arXiv subjects

Yilin Wang

Publications and source records attributed to Yilin Wang.

At least 127 records · Page 7Linked to original sources

arXiv2024

Correlated Electronic Structure and Incipient Flat Bands of the Kagome Superconductor CsCr3Sb5

Kagome materials exhibit many novel phenomena emerging from the interplay between lattice geometry, electronic structure, and topology. A prime example is the vanadium-based kagome materials AV3Sb5 (A = K, Rb, and Cs) with superconductivity and unconventional charge-density wave (CDW). More interestingly, the substitution of vanadium by chromium further introduces magnetism and enhances the correlation effect in CsCr3Sb5 which likewise exhibits superconductivity under pressure and competing density-wave state. Here we systematically investigate the electronic structure of CsCr3Sb5 using high-resolution angle-resolved photoemission spectroscopy (APRES) and ab-initio calculations. Overall, the measured electronic structure agrees with the theoretical calculation. Remarkably, Cr 3d orbitals exhibit incoherent electronic states and contribute to incipient flat bands close to the Fermi level. The electronic structure shows a minor change across the magnetic transition at 55 K, suggesting a weak interplay between the local magnetic moment and itinerant electrons. Furthermore, we reveal a drastic enhancement of the electron scattering rate across the magnetic transition, which is relevant to the semiconducting-like transport property of the system at high temperatures. Our results suggest that CsCr3Sb5 is a strongly correlated Hund's metal with incipient flat bands near the Fermi level, which provides an electronic basis for understanding its novel properties in comparison to the non-magnetic and weakly correlated AV3Sb5.

cond-mat.str-el↗

arXiv2024

Encoding innumerable charge density waves of FeGe into polymorphs of LiFe6Ge6

Kagome metals exhibit rich quantum states by the intertwining of lattice, charge, orbital and spin degrees of freedom. Recently, a novel charge density wave (CDW) ground state was discovered in kagome magnet FeGe and was revealed to be driven by lowering magnetic energy via large Ge1-dimerization. Here, based on DFT calculations, we show that such mechanism will yield infinitely many metastable CDWs in FeGe due to different ways to arrange the Ge1-dimerization in enlarged superstructures. Intriguingly, utilizing these metastable CDWs, innumerable polymorphs of kagome magnet LiFe6Ge6 can be stabilized by filling Li atoms in the voids right above/below the dimerized Ge1-sites in the CDW superstructures. Such polymorphs are very stable due to the presence of magnetic-energy-saving mechanism, in sharp contrast to the non-magnetic "166" kagome compounds. In this way, a one-to-one mapping of the metastable CDWs of FeGe to stable polymorphs of LiFe6Ge6 is established. On one hand, the fingerprints of these metastable CDWs, i.e., the induced in-plane atomic distortions and band gaps, are encoded into the corresponding stable polymorphs of LiFe6Ge6, such that further study of their properties becomes possible. On the other hand, such innumerable polymorphs of LiFe6Ge6 offer great degrees of freedom to explore the rich physics of magnetic kagome metals. We thus reveal a novel connection between the unusually abundant CDWs and structural polymorphism in magnetic kagome materials, and establish a new route to obtain structural polymorphism on top of CDW states.

cond-mat.mtrl-sci↗

arXiv2024

Onsager-Machlup functional for $\text{SLE}_κ$ loop measures

We relate two ways to renormalize the Brownian loop measure on the Riemann sphere. One by considering the Brownian loop measure on the sphere minus a small disk, known as the normalized Brownian loop measure; the other by taking the measure on simple loops induced by the outer boundary of the Brownian loops, known as Werner's measure. This result allows us to interpret the Loewner energy as an Onsager--Machlup functional for SLE$_κ$ loop measure for any fixed $κ\in (0, 4]$, and more generally, for any Malliavin--Kontsevich--Suhov loop measure of the same central charge.

math.PR↗

arXiv2024

Bayesian Diffusion Models for 3D Shape Reconstruction

We present Bayesian Diffusion Models (BDM), a prediction algorithm that performs effective Bayesian inference by tightly coupling the top-down (prior) information with the bottom-up (data-driven) procedure via joint diffusion processes. We show the effectiveness of BDM on the 3D shape reconstruction task. Compared to prototypical deep learning data-driven approaches trained on paired (supervised) data-labels (e.g. image-point clouds) datasets, our BDM brings in rich prior information from standalone labels (e.g. point clouds) to improve the bottom-up 3D reconstruction. As opposed to the standard Bayesian frameworks where explicit prior and likelihood are required for the inference, BDM performs seamless information fusion via coupled diffusion processes with learned gradient computation networks. The specialty of our BDM lies in its capability to engage the active and effective information exchange and fusion of the top-down and bottom-up processes where each itself is a diffusion process. We demonstrate state-of-the-art results on both synthetic and real-world benchmarks for 3D shape reconstruction.

