arXiv subjects

Dongwook Kim

Publications and source records attributed to Dongwook Kim.

At least 19 recordsLinked to original sources

arXiv2026

Two distinct gap structures in the mid-infrared optical conductivity of the Hubbard model

Mid-infrared (MIR) optical conductivities in cuprate superconductors show universal features. In this study, we demonstrate that these originate from two distinct gaps: the pseudogap mediated by antiferromagnetic spin fluctuations and the Mott-Hubbard gap. The MIR spectra from these two gaps show a characteristic and distinct shape and doping dependence, and can thus be distinguished. Specifically, the Mott-Hubbard correlations, while rendering a peak in the visible spectrum for small dopings, yield a broad optical response starting in the MIR when a ``waterfall'' develops in the one-particle spectrum at larger dopings. The numerically and analytically determined doping-dependence of both gaps consistently reproduce experiments, and provide a unified microscopic understanding of the cuprate MIR optical conductivity.

cond-mat.str-el↗

arXiv2026

Universal Signature of Hundness and Its Quantification

Hund's coupling $J$ induces fundamentally different correlation effects from Hubbard $U$. This leads to a violation of the Brinkman--Rice scenario and the emergence of a Janus-faced phase owing to its low-energy effectiveness, in which band renormalization is confined below a characteristic energy scale. We propose a quantitative framework to capture low-energy effectiveness through two correlation factors: $z_L$ for low-energy quasiparticle renormalization and $z_H$ for high-energy charge fluctuations, newly introduced in this study. The discrepancy between $z_L$ and $z_H$ reflects the Hund character of the correlation. By establishing a one-to-one correspondence between correlation factors and the spin and charge susceptibilities, we identify the spin-degree-of-freedom effectiveness as the microscopic origin of low-energy effectiveness. Our framework, validated across multiorbital models and real materials, provides a universal and quantitative measure of Hundness.

cond-mat.str-el↗

arXiv2026

JustDepth: Real-Time Radar-Camera Depth Estimation with Single-Scan LiDAR Supervision

Accurate yet low-latency depth is essential for radar-camera perception in autonomous systems. Cameras provide rich appearance but lack metric scale, whereas automotive radar offers metric range but is sparse and noisy. Many pipelines are multi-stage or depend on auxiliary annotations, increasing latency and limiting portability. We introduce JustDepth, a single-stage radar-camera depth estimator trained only with radar, camera, and single-scan LiDAR. All radar returns are aggregated into a fixed-width 1D representation, decoupling runtime from point count. A Height Fusion Block fuses modalities, a lightweight GNN propagates depth globally, and a training-only confidence decoder stabilizes learning with zero test-time cost. We mitigate stripe artifacts via simple augmentations and quantify them using the Vertical-Horizontal Gradient Ratio (VHGR). On nuScenes, compared to recent state-of-the-art methods, JustDepth maintains accuracy while reducing inference time by 39.7x and stripe artifacts by 66% as measured by VHGR.

cs.CV↗

arXiv2026

TriALS: Triphasic-Aided Liver Lesion Segmentation Benchmark in Non-Contrast CT

Automated segmentation of liver lesions on non-contrast computed tomography (NCCT) is clinically important but fundamentally challenging, particularly in low-resource settings across Africa and Asia where contrast agents are frequently unavailable. Progress has been limited by the absence of annotated NCCT benchmarks. Here we describe the TriALS challenge for automated liver lesion segmentation under contrast-limited conditions, supported by a multi-centre dataset of 150 cases with four-phase CT acquisitions (600 volumes) from Egyptian and Chinese institutions. Algorithms were evaluated on 70 cases from three institutions, including an independent external cohort. The top-performing method achieved a mean venous-phase Dice of 0.754, consistent with human-level performance, yet dropped to 0.57 on NCCT. On external validation, the leading method outperformed off-the-shelf models by up to 28% in Dice on NCCT. Algorithm performance was most strongly predicted by training data scale and pre-training strategy. A cross-year comparison exposed a persistent perceptual barrier on NCCT that scaling pre-training alone cannot overcome. Data, annotations, and code are available at https://github.com/xmed-lab/TriALS.

