1.5}{\percent}$ of its number of parameters.

Aging and chronic conditions affect older adults' daily lives, making early detection of developing health issues crucial. Weakness, common in many conditions, alters physical movements and daily activities subtly. However, detecting such changes can be challenging due to their subtle and gradual nature. To address this, we employ a non-intrusive camera sensor to monitor individuals' daily sitting and relaxing activities for signs of weakness. We simulate weakness in healthy subjects by having them perform physical exercise and observing the behavioral changes in their daily activities before and after workouts. The proposed system captures fine-grained features related to body motion, inactivity, and environmental context in real-time while prioritizing privacy. A Bayesian Network is used to model the relationships between features, activities, and health conditions. We aim to identify specific features and activities that indicate such changes and determine the most suitable time scale for observing the change. Results show 0.97 accuracy in distinguishing simulated weakness at the daily level. Fine-grained behavioral features, including non-dominant upper body motion speed and scale, and inactivity distribution, along with a 300-second window, are found most effective. However, individual-specific models are recommended as no universal set of optimal features and activities was identified across all participants.

Dielectric optical resonators traditionally rely on materials with the combination of high refractive indices and low optical losses. Such materials are scarce for operation in visible spectrum and shorter wavelengths. This limitation can be circumvented by relaxing the requirement of low losses. We demonstrate that highly lossy dielectric materials can be structured to support optical resonances that confine light in air voids. We theoretically design void resonances in the visible spectrum and identify resonant modes supported by void arrays. Experimentally, we fabricate void arrays in tungsten diselenide and characterize the confined resonances using far-field reflectance measurements and scanning near-field optical microscopy. Using van der Waals heterostructure assembly, we encapsulate the voids with hexagonal boron nitride which reduces the void volume causing a large spectral blue shift of the void resonance exceeding 150 nm. Our work demonstrates a versatile optical platform for lossy materials, expanding the range of suitable materials and the spectral range of photonic devices.

Among recent insights into learning quantum states, online learning and shadow tomography procedures are notable for their ability to accurately predict expectation values even of adaptively chosen observables. In contrast to the state case, quantum process learning tasks with a similarly adaptive nature have received little attention. In this work, we investigate online learning tasks for quantum processes. Whereas online learning is infeasible for general quantum channels, we show that channels of bounded gate complexity as well as Pauli channels can be online learned in the regret and mistake-bounded models of online learning. In fact, we can online learn probabilistic mixtures of any exponentially large set of known channels. We also provide a provably sample-efficient shadow tomography procedure for Pauli channels. Our results extend beyond quantum channels to non-Markovian multi-time processes, with favorable regret and mistake bounds, as well as a shadow tomography procedure. We complement our online learning upper bounds with mistake as well as computational lower bounds. On the technical side, we make use of the multiplicative weights update algorithm, classical adaptive data analysis, and Bell sampling, as well as tools from the theory of quantum combs for multi-time quantum processes. Our work initiates a study of online learning for classes of quantum channels and, more generally, non-Markovian quantum processes. Given the importance of online learning for state shadow tomography, this may serve as a step towards quantum channel variants of adaptive shadow tomography., Comment: 14 + 72 pages, 6 figures

Recent advancements in large language models (LLMs) have greatly enhanced natural language processing (NLP) applications. Nevertheless, these models often inherit biases from their training data. Despite the availability of various datasets, most are limited to one or two NLP tasks (typically classification or evaluation) and lack comprehensive evaluations across a broader range of NLP tasks. To address this gap, we introduce the Bias Evaluations Across Domains (BEADs) dataset, designed to support a wide array of NLP tasks, including text classification, token classification, bias quantification, and benign language generation. A key focus of this paper is the gold label subset of BEADs, an important portion of the data verified by experts to ensure high reliability. BEADs provides data for both fine-tuning, including classification and language generation tasks, and for evaluating LLMs. Our findings indicate that BEADs effectively identifies numerous biases when fine-tuned on this dataset. It also reduces biases when used for fine-tuning language generation task, while preserving language quality. The results also reveal some prevalent demographic biases in LLMs when BEADs is used for evaluation in demographic task. The benchmarking results highlight the efficacy of fine-tuning LLMs for bias identification and the necessity of comprehensive bias evaluation. We make BEADs publicly available to promote more responsible AI development. The dataset can be accessed at https://huggingface.co/datasets/shainar/BEAD ., Comment: under review

Large Language Models (LLMs) have advanced various Natural Language Processing (NLP) tasks, such as text generation and translation, among others. However, these models often generate text that can perpetuate biases. Existing approaches to mitigate these biases usually compromise knowledge retention. This study explores whether LLMs can produce safe, unbiased outputs without sacrificing knowledge or comprehension. We introduce the Safe and Responsible Large Language Model (\textbf{SR}$_{\text{LLM}}$), which has been instruction fine-tuned atop an inherently safe fine-tuned LLM to reduce biases in generated texts. We developed a specialized dataset with examples of unsafe and corresponding safe variations to train \textbf{SR}$_{\text{LLM}}$ to identify and correct biased text. Experiments on our specialized dataset and out-of-distribution test sets reveal that \textbf{SR}$_{\text{LLM}}$ effectively reduces biases while preserving knowledge integrity. This performance surpasses that of traditional fine-tuning of smaller language models and base LLMs that merely reply on prompting techniques. Our findings indicate that instruction fine-tuning is an effective strategy for minimizing bias in LLMs while retaining knowledge. The code and dataset are accessible at \href{https://github.com/shainarazavi/Safe-Responsible-LLM}{SR-LLM}.