cs.CV↗

arXiv2024

Learning Mutually Informed Representations for Characters and Subwords

Most pretrained language models rely on subword tokenization, which processes text as a sequence of subword tokens. However, different granularities of text, such as characters, subwords, and words, can contain different kinds of information. Previous studies have shown that incorporating multiple input granularities improves model generalization, yet very few of them outputs useful representations for each granularity. In this paper, we introduce the entanglement model, aiming to combine character and subword language models. Inspired by vision-language models, our model treats characters and subwords as separate modalities, and it generates mutually informed representations for both granularities as output. We evaluate our model on text classification, named entity recognition, POS-tagging, and character-level sequence labeling (intraword code-switching). Notably, the entanglement model outperforms its backbone language models, particularly in the presence of noisy texts and low-resource languages. Furthermore, the entanglement model even outperforms larger pre-trained models on all English sequence labeling tasks and classification tasks. We make our code publically available.

cs.CL↗

arXiv2024

UniHuman: A Unified Model for Editing Human Images in the Wild

Human image editing includes tasks like changing a person's pose, their clothing, or editing the image according to a text prompt. However, prior work often tackles these tasks separately, overlooking the benefit of mutual reinforcement from learning them jointly. In this paper, we propose UniHuman, a unified model that addresses multiple facets of human image editing in real-world settings. To enhance the model's generation quality and generalization capacity, we leverage guidance from human visual encoders and introduce a lightweight pose-warping module that can exploit different pose representations, accommodating unseen textures and patterns. Furthermore, to bridge the disparity between existing human editing benchmarks with real-world data, we curated 400K high-quality human image-text pairs for training and collected 2K human images for out-of-domain testing, both encompassing diverse clothing styles, backgrounds, and age groups. Experiments on both in-domain and out-of-domain test sets demonstrate that UniHuman outperforms task-specific models by a significant margin. In user studies, UniHuman is preferred by the users in an average of 77% of cases. Our project is available at https://github.com/NannanLi999/UniHuman.

cs.CV↗

arXiv2024

KC-GenRe: A Knowledge-constrained Generative Re-ranking Method Based on Large Language Models for Knowledge Graph Completion

The goal of knowledge graph completion (KGC) is to predict missing facts among entities. Previous methods for KGC re-ranking are mostly built on non-generative language models to obtain the probability of each candidate. Recently, generative large language models (LLMs) have shown outstanding performance on several tasks such as information extraction and dialog systems. Leveraging them for KGC re-ranking is beneficial for leveraging the extensive pre-trained knowledge and powerful generative capabilities. However, it may encounter new problems when accomplishing the task, namely mismatch, misordering and omission. To this end, we introduce KC-GenRe, a knowledge-constrained generative re-ranking method based on LLMs for KGC. To overcome the mismatch issue, we formulate the KGC re-ranking task as a candidate identifier sorting generation problem implemented by generative LLMs. To tackle the misordering issue, we develop a knowledge-guided interactive training method that enhances the identification and ranking of candidates. To address the omission issue, we design a knowledge-augmented constrained inference method that enables contextual prompting and controlled generation, so as to obtain valid rankings. Experimental results show that KG-GenRe achieves state-of-the-art performance on four datasets, with gains of up to 6.7% and 7.7% in the MRR and Hits@1 metric compared to previous methods, and 9.0% and 11.1% compared to that without re-ranking. Extensive analysis demonstrates the effectiveness of components in KG-GenRe.

cs.AI↗

arXiv2024

Quasiconformal deformation of the chordal Loewner driving function and first variation of the Loewner energy

We derive the variational formula of the Loewner driving function of a simple chord under infinitesimal quasiconformal deformations with Beltrami coefficients supported away from the chord. As an application, we obtain the first variation of the Loewner energy of a Jordan curve, defined as the Dirichlet energy of the driving function of the curve. This result gives another explanation of the identity between the Loewner energy and the universal Liouville action introduced by Takhtajan and Teo, which has the same variational formula. We also deduce the variation of the total mass of SLE$_{8/3}$ loops touching the Jordan curve under quasiconformal deformations.

math.CV↗

arXiv2024

From the random geometry of conformally invariant systems to the Kähler geometry of universal Teichmüller space

The goal of this expository article is to explain how a fundamental functional on the space of Jordan curves arising from SLE - Loewner energy - is connected to a seemingly far apart subject: the Kähler geometry of universal Teichmüller space. We provide a background on Loewner energy, the universal Liouville action, and the intuition behind the proof of the identity between them.

