arXiv subjects

Praveen Kumar

Publications and source records attributed to Praveen Kumar.

At least 19 recordsLinked to original sources

arXiv2026

Toward Next-Generation AI Data Centers: Power Delivery Architecture Shifts, Emerging Technologies, and Challenges

The rapid growth of AI workloads is driving unprecedented increases in data center power demand, current transients, and thermal stress, exposing fundamental limitations in traditional 48 V rack architectures, low-voltage AC distribution, and line-frequency transformer interfaces. This paper reviews the three stages of architectural shifts required to support next-generation AI data centers and identifies three enabling technological building blocks: high-voltage conversion-ratio DC/DC converters, facility-level low-voltage DC distribution, and medium-voltage solid-state transformers. The advantages, technical challenges, and potential solutions associated with each building block are reviewed. Finally, future research directions and open challenges are discussed.

eess.SY↗

arXiv2026

Isospin Decomposition of Vector and Axial Two-Body Currents via Polarized Electron--Deuteron and Electron--$^3$He Scattering at the Electron-Ion Collider

Two-particle two-hole (2p2h) excitations driven by meson-exchange currents (MEC) are among the leading nuclear uncertainties in long-baseline neutrino oscillation experiments. Three models currently implemented in neutrino event generators disagree by 20--40% on the $\omega$-integrated 2p2h cross section in the dip region on carbon (differential disagreements can reach factors of 2--3), and the axial two-body current has no direct experimental constraint beyond tritium $\beta$-decay at $Q^2 = 0$. We propose a measurement program at the Electron-Ion Collider (EIC) using polarized electron scattering on deuteron and $^3$He. Electromagnetic (EM) scattering ($\gamma^*$ exchange) measures the vector MEC. Charged-current (CC) scattering ($W^-$ exchange) on the same targets measures the vector$+$axial MEC. Subtracting the two provides the first direct sensitivity to the axial two-body current, including the $V$--$A$ interference, as a function of momentum transfer. Using $^3$He (2~$pn$ $+$ 1~$pp$ pair) extends the decomposition to $pp$ pairs. Polarized beams and targets give access to six EM response functions on deuteron, four of which have not been previously measured. The tensor analyzing power provides a sign-flip test for $\Delta$-excitation MEC. We present projected sensitivities at $50 fb^{-1}$ on deuteron ($\sim$5 years at $10^{33}$~cm$^{-2}$s$^{-1}$). The EM program can deliver $\sim\!5\!\times\!10^4$ events per $Q^2$ bin constraining the MEC transverse response to $\sim$2% per bin, the beam--target double-spin asymmetry reaches $6$--$13\sigma$ per bin, and the vector MEC $V_{pn}$ is measured to $\sim$6% per bin. The CC channel is statistics-limited, with $\sim$6--38 events per $Q^2$ bin at $50 fb^{-1}$, requiring a luminosity upgrade beyond the current EIC baseline.

nucl-ex↗

arXiv2026

Socio-Spatial Patterns of Suicide Mortality in the United States

Suicides cause over 49000 deaths yearly in the United States, 55% involving firearms. Suicide mortality exhibits substantial geographical and sociodemographic heterogeneity; yet the role of social networks remains underexplored. To assess how suicide risk and firearm restriction policies propagate through social ties, we integrate county-level suicide mortality data (2010-2022) with the Facebook Social Connectedness Index (SCI). We also examine Extreme Risk Protection Orders (ERPO), state-level policies restricting firearm access for individuals at risk of self-harm. In two-way fixed effects regressions, a one-standard-deviation increase in the SCI-weighted average suicide mortality rate of connected counties was associated with +2.78 deaths per 100,000 in a focal county, while a one-standard-deviation increase in ERPO social exposure was associated with -0.214 deaths per 100,000. These associations persisted when adjusting for geographic proximity and including state-by-year fixed effects, and confirm the effect of social networks on diffusion of both harmful exposures and protective interventions.

stat.AP↗

arXiv2026

Constraining Neutrino--Nucleon Form Factors with Charged-Current Scattering at the Electron-Ion Collider

