arXiv subjects

Dong-Hyun Kim

Publications and source records attributed to Dong-Hyun Kim.

17 recordsLinked to original sources

arXiv2026

Three-temperature atomistic spin-lattice dynamics in LAMMPS: a moment-consistent, fluctuation-dissipation-correct extension and its validation on ultrafast demagnetization and all-optical switching of GdFeCo

Atomistic spin dynamics (ASD) codes like VAMPIRE simulate femtosecond-laser-induced ultrafast demagnetization and all-optical switching (AOS) on a rigid lattice; the SPIN package of the Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) instead propagates spins and lattice together, but until now could couple its spin thermostat only to a single global temperature, precluding three-temperature (3TM) simulations. We present a validated 3TM extension with two new fixes: langevin/spin/ttm, coupling the stochastic spin bath to the local electron-temperature field of fix ttm; and moment/scale/spin, supplying the per-atom moment-dependent ($1/μ_i$) prefactor required for heterogeneous-moment systems, with deterministic terms scaling as $1/μ_i$ and, as required by the fluctuation-dissipation theorem, stochastic noise scaling as $1/\sqrt{μ_i}$. We validate the framework on single-species benchmarks-precession, the bcc-Fe Curie curve, a continuous demagnetization-remagnetization-precession trajectory, and a lattice-strain acoustic-phonon pulse unavailable to spin-only ASD codes-then on heat-induced AOS of GdFeCo with literature parameters (Radu/Ostler exchange constants, moments, and damping). The corrected integrator reproduces the experimentally reported transient ferromagnetic-like sublattice alignment, field-insensitive thermal switching, a switching probability approaching unity across a genuine 8x8 literature damping phase diagram, a non-monotonic critical-cooling-duration boundary that disappears above $T_e^0 \approx 2000 \text{ K}$, a composition-dependent switching window centered on angular-momentum compensation, and convergence in system size with monotonic, physical damping dependence. These results establish LAMMPS as a quantitatively validated platform for three-temperature spin-lattice simulations of ultrafast magnetism.

cond-mat.mtrl-sci↗

arXiv2026

DRIFT: Difficulty-aware Rectified Flows for Through-plane MRI Super-Resolution

Magnetic Resonance Imaging (MRI) is often acquired with anisotropic resolution to reduce scan time, producing stair-step artifacts along the through-plane direction. In through-plane MRI super-resolution, an efficiency-fidelity trade-off arises: feed-forward regressors are fast but oversmooth at large slice-thicknesses, while sampling-based methods improve fidelity at high inference cost. We propose DRIFT, a two-stage thickness-conditioned rectified flow framework for through-plane MRI super-resolution with continuous input slice-thickness. Stage 1 employs an Anatomical Projection Network (APN) to map low-resolution patches to a coarse high-resolution manifold, providing a deterministic anatomical initialization that shortens the residual transport of Stage 2 and stabilizes slice-wise refinement. Stage 2 refines details via rectified flow and introduces a Physics-Aware Difficulty (PAD) metric derived from slice-thickness induced through-plane bandwidth deficit to guide an Adaptive Integration Scheduler (AIS), allocating ODE steps by thickness. A Consistent Endpoint Trajectory Alignment (CETA) loss enforces thickness-consistent reconstructions. Experiments show that DRIFT outperforms super-resolution baselines while reducing inference cost. Code, models, and interactive demos are available at https://yoonseokchoi-ai.github.io/drift-eccv2026/.

cs.CV↗

arXiv2026

Universal Operational Privacy in Distributed Quantum Sensing

Privacy is a fundamental requirement in distributed quantum sensor networks, where multiple clients estimate spatially distributed parameters using shared quantum resources while interacting with potentially untrusted servers. Despite its importance, existing privacy conditions rely on idealized quantum bounds and do not fully capture the operational constraints imposed by realistic measurements. Here, we introduce a universal operational privacy framework for distributed quantum sensing, formulated in terms of the experimentally accessible Fisher information matrix and applicable to arbitrary protocols characterized by singular information structures. The proposed condition provides a protocol-independent criterion, ensuring that no information about individual parameters is accessible to untrusted parties. We further experimentally demonstrate that a distributed quantum sensing protocol employing fewer photons than the number of estimated parameters simultaneously satisfies the universal privacy condition and achieves Heisenberg-limited precision. Our results establish universal operational constraints governing privacy in distributed quantum sensor networks and provide a foundation for practical, privacy-preserving quantum sensing beyond full-rank regimes.

quant-ph↗

arXiv2025

Improving Noise Robust Audio-Visual Speech Recognition via Router-Gated Cross-Modal Feature Fusion