cs.CV↗

arXiv2026

Correlation-driven tunability of altermagnetism in RuO$_2$

RuO$_2$ has been regarded as a prototypical candidate for metallic altermagnet, offering a potential platform for high-speed and high-efficiency spintronics. However, the magnetic ground state of RuO$_2$ remains a topic of active debate due to conflicting experimental reports. In this work, we investigate the effect of electron correlations in RuO$_2$ using density functional theory combined with dynamical mean-field theory (DFT+DMFT). In contrast to previous DFT-based studies, DFT+DMFT captures essential dynamical correlation effects, yielding spectral functions and optical conductivities in excellent quantitative agreement with experiments, and further reveals that RuO$_2$ resides in the close vicinity of both the paramagnetic-altermagnetic phase boundary and the itinerant-localized crossover, rendering the magnetic ground state highly susceptible to external perturbations. Indeed, even a minimal compressive strain of $\sim$0.5% is sufficient to drive the system into an altermagnetic phase. These findings elucidate the origin of the conflicting experimental observations and reveal that dynamical correlation effects are the key driving force behind the highly tunable magnetic ground state of RuO$_2$.

cond-mat.mtrl-sci↗

arXiv2026

DFLOP: A Data-driven Framework for Multimodal LLM Training Pipeline Optimization

Multimodal Large Language Models (MLLMs) have achieved remarkable advances by integrating text, image, and audio understanding within a unified architecture. However, existing distributed training frameworks remain fundamentally data-blind: they parallelize computation without accounting for variations in input data characteristics. This data unawareness leads to severe computation skew across stages and microbatches, where heterogeneous multimodal inputs incur different processing costs. Consequently, GPU resources are unevenly utilized, synchronization delays accumulate, and overall training efficiency degrades. To address this limitation, we present DFLOP, a data-driven framework for multimodal LLM training pipeline optimization. DFLOP continuously profiles runtime behavior to capture data-induced computation variance and employs predictive scheduling to balance workloads across stages and microbatches. By coupling data characteristics with execution planning, DFLOP substantially improves GPU utilization and throughput. Extensive experiments on large-scale multimodal benchmarks show that DFLOP achieves up to 3.6x faster training compared to state-of-the-art distributed training frameworks.

cs.DC↗

arXiv2026

Tight-Binding Device Modeling of 2-D Topological Insulator Field-Effect Transistors With Gate-Induced Phase Transition

Topological insulator field-effect transistors (TIFETs) built on 2-D quantum spin Hall insulators are being considered as advanced logic transistors due to their potentially superior performance originating from the dissipationless edge transport. This paper presents a device modeling based on the tight-binding model and the nonequilibrium Green's function formalism to simulate the current-voltage characteristics of the TIFETs. We then use the device simulator to demonstrate the effect of channel length on device performance. The device modeling will not only enable a direct estimation of TIFET performance but also shed light on the nontraditional switching operation via the topological phase transition.

cond-mat.mes-hall↗

arXiv2026

High-resolution neutron diffraction determination of noncollinear antiferromagnetic order in the honeycomb magnetoelectric Fe$_{4}$Nb$_{2}$O$_{9}$

Magnetoelectric systems offer potential for device applications exploiting coupled states between electric and magnetic properties. Among magnetoelectric materials, \FNO has attracted special attention because of its pronounced dielectric signal at high magnetic transition temperatures. However, the magnetic ground state, which is essential information for understanding its unusual magnetoelectricity, remains unclarified. Here, we report a noncollinear magnetic ground state of Fe$_{4}$Nb$_{2}$O$_{9}$. To examine the magnetoelectric effect associated with sequential magnetic and structural transitions upon cooling, we conducted combined x-ray diffraction, magnetic susceptibility, magnetization, dielectric constant, and magnetodielectric experiments. Powder neutron diffraction experiments revealed a series of magnetic Bragg peaks and clear splitting of peaks via structural transition. Magnetic Rietveld refinements, combined with group theory analysis, determined a noncollinear antiferromagnetic structure including a significant $c$-axis moment component at 1.5 K. This study provides insights into the understanding of its magnetoelectric properties.

cond-mat.mtrl-sci↗

arXiv2025

Unusual ferromagnetic band evolution and high Curie temperature in monolayer 1T-CrTe2 on bilayer graphene