Next-generation neutrino oscillation experiments such as DUNE require percent-level knowledge of neutrino--nucleon interaction cross sections. The nucleon axial form factor $F_A(Q^2)$, parameterized by the axial mass $\MA$, is the dominant source of uncertainty in the quasi-elastic channel, and the parity-violating structure function $xF_3$ is poorly constrained on free nucleons. We propose using charged-current (CC) electron--proton scattering at the Electron-Ion Collider (EIC) to address both problems simultaneously. The measurement exploits three key features of the EIC: (1)~helicity-selective electron bunches provide \emph{in situ} electromagnetic background rejection; (2)~a longitudinally polarized proton target enables extraction of $F_A(Q^2)$ through the target-spin asymmetry $A_{UL}$; and (3)~the $y$-distribution leverage in CC deep inelastic scattering separates $F_2$ and $xF_3$ on a \emph{free proton}, without nuclear corrections. Using a Fisher-information analysis at $\sqrt{s} = 141\GeV$ with $500\fb^{-1}$ of integrated luminosity, we project the Cram\'{e}r--Rao statistical floor of $\delta\MA \approx 0.03\GeV$ (3\%). Incorporating first-order realistic detector effects: ZDC acceptance, $Q^2$ smearing (5\%), and background noise from helicity subtraction, the projected sensitivity is severely background-limited due to the small signal-to-background ratio ($S/B \approx 3 \times 10^{-4}$) in the elastic channel. Achieving competitive sensitivity ($\delta\MA \approx 0.14\GeV$) would require $\sim\!10^{-7}$ background suppression, three orders of magnitude beyond current projections. The CC DIS $y$-distribution provides sub-percent extraction of $xF_3^{\Wminus}$ over $0.05 < x < 0.5$, representing the most robust electroweak measurement in the near term.

hep-ph↗

arXiv2026

Benchmarking Video Foundation Models for Remote Parkinson's Disease Screening

Video-based assessments offer a scalable pathway for remote Parkinson's disease (PD) screening. While traditional approaches rely on handcrafted features mimicking clinical scales, recent advances in video foundation models (VFMs) enable representation learning without task-specific customization. However, the comparative effectiveness of different VFM architectures across diverse clinical tasks remains poorly understood. We present a large-scale systematic study using a novel video dataset from 1,888 participants (727 with PD), comprising 32,847 videos across 16 standardized clinical tasks. We evaluate seven state-of-the-art VFMs -- including VideoPrism, V-JEPA, ViViT, and VideoMAE -- to determine their robustness in clinical screening. By evaluating frozen embeddings with a linear classification head, we demonstrate that task saliency is highly model-dependent: VideoPrism excels in capturing visual speech kinematics (no audio) and facial expressivity, while V-JEPA proves superior for upper-limb motor tasks. Notably, TimeSformer remains highly competitive for rhythmic tasks like finger tapping. Our experiments yield AUCs of 76.4 - 85.3% and accuracies of 71.5 - 80.6%. While high specificity (up to 90.3%) suggests strong potential for ruling out healthy individuals, the lower sensitivity (43.2 - 57.3%) highlights the need for task-aware calibration and integration of multiple tasks and modalities. Overall, this work establishes a rigorous baseline for VFM-based PD screening and provides a roadmap for selecting suitable tasks and architectures in remote neurological monitoring. Code and anonymized structured data are publicly available: https://anonymous.4open.science/r/parkinson\_video\_benchmarking-A2C5

cs.CV↗

arXiv2026

A Hybrid Supervised-LLM Pipeline for Actionable Suggestion Mining in Unstructured Customer Reviews

Extracting actionable suggestions from customer reviews is essential for operational decision-making, yet these directives are often embedded within mixed-intent, unstructured text. Existing approaches either classify suggestion-bearing sentences or generate high-level summaries, but rarely isolate the precise improvement instructions businesses need. We evaluate a hybrid pipeline combining a high-recall RoBERTa classifier trained with a precision-recall surrogate to reduce unrecoverable false negatives with a controlled, instruction-tuned LLM for suggestion extraction, categorization, clustering, and summarization. Across real-world hospitality and food datasets, the hybrid system outperforms prompt-only, rule-based, and classifier-only baselines in extraction accuracy and cluster coherence. Human evaluations further confirm that the resulting suggestions and summaries are clear, faithful, and interpretable. Overall, our results show that hybrid reasoning architectures achieve meaningful improvements fine-grained actionable suggestion mining while highlighting challenges in domain adaptation and efficient local deployment.

