arXiv subjects

Sangyun Lee

Publications and source records attributed to Sangyun Lee.

At least 37 records · Page 2Linked to original sources

arXiv2024

Out-of-plane magnetic phase diagram of Kitaev quantum spin liquid candidate Na2Co2TeO6

We have investigated the magnetic properties and mapped out the phase diagram of the honeycomb magnet Na2Co2TeO6 with Co 3d7 in out-of-plane magnetic fields. This material has previously been proposed to show nearest-neighbor Kitaev interactions between Co spins and maybe even Kitaev quantum spin liquid behavior in high fields. At low magnetic fields, we observe a thermal phase transition at TN = 27 K, transitioning from a paramagnetic state to a canonical ferrimagnetic state. Under the application of magnetic fields, a spin flop-like phase transition occurred before saturation of J = 1/2 between 10 K and TN. Below 10 K, a peak-dip-peak structure emerges between 10 and 17 T in the magnetic susceptibility (dM/dH) before the magnetic saturation, reminiscent of magnetic plateau behavior. The measurement of the magnetocaloric effect also shows dip-peak-dip behavior in this field range. Our data can be explained by an XXZ model with a single ion anisotropy and possibly small Kitaev and Γ exchange interactions. We also unambiguously determined the magnetization saturation field that helps constrain the energy scale of the exchange interactions

cond-mat.str-el↗

arXiv2024

Normal Fermi Surface in the Nodal Superconductor CeCoIn$_5$ Revealed via Thermal Conductivity

The thermal conductivity of heavy-fermion superconductor CeCoIn$_5$ was measured with a magnetic field rotating in the tetragonal a-b plane, with the heat current in the anti-nodal direction, $J$ || [100]. We observe a sharp resonance in thermal conductivity for the magnetic field at an angle $θ$ $\sim$ 12$^{\circ}$, measured from the heat current direction [100]. This resonance corresponds to the reported resonance at an angle $θ'$ $\sim$ 33$^{\circ}$ from the direction of the heat current applied along the nodal direction, $J$ || [110]. Both resonances, therefore, occur when the magnetic field is applied in the same crystallographic orientation in the two experiments, regardless of the direction of the heat current, proving conclusively that these resonances are due to the structure of the Fermi surface of CeCoIn$_5$. We argue that the uncondensed Landau quasiparticles, emerging with field, are responsible for the observed resonance. We support our experimental results with density-functional-theory model calculations of the density of states in a rotating magnetic field. Our calculations, using a model Fermi surface of CeCoIn$_5$, reveal several sharp peaks as a function of the field direction. Our study demonstrates that the thermal-conductivity measurement in rotating magnetic field can probe the normal parts of the Fermi surface deep inside the superconducting state.

cond-mat.supr-con↗

arXiv2024

Can LLMs Produce Faithful Explanations For Fact-checking? Towards Faithful Explainable Fact-Checking via Multi-Agent Debate

Fact-checking research has extensively explored verification but less so the generation of natural-language explanations, crucial for user trust. While Large Language Models (LLMs) excel in text generation, their capability for producing faithful explanations in fact-checking remains underexamined. Our study investigates LLMs' ability to generate such explanations, finding that zero-shot prompts often result in unfaithfulness. To address these challenges, we propose the Multi-Agent Debate Refinement (MADR) framework, leveraging multiple LLMs as agents with diverse roles in an iterative refining process aimed at enhancing faithfulness in generated explanations. MADR ensures that the final explanation undergoes rigorous validation, significantly reducing the likelihood of unfaithful elements and aligning closely with the provided evidence. Experimental results demonstrate that MADR significantly improves the faithfulness of LLM-generated explanations to the evidence, advancing the credibility and trustworthiness of these explanations.

cs.CL↗

arXiv2024

Magnetic field-temperature phase diagram of spin-1/2 triangular lattice antiferromagnet KYbSe$_2$

