BZPEERReading, trust and collaboration for research
Menu

Discover

GraphFollow connections around a paper, topic or saved view.

Workspaces

HomePick up where your research flow left off.InboxHandle requests, replies and notifications from one place.CollectionsCurate reading lists, evidence sets and shareable dossiers.ResearchSee your private provenance graph, trail and imports.CollaborationMove from discovery into focused discussion rooms.PublishDraft, preview and publish full-text research articles.

Network

ExpertsFind specialists worth following or asking for help.CommunitiesJoin research circles organized around shared work.PartnersSee external organizations active around the same topics.

Opportunities

ListingsBrowse roles, calls and collaborations with clearer fit signals.

Account

Log inReturn to your reading and collaboration workspace.Create accountStart saving work, following people and joining teams.
Log inCreate account

Researcher profile

Hui Yu

Hui Yu contributes to research discovery and scholarly infrastructure.

ResearcherAffiliation not importedOpen to collaborate
Computer Visionmath.APArtificial Intelligenceeess.SPInformation Theorymath.ITastro-ph.SRComputation and Languagecond-mat.str-elcond-mat.supr-conCryptography and SecurityDatabaseseess.IVInformation RetrievalMachine Learningmath.CVmath.NANumerical Analysisphysics.ao-phquant-ph

Trust snapshot

Quick read

Trust 21 - EmergingVerification L1Unclaimed author
19works
0followers
20topics
4close collaborators

Actions

Decide how to stay connected

Follow researcher0

Identity and collaboration

How to connect with this researcher

Claiming links this public author record to a researcher profile and unlocks direct collaboration workflows.

Log in to claim

Direct collaboration

Open a focused conversation when the fit is right

Claim this author entity first to unlock direct invitations.

Research graph

See the researcher in context

Open full explorer

Inspect adjacent work, topics, institutions and collaborators without jumping out to a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

Published work

19 published item(s)

preprint2026arXiv

LaCoVL-FER: Landmark-Guided Contrastive Learning Network with Vision-Language Enhancement for Facial Expression Recognition

Facial Expression Recognition (FER) in the wild is still challenging due to uncontrolled variations in pose, occlusion, and illumination. Most existing attention-based methods primarily rely on visual appearance cues, suffering from attention redundancy and instability, which limits their performance in complex scenarios. To address these issues, we propose a novel landmark-guided contrastive learning network with vision-language enhancement for FER (LaCoVL-FER), which integrates geometric priors from facial landmarks and semantic priors from a vision-language model. Specifically, a Landmark-Guided Adaptive Encoder (LGAE) is designed to introduce geometric priors through a Bi-branch Gated Cross Attention (BGCA) mechanism, which achieves adaptive fusion of landmark-based geometric and visual appearance features to produce expression-relevant features, thereby focusing on key facial regions and suppressing noise interference. In parallel, a Vision-Language Enhancement Strategy (VLES) is presented to leverage the expression-relevant features to refine the generalizable visual features extracted by the frozen pretrained CLIP image encoder, yielding expression-specific visual representations. Based on these representations, an Expression-Conditioned Prompting (ECP) mechanism is utilized to further adapt the textual features of fixed class-level prompts from the frozen pretrained CLIP text encoder, generating more instance-aware textual representations. These visual-textual representations are aligned as semantic priors to enhance the robustness and generalization of FER. Quantitative and qualitative experiments demonstrate that our LaCoVL-FER outperforms state-of-the-art methods on three representative real-world FER datasets, including RAF-DB, FERPlus, and AffectNet. The code is available at https://github.com/ylin06804/LaCoVL-FER.

0 citations0 reviewsNo code linkedOpen access
preprint2025arXiv

Topological Degeneracy Induced Flat Bands in two-Dimensional Holed Systems

Systems hosting flat bands offer a powerful platform for exploring strong correlation physics. Theoretically topological degeneracy rising in systems with non-trivial topological orders on periodic manifolds of non-zero genus can generate ideal flat bands. However, experimental realization of such geometrically engineered systems is very difficult. In this work, we demonstrate that flat planes with strategically patterned hole defects can engineer ideal flat bands. We constructing two families of models, singular flat band systems where degeneracy is stabilized by non-contractible loop excitations tied to hole defects and perfectly nested van Hove systems where degeneracy arises from line excitations in momentum space. These models circumvent the need for exotic manifolds while retaining the essential features of topological flat bands. By directly linking defect engineering to degeneracy mechanisms, our results establish a scalable framework for experimentally accessible flat band design.