cs.CL↗

arXiv2026

Beyond Sentiment: A Multi-Agent Pipeline for Actionable Business Advice from Reviews

Customer reviews contain valuable signals about service quality, but converting large-scale review corpora into actionable business recommendations remains difficult. Standard sentiment/aspect analysis is largely descriptive, while direct prompting of large language models (LLMs) often yields generic and repetitive advice that is weakly grounded in user feedback. We propose a hierarchical decision-support pipeline that explicitly separates signal compression, problem abstraction, candidate generation, objective-based evaluation, and cost-aware routing into different agents. This architectural decomposition produces auditable intermediate artifacts and enables controllable trade-offs between advice quality and token budget. Experiments on Yelp reviews from three service domains show consistent improvements over single-pass LLM baselines across multiple advice quality dimensions, including actionability, relevance, and non-redundancy. A human evaluation further indicates that users generally prefer our system's recommendations. These results highlight the value of structured agentic decomposition for scalable, cost-aware business decision support.

cs.AI↗

arXiv2026

Hybrid LSTM-UKF Framework: Ankle Angle and Ground Reaction Force Estimation

Accurate prediction of joint kinematics and kinetics is essential for advancing gait analysis and developing intelligent assistive systems such as prosthetics and exoskeletons. This study presents a hybrid LSTM-UKF framework for estimating ankle angle and ground reaction force (GRF) across varying walking speeds. A multimodal sensor fusion strategy integrates force plate data, knee angle, and GRF signals to enrich biomechanical context. Model performance was evaluated using RMSE and $R^2$ under subject-specific validation. The LSTM-UKF consistently outperformed standalone LSTM and UKF models, achieving up to 18.6\% lower RMSE for GRF prediction at 3 km/h. Additionally, UKF integration improved robustness, reducing ankle angle RMSE by up to 22.4\% compared to UKF alone at 1 km/h. These results underscore the effectiveness of hybrid architectures for reliable gait prediction across subjects and walking conditions.

eess.SY↗

arXiv2026

Computationally Efficient Estimation of Localized Treatment Effects for Multi-Level, Multi-Component Interventions to Address the Opioid Crisis

The opioid epidemic remains a major public health challenge in the United States, requiring a multi-pronged intervention approach to mitigate harms to communities. Given the heterogeneity of the epidemic, it is crucial for policymakers to understand localized treatment effects of different intervention components and utilize limited resources efficiently. While locally calibrated simulation models can project epidemic outcomes for any given intervention policy, collecting simulation results for all intervention combinations to estimate localized treatment effects for each community is impractical because the number of combinations grows exponentially with the number of interventions and the levels at which they are applied. To tackle this, we develop a two-stage metamodel framework with a two-step sequential design for efficient sampling. The metamodel consists of a response function linking health outcomes to each intervention component's treatment effect, and a Gaussian process regression (GPR) to learn spatial and socio-economic structures of the treatment effects based on locally-contextualized covariates. With two-step sequential sampling, we leverage spatial correlations and posterior uncertainty to sequentially sample the most informative counties and treatment conditions. We apply this framework to estimate the treatment effects of buprenorphine dispensing and naloxone distribution on overdose mortality rates using a calibrated agent-based opioid epidemic model in Pennsylvania counties. Our approach achieves less than 5% average relative error using fewer than 2% of the runs required for an exhaustive simulation. Our two-stage framework provides a computationally efficient approach to support policymakers, enabling an efficient evaluation of alternative resource-allocation strategies to mitigate the opioid epidemic in local communities.

stat.AP↗

arXiv2025

Near Ultraviolet Transient Explorer (NUTEx): A CubeSat-Based NUV Imaging Payload for Transient Sky Surveys