2D van der Waals ferromagnets hold immense promise for spintronic applications due to their controllability and versatility. Despite their significance, the realization and in-depth characterization of ferromagnetic materials in atomically thin single layers, close to the true 2D limit, has been scarce. Here, a successful synthesis of monolayer (ML) 1T-CrTe2 is reported on a bilayer graphene (BLG) substrate via molecular beam epitaxy. Using angle-resolved photoemission spectroscopy and magneto-optical Kerr effect measurements, that the ferromagnetic transition is observed at the Curie temperature (TC) of 150 K in ML 1T-CrTe2 on BLG, accompanied by unconventional temperature-dependent band evolutions. The spectroscopic analysis and first-principle calculations reveal that the ferromagnetism may arise from Goodenough-Kanamori super-exchange and double-exchange interactions, enhanced by the lattice distortion and the electron doping from the BLG substrate. These findings provide pivotal insight into the fundamental understanding of mechanisms governing 2D ferromagnetism and offer a pathway for engineering higher TC in 2D materials for future spintronic devices.

cond-mat.mtrl-sci↗

arXiv2025

A New Lens on Homelessness: Daily Tent Monitoring with 311 Calls and Street Images

Homelessness in the United States has surged to levels unseen since the Great Depression. However, existing methods for monitoring it, such as point-in-time (PIT) counts, have limitations in terms of frequency, consistency, and spatial detail. This study proposes a new approach using publicly available, crowdsourced data, specifically 311 Service Calls and street-level imagery, to track and forecast homeless tent trends in San Francisco. Our predictive model captures fine-grained daily and neighborhood-level variations, uncovering patterns that traditional counts often overlook, such as rapid fluctuations during the COVID-19 pandemic and spatial shifts in tent locations over time. By providing more timely, localized, and cost-effective information, this approach serves as a valuable tool for guiding policy responses and evaluating interventions aimed at reducing unsheltered homelessness.

cs.LG↗

arXiv2025

Dataset Distillation of 3D Point Clouds via Distribution Matching

Large-scale datasets are usually required to train deep neural networks, but it increases the computational complexity hindering the practical applications. Recently, dataset distillation for images and texts has been attracting a lot of attention, that reduces the original dataset to a synthetic dataset to alleviate the computational burden of training while preserving essential task-relevant information. However, the dataset distillation for 3D point clouds remains largely unexplored, as the point clouds exhibit fundamentally different characteristics from that of images, making the dataset distillation more challenging. In this paper, we propose a distribution matching-based distillation framework for 3D point clouds that jointly optimizes the geometric structures as well as the orientations of the synthetic 3D objects. To address the semantic misalignment caused by unordered indexing of points, we introduce a Semantically Aligned Distribution Matching loss computed on the sorted features in each channel. Moreover, to address the rotation variation, we jointly learn the optimal rotation angles while updating the synthetic dataset to better align with the original feature distribution. Extensive experiments on widely used benchmark datasets demonstrate that the proposed method consistently outperforms existing dataset distillation methods, achieving superior accuracy and strong cross-architecture generalization.

cs.CV↗

arXiv2024

Unsupervised Outlier Detection using Random Subspace and Subsampling Ensembles of Dirichlet Process Mixtures

Probabilistic mixture models are recognized as effective tools for unsupervised outlier detection owing to their interpretability and global characteristics. Among these, Dirichlet process mixture models stand out as a strong alternative to conventional finite mixture models for both clustering and outlier detection tasks. Unlike finite mixture models, Dirichlet process mixtures are infinite mixture models that automatically determine the number of mixture components based on the data. Despite their advantages, the adoption of Dirichlet process mixture models for unsupervised outlier detection has been limited by challenges related to computational inefficiency and sensitivity to outliers in the construction of outlier detectors. Additionally, Dirichlet process Gaussian mixtures struggle to effectively model non-Gaussian data with discrete or binary features. To address these challenges, we propose a novel outlier detection method that utilizes ensembles of Dirichlet process Gaussian mixtures. This unsupervised algorithm employs random subspace and subsampling ensembles to ensure efficient computation and improve the robustness of the outlier detector. The ensemble approach further improves the suitability of the proposed method for detecting outliers in non-Gaussian data. Furthermore, our method uses variational inference for Dirichlet process mixtures, which ensures both efficient and rapid computation. Empirical analyses using benchmark datasets demonstrate that our method outperforms existing approaches in unsupervised outlier detection.