math.PR↗

arXiv2024

The Loewner-Kufarev Energy and Foliations by Weil-Petersson Quasicircles

We study foliations by chord-arc Jordan curves of the twice punctured Riemann sphere $\mathbb C \smallsetminus \{0\}$ using the Loewner-Kufarev equation. We associate to such a foliation a function on the plane that describes the "local winding" along each leaf. Our main theorem is that this function has finite Dirichlet energy if and only if the Loewner driving measure $ρ$ has finite Loewner-Kufarev energy, defined by $$S(ρ) = \frac{1}{2}\iint_{S^1 \times \mathbb{R}} ν_t'(θ)^2 \, d θd t$$ whenever $ρ$ is of the form $ν_t(θ)^2 d θd t$, and set to $\infty$ otherwise. Moreover, if either of these two energies is finite they are equal up to a constant factor, and in this case, the foliation leaves are Weil-Petersson quasicircles. This duality between energies has several consequences. The first is that the Loewner-Kufarev energy is reversible, that is, invariant under inversion and time-reversal of the foliation. Furthermore, the Loewner energy of a Jordan curve can be expressed using the minimal Loewner-Kufarev energy of those measures that generate the curve as a leaf. This provides a new and quantitative characterization of Weil-Petersson quasicircles. Finally, we consider conformal distortion of the foliation and show that the Loewner-Kufarev energy satisfies an exact transformation law involving the Schwarzian derivative. The proof of our main theorem uses an isometry between the Dirichlet energy space on the unit disc and $L^2(2ρ)$ that we construct using Hadamard's variational formula expressed by means of the Loewner-Kufarev equation. Our results are related to $κ$-parameter duality and large deviations of Schramm-Loewner evolutions coupled with Gaussian random fields.

math.CV↗

arXiv2023

Quantum-Enhanced Forecasting: Leveraging Quantum Gramian Angular Field and CNNs for Stock Return Predictions

We propose a time series forecasting method named Quantum Gramian Angular Field (QGAF). This approach merges the advantages of quantum computing technology with deep learning, aiming to enhance the precision of time series classification and forecasting. We successfully transformed stock return time series data into two-dimensional images suitable for Convolutional Neural Network (CNN) training by designing specific quantum circuits. Distinct from the classical Gramian Angular Field (GAF) approach, QGAF's uniqueness lies in eliminating the need for data normalization and inverse cosine calculations, simplifying the transformation process from time series data to two-dimensional images. To validate the effectiveness of this method, we conducted experiments on datasets from three major stock markets: the China A-share market, the Hong Kong stock market, and the US stock market. Experimental results revealed that compared to the classical GAF method, the QGAF approach significantly improved time series prediction accuracy, reducing prediction errors by an average of 25% for Mean Absolute Error (MAE) and 48% for Mean Squared Error (MSE). This research confirms the potential and promising prospects of integrating quantum computing with deep learning techniques in financial time series forecasting.

cs.LG↗

arXiv2023

Circle homeomorphisms with square summable diamond shears

We introduce and the study the space of homeomorphisms of the circle (up to Möbius transformations) which are in $\ell^2$ with respect to modular coordinates called diamond shears along the edges of the Farey tessellation. Diamond shears are related combinatorially to shear coordinates, and are also closely related to the $\log Λ$-lengths of decorated Teichmüller space introduced by Penner. We obtain sharp results comparing this new class to the Weil-Petersson class and Hölder classes of circle homeomorphisms. We also express the Weil-Petersson metric tensor and symplectic form in terms of infinitesimal shears and diamond shears.

math.CV↗

arXiv2023

Dolfin: Diffusion Layout Transformers without Autoencoder

In this paper, we introduce a novel generative model, Diffusion Layout Transformers without Autoencoder (Dolfin), which significantly improves the modeling capability with reduced complexity compared to existing methods. Dolfin employs a Transformer-based diffusion process to model layout generation. In addition to an efficient bi-directional (non-causal joint) sequence representation, we further propose an autoregressive diffusion model (Dolfin-AR) that is especially adept at capturing rich semantic correlations for the neighboring objects, such as alignment, size, and overlap. When evaluated against standard generative layout benchmarks, Dolfin notably improves performance across various metrics (fid, alignment, overlap, MaxIoU and DocSim scores), enhancing transparency and interoperability in the process. Moreover, Dolfin's applications extend beyond layout generation, making it suitable for modeling geometric structures, such as line segments. Our experiments present both qualitative and quantitative results to demonstrate the advantages of Dolfin.