Robust audio-visual speech recognition (AVSR) in noisy environments remains challenging, as existing systems struggle to estimate audio reliability and dynamically adjust modality reliance. We propose router-gated cross-modal feature fusion, a novel AVSR framework that adaptively reweights audio and visual features based on token-level acoustic corruption scores. Using an audio-visual feature fusion-based router, our method down-weights unreliable audio tokens and reinforces visual cues through gated cross-attention in each decoder layer. This enables the model to pivot toward the visual modality when audio quality deteriorates. Experiments on LRS3 demonstrate that our approach achieves an 16.51-42.67% relative reduction in word error rate compared to AV-HuBERT. Ablation studies confirm that both the router and gating mechanism contribute to improved robustness under real-world acoustic noise.

cs.CV↗

arXiv2025

Distributed quantum sensing with multi-mode $N00N$ states

Distributed quantum sensing, which estimates a global parameter across distant nodes, has attracted significant interest for applications such as quantum imaging, sensor networks, and global-scale clock synchronization. $N00N$ states are regarded as one of the optimal quantum resources for quantum metrology, enabling the Heisenberg scaling. Recently, the concept of $N00N$ states has been extended to multi-mode $N00N$ states for quantum-enhanced multiple-parameter estimation. However, the application of multi-mode $N00N$ states in distributed quantum sensing remains unexplored. Here, we propose a distributed quantum sensing scheme that achieves the Heisenberg scaling using multi-mode $N00N$ states. We theoretically show that multi-mode $N00N$ states can reach the Heisenberg scaling by examining both the Cramér-Rao bound and the quantum Cramér-Rao bound. For experimental demonstration, we employ a four-mode $2002$ state to estimate the average of two spatially distributed phases, achieving a 2.74 dB sensitivity enhancement over the standard quantum limit. We believe that utilizing multi-mode $N00N$ states for distributed quantum sensing offers a promising approach for developing entanglement-enhanced sensor networks.

quant-ph↗

arXiv2024

Deformation-Aware Segmentation Network Robust to Motion Artifacts for Brain Tissue Segmentation using Disentanglement Learning

Motion artifacts caused by prolonged acquisition time are a significant challenge in Magnetic Resonance Imaging (MRI), hindering accurate tissue segmentation. These artifacts appear as blurred images that mimic tissue-like appearances, making segmentation difficult. This study proposes a novel deep learning framework that demonstrates superior performance in both motion correction and robust brain tissue segmentation in the presence of artifacts. The core concept lies in a complementary process: a disentanglement learning network progressively removes artifacts, leading to cleaner images and consequently, more accurate segmentation by a jointly trained motion estimation and segmentation network. This network generates three outputs: a motioncorrected image, a motion deformation map that identifies artifact-affected regions, and a brain tissue segmentation mask. This deformation serves as a guidance mechanism for the disentanglement process, aiding the model in recovering lost information or removing artificial structures introduced by the artifacts. Extensive in-vivo experiments on pediatric motion data demonstrate that our proposed framework outperforms state-of-the-art methods in segmenting motion-corrupted MRI scans.

eess.IV↗

arXiv2024

Investigation of Electrical Conductivity Changes during Brain Functional Activity in 3T MRI

Blood oxygenation level-dependent (BOLD) functional magnetic resonance imaging (fMRI) is widely used to visualize brain activation regions by detecting hemodynamic responses associated with increased metabolic demand. While alternative MRI methods have been employed to monitor functional activities, the investigation of in-vivo electrical property changes during brain function remains limited. In this study, we explored the relationship between fMRI signals and electrical conductivity (measured at the Larmor frequency) changes using phase-based electrical properties tomography (EPT). Our results revealed consistent patterns: conductivity changes showed negative correlations, with conductivity decreasing in the functionally active regions whereas B1 phase mapping exhibited positive correlations around activation regions. These observations were consistent across both motor and visual cortex activations. To further substantiate these findings, we conducted electromagnetic radio-frequency simulations that modeled activation states with varying conductivity, which demonstrated trends similar to our in-vivo results for both B1 phase and conductivity. These findings suggest that in-vivo electrical conductivity changes can indeed be measured during brain activity. However, further investigation is needed to fully understand the underlying mechanisms driving these measurements.

q-bio.QM↗

arXiv2024

Persistent quantum advantage with definite photon-number states in lossy multiple-phase estimation

Multiple-phase estimation exploiting quantum states has broad applications in novel sensing and imaging technologies. However, the unavoidable presence of lossy environments in practical settings often diminishes the precision of phase estimations. To address this challenge, we propose an optimal multiple-phase estimation scheme that is inherently robust against photon loss, ensuring a persistent quantum advantage across all levels of photon loss. The scheme employs a multi-mode definite photon-number (DPN) state with weights optimized for given levels of photon loss. We theoretically demonstrate that the DPN state can sustain quantum enhancement in estimation precision under all levels of photon loss, compared to the classical benchmark that employs a coherent state input. The proposed scheme using DPN states generalizes earlier studies employing NOON states, which are only optimal when photon loss is small. We believe that our study, demonstrating persistent robustness to photon loss, paves the way for significant advancements in quantum-enhanced sensing technologies, enabling practical applications and quantum advantages in real-world scenarios.