The Near Ultraviolet Transient Explorer (NUTEx) is a CubeSat-based near-ultraviolet (NUV) imaging payload designed for transient sky surveys and is currently under development. CubeSats are compact and cost-effective satellite platforms that have emerged as versatile tools for scientific exploration and technology demonstrations in space. NUTEx is an imaging telescope operating in the 200-300 nm wavelength range, intended for deployment on a micro-satellite bus. The optical system is based on a Ritchey Chretien (RC) telescope configuration, featuring a 146 mm primary mirror. The detector is a photon-counting microchannel plate (MCP) device with a solar-blind photocathode, paired with an in-house developed readout unit. The instrument has a wide field of view (FoV) of 4 deg, a peak effective area of approximately 18 sq cm at 260 nm, and can reach a sensitivity of 21 AB magnitude (SNR = 5) in a 1200 second exposure. The primary scientific objective of NUTEx is to monitor the night sky for transient phenomena, such as supernova remnants, flaring M-dwarf stars, and other short-timescale events. The payload is currently scheduled for launch in Q2 2026. This paper presents the NUTEx instrument design, outlines its scientific goals and capabilities, and provides an overview of the electronics and mechanical subsystems, including structural analysis.

astro-ph.IM↗

arXiv2025

Vision-Aided Online A* Path Planning for Efficient and Safe Navigation of Service Robots

The deployment of autonomous service robots in human-centric environments is hindered by a critical gap in perception and planning. Traditional navigation systems rely on expensive LiDARs that, while geometrically precise, are semantically unaware, they cannot distinguish a important document on an office floor from a harmless piece of litter, treating both as physically traversable. While advanced semantic segmentation exists, no prior work has successfully integrated this visual intelligence into a real-time path planner that is efficient enough for low-cost, embedded hardware. This paper presents a framework to bridge this gap, delivering context-aware navigation on an affordable robotic platform. Our approach centers on a novel, tight integration of a lightweight perception module with an online A* planner. The perception system employs a semantic segmentation model to identify user-defined visual constraints, enabling the robot to navigate based on contextual importance rather than physical size alone. This adaptability allows an operator to define what is critical for a given task, be it sensitive papers in an office or safety lines in a factory, thus resolving the ambiguity of what to avoid. This semantic perception is seamlessly fused with geometric data. The identified visual constraints are projected as non-geometric obstacles onto a global map that is continuously updated from sensor data, enabling robust navigation through both partially known and unknown environments. We validate our framework through extensive experiments in high-fidelity simulations and on a real-world robotic platform. The results demonstrate robust, real-time performance, proving that a cost-effective robot can safely navigate complex environments while respecting critical visual cues invisible to traditional planners.

cs.RO↗

arXiv2025

Fabrication and Structural Analysis of Trilayers for Tantalum Josephson Junctions with Ta$_2$O$_5$ Barriers

Tantalum (Ta) has emerged as a promising low-loss material, enabling record coherence times in superconducting qubits. This enhanced performance is largely attributed to its stable native oxide, which may host fewer two-level system (TLS) defects, which are the key contributors to decoherence in superconducting circuits. Nevertheless, aluminum oxide remains the predominant choice for Josephson junction (JJ) barriers in most qubit architectures. Here, we investigate techniques for forming high-quality oxide layers on $\alpha$-phase tantalum films to develop tantalum-oxide JJ barriers. We explore thermal oxidation in a tube furnace, rapid thermal annealing, and plasma oxidation of both room-temperature and heated Ta films, characterize the resulting structures using X-ray techniques and electron microscopy, and propose a mechanistic picture of the oxidation pathways. We find that plasma oxidation provides the smoothest Ta$_2$O$_5$ layers, is compatible with in situ Ta deposition, and offers thickness control through the annealing temperature, advantageous for JJ fabrication. Lastly, we evaluate methods for growing Ta/TaO$_x$/Ta trilayers. All trilayers showed c-axis-oriented columnar growth of the bottom Ta layer, with sapphire substrates producing larger, better-aligned grains yet higher dislocation densities than silicon. Nucleation of c-axis-oriented $\alpha$-Ta on tantalum-oxide required an Nb seed layer, as direct Ta deposition yielded amorphous Ta. These results demonstrate the feasibility of $\alpha$-Ta/Nb/TaO$_x$/$\alpha$-Ta stacks for JJs with clean interfaces.