cs.LG↗

arXiv2023

Graph Theorem for Chiral Exact Flat Bands at Charge Neutrality

Chiral exact flat bands (FBs) at charge neutrality have attracted much recent interest, presenting an intriguing condensed-matter system to realize exact many-body phenomena, as specifically shown in "magic angle" twisted bilayer graphene for superconductivity and triangulene-based superatomic graphene for excitonic condensation. Yet, no generic physical model to realize such FBs has been developed. Here we present a new mathematical theorem, called bipartite double cover (BDC) theorem, and prove that the BDC of line-graph (LG) lattices hosts at least two chiral exact FBs of opposite chirality, i.e., yin-yang FBs, centered-around/at charge neutrality (E = 0) akin to the "chiral limit" of twisted bilayer graphene. We illustrate this theorem by mapping it exactly onto tight-binding lattice models of the BDC of LGs of hexagonal lattice for strong topological and of triangular lattice for fragile topological FBs, respectively. Moreover, we use orbital design principle to realize such exotic yin-yang FBs in non-BDC lattices to instigate their real material discovery. This work not only enables the search for exact chiral FBs at zero energy beyond moiré heterostructures, but also opens the door to discovering quantum semiconductor features with FB-enabled strongly correlated carriers.

cond-mat.mtrl-sci↗

arXiv2022

Realization of ideal flat band by rotated d-orbitals in Kagome metals

Recently there has been intense interest in Kagome metals, which are expected to host flat bands (FBs). However, the observed FBs are non-ideal as they are not flat over the whole 2D Brillouin zone and overlap strongly with other bands. Most critically, the theoretical conditions for the existence of ideal FB in Kagome metals, beyond the simplest Kagome lattice model, are unknown. Here, based on tight-binding model analyses of the interplay between d-orbital symmetry and underlying Kagome lattice symmetry, we establish such conditions. We show that for a pure transition-metal (TM) Kagome lattice, only dz2 orbital gives rise to a FB; while dxy, dx2-y2, dxz, and dyz orbitals do not. The condition for the latter four orbitals to produce a FB is to rotate them so that they will conform with the underlying Kagome lattice symmetry. Interestingly, the lattice having rotated dxy (dxz) and dx2-y2 (dyz) orbitals leads to a FB of opposite chirality sitting above and below the Dirac bands, respectively. For intercalated TM Kagome lattices, the Kagome-hexagonal intercalation exhibits always a FB, while the case for the Kagome-trigonal intercalation is conditional. Another condition to isolate the FB arising from individual d-orbitals is to increase the crystal field splitting, such as in Kagome metals of layered metalorganic frameworks.

cond-mat.mtrl-sci↗

arXiv2022

Nanomechanical Characterization of an Antiferromagnetic Topological Insulator

The antiferromagnetic topological insulator MnBi2Te4 (MBT) exhibits an ideal platform to study exotic topological phenomena and magnetic properties. The transport signatures of magnetic phase transitions in the MBT family materials have been well-studied. However, their mechanical properties and magneto-mechanical coupling have not been well-explored. We use nanoelectromechanical systems to study the intrinsic magnetism in MBT thin flakes via their magnetostrictive coupling. We investigate mechanical resonance signatures of magnetic phase transitions from antiferromagnetic (AFM) to canted antiferromagnetic (cAFM) to ferromagnetic (FM) phases versus magnetic field at different temperatures. The spin-flop transitions in MBT are revealed by frequency shifts of mechanical resonance. With temperatures going above TN, the transitions disappear in the resonance frequency map, consistent with transport measurements. We use a magnetostrictive model to correlate the frequency shifts with the spin-canting states. Our work demonstrates a technique to study magnetic phase transitions, magnetization and magnetoelastic properties of the magnetic topological insulator.

cond-mat.mes-hall↗

arXiv2022

Stable topological phase transitions without symmetry indications in NaZnSb$_{1-x}$Bi$_x$