quant-ph↗

arXiv2024

Optimal multiple-phase estimation with multi-mode NOON states against photon loss

Multi-mode NOON states can quantum-enhance multiple-phase estimation in the absence of photon loss. However, a multi-mode NOON state is known to be vulnerable to photon loss, and its quantum-enhancement can be dissipated by lossy environment. In this work, we demonstrate that a quantum advantage in estimate precision can still be achieved in the presence of photon loss. This is accomplished by optimizing the weights of the multi-mode NOON states according to photon loss rates in the multiple modes, including the reference mode which defines the other phases. For practical relevance, we also show that photon-number counting via a multi-mode beam-splitter achieves the useful, albeit sub-optimal, quantum advantage. We expect this work to provide valuable guidance for developing quantum-enhanced multiple-phase estimation techniques in lossy environments.

quant-ph↗

arXiv2023

Toward Automated Detection of Microbleeds with Anatomical Scale Localization: A Complete Clinical Diagnosis Support Using Deep Learning

Cerebral Microbleeds (CMBs) are chronic deposits of small blood products in the brain tissues, which have explicit relation to various cerebrovascular diseases depending on their anatomical location, including cognitive decline, intracerebral hemorrhage, and cerebral infarction. However, manual detection of CMBs is a time-consuming and error-prone process because of their sparse and tiny structural properties. The detection of CMBs is commonly affected by the presence of many CMB mimics that cause a high false-positive rate (FPR), such as calcification and pial vessels. This paper proposes a novel 3D deep learning framework that does not only detect CMBs but also inform their anatomical location in the brain (i.e., lobar, deep, and infratentorial regions). For the CMB detection task, we propose a single end-to-end model by leveraging the U-Net as a backbone with Region Proposal Network (RPN). To significantly reduce the FPs within the same single model, we develop a new scheme, containing Feature Fusion Module (FFM) that detects small candidates utilizing contextual information and Hard Sample Prototype Learning (HSPL) that mines CMB mimics and generates additional loss term called concentration loss using Convolutional Prototype Learning (CPL). The anatomical localization task does not only tell to which region the CMBs belong but also eliminate some FPs from the detection task by utilizing anatomical information. The results show that the proposed RPN that utilizes the FFM and HSPL outperforms the vanilla RPN and achieves a sensitivity of 94.66% vs. 93.33% and an average number of false positives per subject (FPavg) of 0.86 vs. 14.73. Also, the anatomical localization task further improves the detection performance by reducing the FPavg to 0.56 while maintaining the sensitivity of 94.66%.

eess.IV↗

arXiv2021

Stacked U-Nets with Self-Assisted Priors Towards Robust Correction of Rigid Motion Artifact in Brain MRI

In this paper, we develop an efficient retrospective deep learning method called stacked U-Nets with self-assisted priors to address the problem of rigid motion artifacts in MRI. The proposed work exploits the usage of additional knowledge priors from the corrupted images themselves without the need for additional contrast data. The proposed network learns missed structural details through sharing auxiliary information from the contiguous slices of the same distorted subject. We further design a refinement stacked U-Nets that facilitates preserving of the image spatial details and hence improves the pixel-to-pixel dependency. To perform network training, simulation of MRI motion artifacts is inevitable. We present an intensive analysis using various types of image priors: the proposed self-assisted priors and priors from other image contrast of the same subject. The experimental analysis proves the effectiveness and feasibility of our self-assisted priors since it does not require any further data scans.

eess.IV↗

arXiv2020

The dynamics of a domain wall in ferrimagnets driven by spin-transfer torque

The spin-transfer-torque-driven (STT-driven) dynamics of a domain wall in an easy-axis rare-earth transition-metal ferrimagnet is investigated theoretically and numerically in the vicinity of the angular momentum compensation point $T_A$, where the net spin density vanishes. The particular focus is given on the unusual interaction of the antiferromagnetic dynamics of a ferrimagnetic domain wall and the adiabatic component of STT, which is absent in antiferromagnets but exists in the ferrimagnets due to the dominant coupling of conduction electrons to transition-metal spins. Specifically, we first show that the STT-induced domain-wall velocity changes its sign across $T_A$ due to the sign change of the net spin density, giving rise to a phenomenon unique to ferrimagnets that can be used to characterize $T_A$ electrically. It is also shown that the frequency of the STT-induced domain-wall precession exhibits its maximum at $T_A$ and it can approach the spin-wave gap at sufficiently high currents. Lastly, we report a numerical observation that, as the current density increases, the domain-wall velocity starts to deviate from the linear-response result, calling for a more comprehensive theory for the domain-wall dynamics in ferrimagnets driven by a strong current.