A quantum spin liquid (QSL) is a state of matter characterized by fractionalized quasiparticle excitations, quantum entanglement, and a lack of long-range magnetic order. However, QSLs have evaded definitive experimental observation. Several Yb$^{3+}$-based triangular lattice antiferromagnets with effective $S$ = $\frac{1}{2}$ have been suggested to stabilize the QSL state as the ground state. Here, we build a comprehensive magnetic temperature phase diagram of a high-quality single crystalline KYbSe$_2$ via heat capacity and magnetocaloric effect down to 30 mK with magnetic field applied along the $a$-axis. At zero magnetic field, we observe the magnetic long-range order at $T_N$ = 0.29 K entering 120 degrees ordered state in heat capacity, consistent with neutron scattering studies. Analysis of the low-temperature ($T$) specific heat ($C$) at zero magnetic field indicates linear $T$-dependence of $C/T$ and a broad hump of $C/T$ in the proximate QSL region above $T_N$. By applying magnetic field, we observe the up-up-down phase with 1/3 magnetization plateau and oblique phases, in addition to two new phases. These observations strongly indicate that while KYbSe$_2$ closely exhibits characteristics resembling an ideal triangular lattice, deviations may exist, such as the effect of the next-nearest-neighbor exchange interaction, calling for careful consideration for spin Hamiltonian modeling. Further investigations into tuning parameters, such as chemical pressure, could potentially induce an intriguing QSL phase in the material.

cond-mat.str-el↗

arXiv2023

Estimating Entanglement Entropy via Variational Quantum Circuits with Classical Neural Networks

Entropy plays a crucial role in both physics and information science, encompassing classical and quantum domains. In this work, we present the Quantum Neural Entropy Estimator (QNEE), a novel approach that combines classical neural network (NN) with variational quantum circuits to estimate the von Neumann and Renyi entropies of a quantum state. QNEE provides accurate estimates of entropy while also yielding the eigenvalues and eigenstates of the input density matrix. Leveraging the capabilities of classical NN, QNEE can classify different phases of quantum systems that accompany the changes of entanglement entropy. Our numerical simulation demonstrates the effectiveness of QNEE by applying it to the 1D XXZ Heisenberg model. In particular, QNEE exhibits high sensitivity in estimating entanglement entropy near the phase transition point. We expect that QNEE will serve as a valuable tool for quantum entropy estimation and phase classification.

quant-ph↗

arXiv2023

Non-linear magnons and exchange Hamiltonians of delafossite proximate quantum spin liquids

Quantum spin liquids (QSL) are theoretical states of matter with long-range entanglement and exotic quasiparticles. However, they generally elude quantitative theory, rendering their underlying phases mysterious and hampering efforts to identify experimental QSL states. Here we study triangular lattice resonating valence bond QSL candidate materials KYbSe$_2$ and NaYbSe$_2$. We measure the magnon modes in their 1/3 plateau phase, where quantitative theory is tractable, using inelastic neutron scattering and fit them using nonlinear spin wave theory. We also fit the KYbSe$_2$ heat capacity using high temperature series expansion. Both KYbSe$_2$ fits yield the same magnetic Hamiltonian to within uncertainty, confirming previous estimates and showing the Heisenberg $J_2/J_1$ to be an accurate model for these materials. Most importantly, comparing KYbSe$_2$ and NaYbSe$_2$ shows that smaller $A$-site Na$^+$ ion has a larger $J_2/J_1$ ratio. However, hydrostatic pressure applied to KYbSe$_2$ increases the ordering temperature (a result consistent with density functional theory calculations), indicating that pressure decreases $J_2/J_1$. These results show how periodic table and hydrostatic pressure can tune the $A$YbSe$_2$ materials in a controlled way.