We study topological phase transitions in tetragonal NaZnSb$_{1-x}$Bi$_x$, driven by the chemical composition $x$. Notably, we examine mirror Chern numbers that change without symmetry indicators. First-principles calculations are performed to show that NaZnSb$_{1-x}$Bi$_x$ experiences consecutive topological phase transitions, diagnosed by the strong $\mathbb Z_{2}$ topological index $ν_{0}$ and two mirror Chern numbers $μ_{x}$ and $μ_{xy}$. As the chemical composition $x$ increases, the topological invariants ($μ_{x}μ_{xy}ν_{0}$) change from $(000)$, $(020)$, $(220)$, to $(111)$ at $x$ = 0.15, 0.20, and 0.53, respectively. A simplified low-energy effective model is developed to examine the mirror Chern number changes, highlighting the central role of spectator Dirac fermions in avoiding symmetry indicators. Our findings suggest that NaZnSb$_{1-x}$Bi$_{x}$ can be an exciting testbed for the exploration of the interplay between the topology and symmetry.

cond-mat.mtrl-sci↗

arXiv2019

Stiefel-Whitney classes and topological phases in band theory

In this article, we review the recent progress in the study of topological phases in systems with space-time inversion symmetry $I_{\text{ST}}$. $I_{\text{ST}}$ is an anti-unitary symmetry which is local in momentum space and satisfies $I_{\text{ST}}^2=1$ such as $PT$ or $C_{2}T$ symmetry where $P$, $T$, $C_2$ indicate inversion, time-reversal, and two-fold rotation symmetries, respectively. Under $I_{\text{ST}}$, the Hamiltonian and the Bloch wave function can be constrained to be real-valued, which makes the Berry curvature and the Chern number to vanish. In this class of systems, gapped band structures of real wave functions can be topologically distinguished by Stiefel-Whitney numbers instead. The first and second Stiefel-Whitney numbers $w_1$ and $w_2$, respectively, are the corresponding invariants in 1D and 2D, which are equivalent to the quantized Berry phase and the $Z_2$ monopole charge, respectively. We first describe the topological phases characterized by the first Stiefel-Whitney number, including 1D topological insulators with quantized charge polarization, 2D Dirac semimetals, and 3D nodal line semimetals. Next we review how the second Stiefel-Whitney class characterizes the 3D nodal line semimetals carrying a $Z_{2}$ monopole charge. In particular, we explain how the second Stiefel-Whitney number $w_2$, the $Z_{2}$ monopole charge, and the linking number between nodal lines are related. Finally, we review the properties of 2D and 3D topological insulators characterized by the nontrivial second Stiefel Whitney class.

cond-mat.mes-hall↗

arXiv2018

Band Topology and Linking Structure of Nodal Line Semimetals with Z2 Monopole Charges

We study the band topology and the associated linking structure of topological semimetals with nodal lines carrying $Z_{2}$ monopole charges, which can be realized in three-dimensional systems invariant under the combination of inversion $P$ and time reversal $T$ when spin-orbit coupling is negligible. In contrast to the well-known $PT$-symmetric nodal lines protected only by $π$ Berry phase in which a single nodal line can exist, the nodal lines with $Z_{2}$ monopole charges should always exist in pairs. We show that a pair of nodal lines with $Z_{2}$ monopole charges is created by a {\it double band inversion} (DBI) process, and that the resulting nodal lines are always {\it linked by another nodal line} formed between the two topmost occupied bands. It is shown that both the linking structure and the $Z_{2}$ monopole charge are the manifestation of the nontrivial band topology characterized by the {\it second Stiefel-Whitney class}, which can be read off from the Wilson loop spectrum. We show that the second Stiefel-Whitney class can serve as a well-defined topological invariant of a $PT$-invariant two-dimensional (2D) insulator in the absence of Berry phase. Based on this, we propose that pair creation and annihilation of nodal lines with $Z_{2}$ monopole charges can mediate a topological phase transition between a normal insulator and a three-dimensional weak Stiefel-Whitney insulator (3D weak SWI). Moreover, using first-principles calculations, we predict ABC-stacked graphdiyne as a nodal line semimetal (NLSM) with $Z_{2}$ monopole charges having the linking structure. Finally, we develop a formula for computing the second Stiefel-Whitney class based on parity eigenvalues at inversion invariant momenta, which is used to prove the quantized bulk magnetoelectric response of NLSMs with $Z_2$ monopole charges under a $T$-breaking perturbation.

cond-mat.mes-hall↗