0 citations0 reviewsNo code linkedOpen access
preprint2024arXiv

Attention-Guided Erasing: A Novel Augmentation Method for Enhancing Downstream Breast Density Classification

The assessment of breast density is crucial in the context of breast cancer screening, especially in populations with a higher percentage of dense breast tissues. This study introduces a novel data augmentation technique termed Attention-Guided Erasing (AGE), devised to enhance the downstream classification of four distinct breast density categories in mammography following the BI-RADS recommendation in the Vietnamese cohort. The proposed method integrates supplementary information during transfer learning, utilizing visual attention maps derived from a vision transformer backbone trained using the self-supervised DINO method. These maps are utilized to erase background regions in the mammogram images, unveiling only the potential areas of dense breast tissues to the network. Through the incorporation of AGE during transfer learning with varying random probabilities, we consistently surpass classification performance compared to scenarios without AGE and the traditional random erasing transformation. We validate our methodology using the publicly available VinDr-Mammo dataset. Specifically, we attain a mean F1-score of 0.5910, outperforming values of 0.5594 and 0.5691 corresponding to scenarios without AGE and with random erasing (RE), respectively. This superiority is further substantiated by t-tests, revealing a p-value of p<0.0001, underscoring the statistical significance of our approach.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

A Deep Learning Method for Real-time Bias Correction of Wind Field Forecasts in the Western North Pacific

Forecasts by the European Centre for Medium-Range Weather Forecasts (ECMWF; EC for short) can provide a basis for the establishment of maritime-disaster warning systems, but they contain some systematic biases.The fifth-generation EC atmospheric reanalysis (ERA5) data have high accuracy, but are delayed by about 5 days. To overcome this issue, a spatiotemporal deep-learning method could be used for nonlinear mapping between EC and ERA5 data, which would improve the quality of EC wind forecast data in real time. In this study, we developed the Multi-Task-Double Encoder Trajectory Gated Recurrent Unit (MT-DETrajGRU) model, which uses an improved double-encoder forecaster architecture to model the spatiotemporal sequence of the U and V components of the wind field; we designed a multi-task learning loss function to correct wind speed and wind direction simultaneously using only one model. The study area was the western North Pacific (WNP), and real-time rolling bias corrections were made for 10-day wind-field forecasts released by the EC between December 2020 and November 2021, divided into four seasons. Compared with the original EC forecasts, after correction using the MT-DETrajGRU model the wind speed and wind direction biases in the four seasons were reduced by 8-11% and 9-14%, respectively. In addition, the proposed method modelled the data uniformly under different weather conditions. The correction performance under normal and typhoon conditions was comparable, indicating that the data-driven mode constructed here is robust and generalizable.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Application of Data Encryption in Chinese Named Entity Recognition

Recently, with the continuous development of deep learning, the performance of named entity recognition tasks has been dramatically improved. However, the privacy and the confidentiality of data in some specific fields, such as biomedical and military, cause insufficient data to support the training of deep neural networks. In this paper, we propose an encryption learning framework to address the problems of data leakage and inconvenient disclosure of sensitive data in certain domains. We introduce multiple encryption algorithms to encrypt training data in the named entity recognition task for the first time. In other words, we train the deep neural network using the encrypted data. We conduct experiments on six Chinese datasets, three of which are constructed by ourselves. The experimental results show that the encryption method achieves satisfactory results. The performance of some models trained with encrypted data even exceeds the performance of the unencrypted method, which verifies the effectiveness of the introduced encryption method and solves the problem of data leakage to a certain extent.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Filter-enhanced MLP is All You Need for Sequential Recommendation

Recently, deep neural networks such as RNN, CNN and Transformer have been applied in the task of sequential recommendation, which aims to capture the dynamic preference characteristics from logged user behavior data for accurate recommendation. However, in online platforms, logged user behavior data is inevitable to contain noise, and deep recommendation models are easy to overfit on these logged data. To tackle this problem, we borrow the idea of filtering algorithms from signal processing that attenuates the noise in the frequency domain. In our empirical experiments, we find that filtering algorithms can substantially improve representative sequential recommendation models, and integrating simple filtering algorithms (eg Band-Stop Filter) with an all-MLP architecture can even outperform competitive Transformer-based models. Motivated by it, we propose \textbf{FMLP-Rec}, an all-MLP model with learnable filters for sequential recommendation task. The all-MLP architecture endows our model with lower time complexity, and the learnable filters can adaptively attenuate the noise information in the frequency domain. Extensive experiments conducted on eight real-world datasets demonstrate the superiority of our proposed method over competitive RNN, CNN, GNN and Transformer-based methods. Our code and data are publicly available at the link: \textcolor{blue}{\url{https://github.com/RUCAIBox/FMLP-Rec}}.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