cond-mat.str-el↗

arXiv2023

Anisotropic field-induced changes in the superconducting order parameter of UTe2

UTe2 is a newly discovered unconventional superconductor, where electron Cooper pairs combine into a spin-triplet ground state. Here we report the specific heat C(H,T) of a high-quality single crystal of UTe2 with a single specific heat anomaly at the superconducting transition temperature T_c {\approx} 2 K and a small zero-field residual Sommerfeld coefficient γ_0 = C/T (T=0) = 10 mJ/mol-K^2. We applied magnetic field up to 12 T along the three principal crystallographic axes of UTe2 to probe the nature of the superconducting state. The evolution of the residual Sommerfeld coefficient as a function of magnetic field, γ_0 (H), is highly anisotropic and reveals distinct regions. In magnetic field up to 4 T applied along a, b, and c axes, we find γ_0{\approx}α_i {\square}H, with i=a,b,c, as expected for an unconventional superconductor with nodes (zeros) of the superconducting order parameter on the Fermi surface. A pronounced kink in γ_0(H), however, is observed at roughly 4 T for field applied along both a and b axes, whereas a smooth change from square-root to linear behaviour is observed at 4 T for H//c. These results strongly indicate that a zero-field ground state is stable up to 4 T and undergoes a field-induced evolution above 4 T. {α_c} > {α_a} > {α_b}, indicating that the nodes in the low-field state are predominantly located in the vicinity of the a-b plane. The modification of the order parameter is strongest when field is applied in the a-b plane, which causes nodes to move away from the direction of the applied field. Both d_(B_2u)+id_(B_1u) and d_(B_2u)+id_(A_u) two-component order parameters can account for our observations, with d_(B_2u)+id_(B_1u) a more likely candidate. In either scenario, our measurements indicate that B_2u is the primary superconducting order parameter in UTe2.

cond-mat.supr-con↗

arXiv2023

Sequential Data Generation with Groupwise Diffusion Process

We present the Groupwise Diffusion Model (GDM), which divides data into multiple groups and diffuses one group at one time interval in the forward diffusion process. GDM generates data sequentially from one group at one time interval, leading to several interesting properties. First, as an extension of diffusion models, GDM generalizes certain forms of autoregressive models and cascaded diffusion models. As a unified framework, GDM allows us to investigate design choices that have been overlooked in previous works, such as data-grouping strategy and order of generation. Furthermore, since one group of the initial noise affects only a certain group of the generated data, latent space now possesses group-wise interpretable meaning. We can further extend GDM to the frequency domain where the forward process sequentially diffuses each group of frequency components. Dividing the frequency bands of the data as groups allows the latent variables to become a hierarchical representation where individual groups encode data at different levels of abstraction. We demonstrate several applications of such representation including disentanglement of semantic attributes, image editing, and generating variations.

cs.CV↗

arXiv2023

Signatures of a Majorana-Fermi surface in the Kitaev magnet Ag$_3$LiIr$_2$O$_6$

Detecting Majorana fermions in experimental realizations of the Kitaev honeycomb model is often complicated by non-trivial interactions inherent to potential spin liquid candidates. In this work, we identify several distinct thermodynamic signatures of massive, itinerant Majorana fermions within the well-established analytical paradigm of Landau-Fermi liquid theory. We find a qualitative and quantitative agreement between the salient features of our Landau-Majorana liquid theory and the Kitaev spin liquid candidate Ag$_3$LiIr$_2$O$_6$. Our study presents strong evidence for a Fermi liquid-like ground state in the fundamental excitations of a honeycomb iridate, and opens new experimental avenues to detect itinerant Majorana fermions in condensed matter systems.

cond-mat.str-el↗

arXiv2023

Electronic and magnetic phase diagrams of Kitaev quantum spin liquid candidate Na$_2$Co$_2$TeO$_6$

The 3$d^7$ Co$^{2+}$-based insulating magnet \NCTO{} has recently been reported to have strong Kitaev interactions on a honeycomb lattice, and is thus being considered as a Kitaev quantum spin liquid candidate. However, due to the existence of other types of interactions, a spontaneous long-range magnetic order occurs. This order is suppressed by applied magnetic fields leading to a succession of phases and ultimately saturation of the magnetic moments. The precise phase diagram, the nature of the phases, and the possibility that one of the field-induced phases is a Kitaev quantum spin liquid phase are still a matter of debate. Here we measured an extensive set of physical properties to build the complete temperature-field phase diagrams to magnetic saturation at 10 T for magnetic fields along the $a$- and $a^*$-axes, and a partial phase diagram up to 60 T along $c$. We probe the phases using magnetization, specific heat, magnetocaloric effect, magnetostriction, dielectric constant, and electric polarization, which is a symmetry-sensitive probe. With these measurements we identify all the previously incomplete phase boundaries and find new high-field phase boundaries. We find strong magnetoelectric coupling in the dielectric constant and moderate magnetostrictive coupling at several phase boundaries. Furthermore, we detect the symmetry of the magnetic order using electrical polarization measurements under magnetic fields. Based on our analysis, the absence of electric polarization under zero or finite magnetic field in any of the phases or after...