cs.CV↗

arXiv2023

Low-energy electronic interactions in ferrimagnetic Sr2CrReO6 thin films

We reveal in this study the fundamental low-energy landscape in the ferrimagnetic Sr2CrReO6 double perovskite and describe the underlying mechanisms responsible for the three low-energy excitations below 1.4 eV. Based on resonant inelastic x-ray scattering and magnetic dynamics calculations, and experiments collected from both Sr2CrReO6 powders and epitaxially strained thin films, we reveal a strong competition between spin-orbit coupling, Hund's coupling, and the strain-induced tetragonal crystal field. We also demonstrate that a spin-flip process is at the origin of the lowest excitation at 200 meV, and we bring insights into the predicted presence of orbital ordering in this material. We study the nature of the magnons through a combination of ab initio and spin-wave theory calculations, and show that two nondegenerate magnon bands exist and are dominated either by rhenium or chromium spins. The rhenium band is found to be flat at about 200 meV ($\pm$25 meV) through X-L-W-U high-symmetry points and is dispersive toward $Γ$

cond-mat.str-el↗

arXiv2023

Photoemission Evidence of a Novel Charge Order in Kagome Metal FeGe

A charge order has been discovered to emerge deep into the antiferromagnetic phase of the kagome metal FeGe. To study its origin, the evolution of the low-lying electronic structure across the charge order phase transition is investigated with angle-resolved photoemission spectroscopy. We do not find signatures of nesting between Fermi surface sections or van-Hove singularities in zero-frequency joint density of states, and there are no obvious energy gaps at the Fermi level, which exclude the nesting mechanism for the charge order formation in FeGe. However, two obvious changes in the band structure have been detected, i.e., one electron-like band around the K point and another one around the A point move upward in energy position when the charge order forms. These features can be well reproduced by our density-functional theory calculations, where the charge order is primarily driven by magnetic energy saving via large dimerizations of a quarter of Ge1-sites (in the kagome plane) along the c-axis. Our results provide strong support for this novel charge order formation mechanism in FeGe, in contrast to the conventional nesting mechanism.

cond-mat.str-el↗

arXiv2023

An Empirical Study on Real Bug Fixes from Solidity Smart Contract Projects

Smart contracts are pieces of code that reside inside the blockchains and can be triggered to execute any transaction when specifically predefined conditions are satisfied. Being commonly used for commercial transactions in blockchain makes the security of smart contracts particularly important. Over the last few years, we have seen a great deal of academic and practical interest in detecting and fixing the bugs in smart contracts written by Solidity. But little is known about the real bug fixes in Solidity smart contract projects. To understand the bug fixes and enrich the knowledge of bug fixes in real-world projects, we conduct an empirical study on historical bug fixes from 46 real-world Solidity smart contract projects in this paper. We provide a multi-faceted discussion and mainly explore the following four questions: File Type and Amount, Fix Complexity, Bug distribution, and Fix Patches. We distill four findings during the process to explore these four questions. Finally, based on these findings, we provide actionable implications to improve the current approaches to fixing bugs in Solidity smart contracts from three aspects: Automatic repair techniques, Analysis tools, and Solidity developers.

cs.CR↗

arXiv2023

Scalable Optimal Margin Distribution Machine

Optimal margin Distribution Machine (ODM) is a newly proposed statistical learning framework rooting in the novel margin theory, which demonstrates better generalization performance than the traditional large margin based counterparts. Nonetheless, it suffers from the ubiquitous scalability problem regarding both computation time and memory as other kernel methods. This paper proposes a scalable ODM, which can achieve nearly ten times speedup compared to the original ODM training method. For nonlinear kernels, we propose a novel distribution-aware partition method to make the local ODM trained on each partition be close and converge fast to the global one. When linear kernel is applied, we extend a communication efficient SVRG method to accelerate the training further. Extensive empirical studies validate that our proposed method is highly computational efficient and almost never worsen the generalization.

cs.LG↗

arXiv2023

Photoswap: Personalized Subject Swapping in Images

In an era where images and visual content dominate our digital landscape, the ability to manipulate and personalize these images has become a necessity. Envision seamlessly substituting a tabby cat lounging on a sunlit window sill in a photograph with your own playful puppy, all while preserving the original charm and composition of the image. We present Photoswap, a novel approach that enables this immersive image editing experience through personalized subject swapping in existing images. Photoswap first learns the visual concept of the subject from reference images and then swaps it into the target image using pre-trained diffusion models in a training-free manner. We establish that a well-conceptualized visual subject can be seamlessly transferred to any image with appropriate self-attention and cross-attention manipulation, maintaining the pose of the swapped subject and the overall coherence of the image. Comprehensive experiments underscore the efficacy and controllability of Photoswap in personalized subject swapping. Furthermore, Photoswap significantly outperforms baseline methods in human ratings across subject swapping, background preservation, and overall quality, revealing its vast application potential, from entertainment to professional editing.

cs.CV↗