cond-mat.supr-con↗

arXiv2025

ReviewSense: Transforming Customer Review Dynamics into Actionable Business Insights

As customer feedback becomes increasingly central to strategic growth, the ability to derive actionable insights from unstructured reviews is essential. While traditional AI-driven systems excel at predicting user preferences, far less work has focused on transforming customer reviews into prescriptive, business-facing recommendations. This paper introduces ReviewSense, a novel prescriptive decision support framework that leverages advanced large language models (LLMs) to transform customer reviews into targeted, actionable business recommendations. By identifying key trends, recurring issues, and specific concerns within customer sentiments, ReviewSense extends beyond preference-based systems to provide businesses with deeper insights for sustaining growth and enhancing customer loyalty. The novelty of this work lies in integrating clustering, LLM adaptation, and expert-driven evaluation into a unified, business-facing pipeline. Preliminary manual evaluations indicate strong alignment between the model's recommendations and business objectives, highlighting its potential for driving data-informed decision-making. This framework offers a new perspective on AI-driven sentiment analysis, demonstrating its value in refining business strategies and maximizing the impact of customer feedback.

cs.AI↗

arXiv2025

Compact Binary Coalescence Sensitivity Estimates with Injection Campaigns during the LIGO-Virgo-KAGRA Collaborations' Fourth Observing Run

We describe the effort to characterize gravitational-wave searches and detector sensitivity to different types of compact binary coalescences during the LIGO-Virgo-KAGRA Collaborations' fourth observing run. We discuss the design requirements and example use cases for this data product, constructed from $> 4.33\times10^8$ injections during O4a alone. We also identify subtle effects with high confidence, like diurnal duty cycles within detectors. This paper accompanies a public data release of the curated injection set, and the appendixes give detailed examples of how to use the publicly available data.

gr-qc↗

arXiv2025

Exploiting solute segregation and partitioning to the deformation-induced planar defects and nano-martensite in designing ultra-strong Co-Ni base alloys

Single-phase, multi-elements (three or more) with high concentrations show exceptional tensile strength up to ~ 0.8-1.2 GPa. However, they possess a very low 0.2% yield strength (YS), i.e., they can be permanently deformed at very low-stress levels of 300 to 600 MPa. Here, we reveal by exploiting atomic-scale solute interactions with the deformation-induced structures to design ultra-strong single-phase alloys with YS > 2 GPa. This was achieved by controlled thermomechanical processing that introduces stacking-faults (SFs), nano-twins (NTs), and nano-martensite {\epsilon}-laths (NMLs) during cold deformation followed by facilitating solute segregation/partitioning to them by tempering at intermediate temperature. We demonstrate the phenomena in a low stacking faulty energy multi-component (face-centered-cubic, fcc structured) Co-33Ni-24Cr alloy (all in at.%) containing 5at.% Mo as a solute. It is also shown that the degree of strengthening after tempering scales up with the fraction of these structures (before tempering) in the alloy microstructure that can be tuned by the amount and temperature of cold deformation. Cold-rolling with 45% and 65% thickness reduction, followed by tempering at 600{\deg}C for 4 hours, led to an YS of 1.5 GPa and 2 GPa with elongation to fracture (%El) 14% and 7%, respectively. The YS is further enhanced to ~ 2.2 GPa without reduction in %El upon cryo-rolling followed by tempering. The alloy microstructure is stable at 600{\deg}C up to 100 hours and also retains an YS of ~ 1.5 GPa with %El of 18% during tensile test at 600{\deg}C. The derived high YS and high-temperature stability are critically a consequence of solute partitioning to the NMLs that we termed as Solute-Partitioned NMLs (SP-NMLs) in the microstructure.

cond-mat.mtrl-sci↗

arXiv2025

High-strength and ductile lightweight cast aluminium alloys with superlattice nano-layered fibres (SNL) and core-shell nano-particles