cond-mat.str-el↗

arXiv2023

Minimizing Trajectory Curvature of ODE-based Generative Models

Recent ODE/SDE-based generative models, such as diffusion models, rectified flows, and flow matching, define a generative process as a time reversal of a fixed forward process. Even though these models show impressive performance on large-scale datasets, numerical simulation requires multiple evaluations of a neural network, leading to a slow sampling speed. We attribute the reason to the high curvature of the learned generative trajectories, as it is directly related to the truncation error of a numerical solver. Based on the relationship between the forward process and the curvature, here we present an efficient method of training the forward process to minimize the curvature of generative trajectories without any ODE/SDE simulation. Experiments show that our method achieves a lower curvature than previous models and, therefore, decreased sampling costs while maintaining competitive performance. Code is available at https://github.com/sangyun884/fast-ode.

cs.LG↗

arXiv2023

Witnessing quantum criticality and entanglement in the triangular antiferromagnet KYbSe$_2$

The Heisenberg triangular lattice quantum spin liquid and the phase transitions to nearby magnetic orders have received much theoretical attention, but clear experimental manifestations of these states are rare. This work investigates a new spin-half Yb$^{3+}$ delafossite material, KYbSe$_2$, whose inelastic neutron scattering spectra reveal a diffuse continuum with a sharp lower bound. Applying entanglement witnesses to the data reveals significant multipartite entanglement spread between its neighbors, and analysis of its magnetic exchange couplings shows close proximity to the triangular lattice Heisenberg quantum spin liquid. Key features of the data are reproduced by Schwinger-boson theory and tensor network calculations with a significant second-neighbor coupling $J_2$. The strength of the dynamical structure factor at the $K$ point shows a scaling collapse in $\hbarω/k_\mathrm{B}T$ down to 0.3 K, indicating a second-order quantum phase transition. Comparing this to previous theoretical work suggests that the proximate phase at larger $J_2$ is a gapped $\mathbb{Z}_2$ spin liquid, resolving a long-debated issue. We thus show that KYbSe$_2$ is close to a spin liquid phase, which in turn sheds light on the theoretical phase diagram itself.

cond-mat.str-el↗

arXiv2023

Multidimensional entropic bound: Estimator of entropy production for general Langevin dynamics with an arbitrary time-dependent protocol

Entropy production (EP) is a key quantity in thermodynamics, and yet measuring EP has remained a challenging task. Here we introduce an EP estimator, called multidimensional entropic bound (MEB), utilizing an ensemble of trajectories without knowing the details of a given system. MEB is a unified method in the sense that it is applicable to both overdamped and underdamped Langevin dynamics, irrespective of the time dependence of the driving protocol. In addition, MEB is computationally efficient because optimization is unnecessary. We apply our developed estimator to three physical systems driven by time-dependent protocols pertaining to experiments using optical tweezers: a dragged Brownian particle, a pulling process of a harmonic chain, and an unfolding process of an RNA hairpin. Numerical simulations confirm the validity and efficiency of our method.

cond-mat.stat-mech↗

arXiv2022

Progressive Deblurring of Diffusion Models for Coarse-to-Fine Image Synthesis

Recently, diffusion models have shown remarkable results in image synthesis by gradually removing noise and amplifying signals. Although the simple generative process surprisingly works well, is this the best way to generate image data? For instance, despite the fact that human perception is more sensitive to the low frequencies of an image, diffusion models themselves do not consider any relative importance of each frequency component. Therefore, to incorporate the inductive bias for image data, we propose a novel generative process that synthesizes images in a coarse-to-fine manner. First, we generalize the standard diffusion models by enabling diffusion in a rotated coordinate system with different velocities for each component of the vector. We further propose a blur diffusion as a special case, where each frequency component of an image is diffused at different speeds. Specifically, the proposed blur diffusion consists of a forward process that blurs an image and adds noise gradually, after which a corresponding reverse process deblurs an image and removes noise progressively. Experiments show that the proposed model outperforms the previous method in FID on LSUN bedroom and church datasets. Code is available at https://github.com/sangyun884/blur-diffusion.