High-Pressure NMR Enabled by Diamond Nitrogen-Vacancy Centers

The integration of NMR and high pressure technique brings unique opportunities to study electronic, structural and dynamical properties under extreme conditions. Despite a great degree of success has been achieved using coil-based schemes, the contradictory requirement on sample volume of these two techniques remains an outstanding challenge. In this letter, we introduce diamond nitrogen-vacancy (NV) centers, as the source and probe of in-situ nuclear spin polarization, to address the sample volume issue. We demonstrate hyperpolarization and coherent control of $^{14}$N nuclear spins under high pressures. NMR spectra of a micro-diamond are measured up to 16.6 GPa, and unexpected pressure shift of the $^{14}$N nuclear quadrupole and hyperfine coupling terms are observed. Our work contributes to quantum sensing enhanced spectrometry under extreme conditions.

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Nebula Graph: An open source distributed graph database

This paper introduces the recent work of Nebula Graph, an open-source, distributed, scalable, and native graph database. We present a system design trade-off and a comprehensive overview of Nebula Graph internals, including graph data models, partitioning strategies, secondary indexes, optimizer rules, storage-side transactions, graph query languages, observability, graph processing frameworks, and visualization tool-kits. In addition, three sets of large-scale graph b

0 citations0 reviewsNo code linkedOpen access
preprint2022arXiv

Quantum critical points, lines and surfaces

In this paper we promote the idea of quantum critical lines ({\em inter alia} surfaces) as opposed to points. A quantum critical line obtains when criticality at zero temperature is extended over a continuum in a one-dimensional line. We base our ideas on a simple but exactly solved model introduced by one of the authors involving a one-dimensional quantum transverse field Ising model with added 3-spin interaction. While many of the ideas are quite general, there are other aspects that are not. In particular, a line of criticality with continuously varying exponents is not captured. However, the exact solvability of the model gives us considerable confidence in our results. Although the pure system is analytically exactly solved, the disorder case requires numerical analyses based on exact computation of the correlation function in the Pfaffian representation. The disorder case leads to dynamic structure factor as a function of frequency and wave vector. We expect that the model is experientally realizable and perhaps many other similar models will be found to explore quantum critical lines.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

On the fine regularity of the singular set in the nonlinear obstacle problem

We revisit and sharpen the results from our previous work, where we investigated the regularity of the singular set of the free boundary in the nonlinear obstacle problem. As in the work of Figalli-Serra on the classical obstacle problem, we show that each stratum can be further decomposed into a `good&#39; part and an `anomalous&#39; part, where the former is covered by $C^{1,1-}$ manifolds, and the latter is of lower dimension.

0 citations0 reviewsNo code linkedOpen access
preprint2021arXiv

Positivity-preserving third order DG schemes for Poisson--Nernst--Planck equations

In this paper, we design and analyze third order positivity-preserving discontinuous Galerkin (DG) schemes for solving the time-dependent system of Poisson--Nernst--Planck (PNP) equations, which has found much use in diverse applications. Our DG method with Euler forward time discretization is shown to preserve the positivity of cell averages at all time steps. The positivity of numerical solutions is then restored by a scaling limiter in reference to positive weighted cell averages. The method is also shown to preserve steady states. Numerical examples are presented to demonstrate the third order accuracy and illustrate the positivity-preserving property in both one and two dimensions.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

3D Facial Geometry Recovery from a Depth View with Attention Guided Generative Adversarial Network