Lightweight, high-strength structural materials are component enablers in transportation and aerospace, reducing carbon footprints and enhancing fuel efficiency. Cast aluminium alloys, mainly based on eutectic compositions, make up 85% of these materials but often fail catastrophically due to inefficient load transfer across the interfaces between the brittle eutectic phase and the ductile matrix. Here, we discovered that promoting a superlattice nano-layer (SNL) around the eutectic fibres, achieved by adding Zr to an Al-Gd near-eutectic alloy, enables excellent load transfer capabilities, resulting in a 400% increase in tensile ductility. The primary Al matrix also contains a high number density of superlattice core-shell nano-particles. This exceptional increase in formability is attributed to the ability of the SNL to prevent dislocations from accumulating at the weak and brittle eutectic fibre/matrix interfaces, thereby avoiding stress concentrations that would otherwise initiate fibre breakage and debonding. The core-shell nano-particles in Al cause a large number of dislocation cross/multiple-slips on {111} planes, forming ultra-fine (12 nm) dislocation networks that leverage substantial plastic strain accumulation. This atomic interface design overcomes the ductility limitations of cast-eutectic alloys, enabling them for structural applications.

cond-mat.mtrl-sci↗

arXiv2025

Unsupervised Latent Pattern Analysis for Estimating Type 2 Diabetes Risk in Undiagnosed Populations

The global prevalence of diabetes, particularly type 2 diabetes mellitus (T2DM), is rapidly increasing, posing significant health and economic challenges. T2DM not only disrupts blood glucose regulation but also damages vital organs such as the heart, kidneys, eyes, nerves, and blood vessels, leading to substantial morbidity and mortality. In the US alone, the economic burden of diagnosed diabetes exceeded \$400 billion in 2022. Early detection of individuals at risk is critical to mitigating these impacts. While machine learning approaches for T2DM prediction are increasingly adopted, many rely on supervised learning, which is often limited by the lack of confirmed negative cases. To address this limitation, we propose a novel unsupervised framework that integrates Non-negative Matrix Factorization (NMF) with statistical techniques to identify individuals at risk of developing T2DM. Our method identifies latent patterns of multimorbidity and polypharmacy among diagnosed T2DM patients and applies these patterns to estimate the T2DM risk in undiagnosed individuals. By leveraging data-driven insights from comorbidity and medication usage, our approach provides an interpretable and scalable solution that can assist healthcare providers in implementing timely interventions, ultimately improving patient outcomes and potentially reducing the future health and economic burden of T2DM.

cs.LG↗

arXiv2025

Harnessing Layer-Controlled Two-dimensional Semiconductors for Photoelectrochemical Energy Storage via Quantum Capacitance and Band Nesting

Two-dimensional (2D) transition metal dichalcogenides like molybdenum diselenide (MoSe$_2$) have shown great potential in optoelectronics and energy storage due to their layer-dependent bandgap. However, producing high-quality 2D MoSe$_2$ layers in a scalable and controlled manner remains challenging. Traditional methods, such as hydrothermal and liquid-phase exfoliation, lack precision and understanding at the nanoscale, limiting further applications. Atmospheric pressure chemical vapor deposition (APCVD) offers a scalable solution for growing high-quality, large-area, layer-controlled 2D MoSe$_2$. Despite this, the photoelectrochemical performance of APCVD-grown 2D MoSe$_2$, particularly in energy storage, has not been extensively explored. This study addresses this by examining MoSe$_2$'s layer-dependent quantum capacitance and photo-induced charge storage properties. Using a three-electrode setup in 0.5M H$_2$SO$_4$, we observed a layer-dependent increase in areal capacitance under both dark and illuminated conditions. A six-layer MoSe$_2$ film exhibited the highest capacitance, reaching $96 \mu\mathrm{F/cm^2}$ in the dark and $115 \mu\mathrm{F/cm^2}$ under illumination at a current density of $5 \mu\mathrm{A/cm^2}$. Density Functional Theory (DFT) and Many-Body Perturbation Theory calculations reveal that Van Hove singularities and band nesting significantly enhance optical absorption and quantum capacitance. These results highlight APCVD-grown 2D MoSe$_2$'s potential as light-responsive, high-performance energy storage electrodes, paving the way for innovative energy storage systems.

cond-mat.mtrl-sci↗