cs.CV↗

arXiv2022

Learning Multiple Probabilistic Degradation Generators for Unsupervised Real World Image Super Resolution

Unsupervised real world super resolution (USR) aims to restore high-resolution (HR) images given low-resolution (LR) inputs, and its difficulty stems from the absence of paired dataset. One of the most common approaches is synthesizing noisy LR images using GANs (i.e., degradation generators) and utilizing a synthetic dataset to train the model in a supervised manner. Although the goal of training the degradation generator is to approximate the distribution of LR images given a HR image, previous works have heavily relied on the unrealistic assumption that the conditional distribution is a delta function and learned the deterministic mapping from the HR image to a LR image. In this paper, we show that we can improve the performance of USR models by relaxing the assumption and propose to train the probabilistic degradation generator. Our probabilistic degradation generator can be viewed as a deep hierarchical latent variable model and is more suitable for modeling the complex conditional distribution. We also reveal the notable connection with the noise injection of StyleGAN. Furthermore, we train multiple degradation generators to improve the mode coverage and apply collaborative learning for ease of training. We outperform several baselines on benchmark datasets in terms of PSNR and SSIM and demonstrate the robustness of our method on unseen data distribution. Code is available at https://github.com/sangyun884/MSSR.

eess.IV↗

arXiv2022

High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions

Image-based virtual try-on aims to synthesize an image of a person wearing a given clothing item. To solve the task, the existing methods warp the clothing item to fit the person's body and generate the segmentation map of the person wearing the item before fusing the item with the person. However, when the warping and the segmentation generation stages operate individually without information exchange, the misalignment between the warped clothes and the segmentation map occurs, which leads to the artifacts in the final image. The information disconnection also causes excessive warping near the clothing regions occluded by the body parts, so-called pixel-squeezing artifacts. To settle the issues, we propose a novel try-on condition generator as a unified module of the two stages (i.e., warping and segmentation generation stages). A newly proposed feature fusion block in the condition generator implements the information exchange, and the condition generator does not create any misalignment or pixel-squeezing artifacts. We also introduce discriminator rejection that filters out the incorrect segmentation map predictions and assures the performance of virtual try-on frameworks. Experiments on a high-resolution dataset demonstrate that our model successfully handles the misalignment and occlusion, and significantly outperforms the baselines. Code is available at https://github.com/sangyun884/HR-VITON.

cs.CV↗

arXiv2022

Speed limit for a highly irreversible process and tight finite-time Landauer's bound

Landauer's bound is the minimum thermodynamic cost for erasing one bit of information. As this bound is achievable only for quasistatic processes, finite-time operation incurs additional energetic costs. We find a tight finite-time Landauer's bound by establishing a general form of the classical speed limit. This tight bound well captures the divergent behavior associated with the additional cost of a highly irreversible process, which scales differently from a nearly irreversible process. We also find an optimal dynamics which saturates the equality of the bound. We demonstrate the validity of this bound via discrete one-bit and coarse-grained bit systems. Our work implies that more heat dissipation than expected occurs during high-speed irreversible computation.

cond-mat.stat-mech↗

arXiv2021

Estimating entropy production in a stochastic system with odd-parity variables

Entropy production (EP) is a central measure in nonequilibrium thermodynamics, as it can quantify the irreversibility of a process as well as its energy dissipation in special cases. Using the time-reversal asymmetry in a system's path probability distribution, many methods have been developed to estimate EP from only trajectory data. However, estimating the EP of a system with odd-parity variables, which prevails in nonequilibrium systems, has not been covered. In this study, we develop a machine learning method for estimating the EP in a stochastic system with odd-parity variables through multiple neural networks. We demonstrate our method with two systems, an underdamped bead-spring model and a one-particle odd-parity Markov jump process.

cond-mat.stat-mech↗