cond-mat.mes-hall↗

arXiv2019

Magnetic Domain Wall Motion due to AC Bias-Driven Resonances

Most of the existing researches on the dynamics of a domain wall (DW) have focused on the effect of DC biases, where the induced velocity is determined by the bias strength. Here we show that AC biases such as a field or a current are also able to move a DW via synchronization between the DW angle and the phase of the AC bias. The resulting DW velocity is proportional to the driving frequency of the AC bias, but independent of the bias strength, offering potentially low-power operations of DW devices. The AC-bias-driven DW motion is shown to exhibit a phase locking-unlocking transition, a critical phenomenon akin to the Walker breakdown of a DC-bias-driven DW motion. Our work shows that a DW can be driven resonantly by synchronizing its angle to AC biases, shedding a light on hitherto overlooked utility of internal degree of freedom for driving magnetic textures.

cond-mat.mes-hall↗

arXiv2015

RC-circuit-like dynamic characteristic of the magnetic domain wall in ferromagnetic nanowires

We have investigated dynamic behaviors of the magnetic domain wall under perpendicular magnetic field pulses in ferromagnetic nanowires using micromagnetic simulations. It has been found that the perpendicular magnetic field pulse can trigger the magnetic domain wall motion, where all the field torques are kept to be on the plane of nanowire strip. The magnetic domain wall speed faster than several hundreds meters per second is predicted without the Walker breakdown for the perpendicular magnetic driving field stronger than $200~\mathrm{mT}$. Interestingly, the dynamic behavior of the moving magnetic domain wall driven by perpendicular magnetic field pulses is explained by charging- and discharging-like behaviors of an electrical RC-circuit model, where the charging and the discharging of "magnetic charges" on the nanowire planes are considered. The concept of the RC-model-like dynamic characteristic of the magnetic domain wall might be promising for spintronic functional device applications based on the magnetic domain wall motion.

cond-mat.mes-hall↗

arXiv2014

Perpendicular magnetic anisotropy and magnetization process in CoFeB/Pd multilayer films

Perpendicular magnetic anisotropy (PMA) and dynamic magnetization reversal process in [CoFeB $t$ nm/Pd 1.0 nm]$_n$ ($t$ = 0.4, 0.6, 0.8, 1.0, and 1.2 nm; $n$ = 2 - 20) multilayer films have been studied by means of magnetic hysteresis and Kerr effect measurements. Strong and controllable PMA with an effective uniaxial anisotropy up to 7.7$\times$ 10$^6$ J.m$^{-3}$ and a saturation magnetization as low as 200 emu/cc are achieved. Surface/interfacial anisotropy of CoFeB/Pd interfaces, the main contribution to the PMA, is separated from the effective uniaxial anisotropy of the films, and appears to increase with the number of the CoFeB/Pd bilayers. Observation of the magnetic domains during a magnetization reversal process using polar magneto-optical Kerr microscopy shows the detailed behavior of nucleation and displacement of the domain walls.

cond-mat.mes-hall↗

arXiv2010

Hybrid Workflow Policy Management for Heart Disease Identification

As science technology grows, medical application is becoming more complex to solve the physiological problems within expected time. Workflow management systems (WMS) in Grid computing are promising solution to solve the sophisticated problem such as genomic analysis, drug discovery, disease identification, etc. Although existing WMS can provide basic management functionality in Grid environment, consideration of user requirements such as performance, reliability and interaction with user is missing. In this paper, we propose hybrid workflow management system for heart disease identification and discuss how to guarantee different user requirements according to user SLA. The proposed system is applied to Physio-Grid e-health platform to identify human heart disease with ECG analysis and Virtual Heart Simulation (VHS) workflow applications.

cs.OH↗

arXiv2008

Current-Induced Resonant Motion of a Magnetic Vortex Core: Effect of Nonadiabatic Spin Torque

The current-induced resonant excitation of a magnetic vortex core is investigated by means of analytical and micromagnetic calculations. We find that the radius and the phase shift of the resonant motion are not correctly described by the analytical equations because of the dynamic distortion of a vortex core. In contrast, the initial tilting angle of a vortex core is free from the distortion and determined by the nonadiabaticity of the spin torque. It is insensitive to experimentally uncontrollable current-induced in-plane Oersted field. We propose that a time-resolved imaging of the very initial trajectory of a core is essential to experimentally estimate the nonadiabaticity.

cond-mat.other↗