We present to recover the complete 3D facial geometry from a single depth view by proposing an Attention Guided Generative Adversarial Networks (AGGAN). In contrast to existing work which normally requires two or more depth views to recover a full 3D facial geometry, the proposed AGGAN is able to generate a dense 3D voxel grid of the face from a single unconstrained depth view. Specifically, AGGAN encodes the 3D facial geometry within a voxel space and utilizes an attention-guided GAN to model the illposed 2.5D depth-3D mapping. Multiple loss functions, which enforce the 3D facial geometry consistency, together with a prior distribution of facial surface points in voxel space are incorporated to guide the training process. Both qualitative and quantitative comparisons show that AGGAN recovers a more complete and smoother 3D facial shape, with the capability to handle a much wider range of view angles and resist to noise in the depth view than conventional methods

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Damping of slow surface kink modes in solar photospheric waveguides modeled by one-dimensional inhomogeneities

Given the recent interest in magnetohydrodynamic (MHD) waves in pores and sunspot umbrae, we examine the damping of slow surface kink modes (SSKMs) by modeling solar photospheric waveguides with a cylindrical inhomogeneity comprising a uniform interior, a uniform exterior, and a continuous transition layer (TL) in between. Performing an eigen-mode analysis in linear, resistive, gravity-free MHD, our approach is idealized in that, among other things, our equilibrium is structured only in the radial direction. We can nonetheless address two damping mechanisms simultaneously, one being the Ohmic resistivity, and the other being the resonant absorption of SSKMs in the cusp and Alfv$\acute{\rm e}$n continua. We find that the relative importance of the two mechanisms depends sensitively on the magnetic Reynolds number ($R_{\rm m}$). Resonant absorption is the sole damping mechanism for realistically large values of $R_{\rm m}$, and the cusp resonance in general dominates the Alfv$\acute{\rm e}$n one unless the axial wavenumbers are at the lower end of the observationally relevant range. We also find that the thin-boundary approximation holds only when the TL-width-to-radius ratios are much smaller than nominally expected. The Ohmic resistivity is far more important for realistically small $R_{\rm m}$. Even in this case, SSKMs are only marginally damped, with damping-time-to-period-ratios reaching $\sim 10$ in the parameter range we examine.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

PIoU Loss: Towards Accurate Oriented Object Detection in Complex Environments

Object detection using an oriented bounding box (OBB) can better target rotated objects by reducing the overlap with background areas. Existing OBB approaches are mostly built on horizontal bounding box detectors by introducing an additional angle dimension optimized by a distance loss. However, as the distance loss only minimizes the angle error of the OBB and that it loosely correlates to the IoU, it is insensitive to objects with high aspect ratios. Therefore, a novel loss, Pixels-IoU (PIoU) Loss, is formulated to exploit both the angle and IoU for accurate OBB regression. The PIoU loss is derived from IoU metric with a pixel-wise form, which is simple and suitable for both horizontal and oriented bounding box. To demonstrate its effectiveness, we evaluate the PIoU loss on both anchor-based and anchor-free frameworks. The experimental results show that PIoU loss can dramatically improve the performance of OBB detectors, particularly on objects with high aspect ratios and complex backgrounds. Besides, previous evaluation datasets did not include scenarios where the objects have high aspect ratios, hence a new dataset, Retail50K, is introduced to encourage the community to adapt OBB detectors for more complex environments.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

QoS-Based Source and Relay Secure Optimization Design with Presence of Channel Uncertainty

In this letter, we study relay-aided networks with presence of single eavesdropper. We provide joint beamforming design of the source and relay that can minimize the overall power consumption while satisfying our predefined quality-of-service (QoS) requirements. Additionally, we investigate the case that the channel between relay and eavesdropper suffers from channel uncertainty. Finally, simulation results are provided to verify the effectiveness of our algorithm.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Regularity of the singular set in the fully nonlinear obstacle problem

For the obstacle problem involving a convex fully nonlinear elliptic operator, we show that the singular set in the free boundary stratifies. The top stratum is locally covered by a $C^{1,α}$-manifold, and the lower strata are covered by $C^{1,\log^\varepsilon}$-manifolds. This essentially recovers the regularity result obtained by Figalli-Serra when the operator is the Laplacian.

0 citations0 reviewsNo code linkedOpen access
preprint2020arXiv

Robust Beamforming Design for Sum Secrecy Rate Optimization in MU-MISO Networks

This paper studies the beamforming design problem of a multi-user downlink network, assuming imperfect channel state information known to the base station. In this scenario, the base station is equipped with multiple antennas, and each user is wiretapped by a specific eavesdropper where each user or eavesdropper is equipped with one antenna. It is supposed that the base station employs transmit beamforming with a given requirement on sum transmitting power. The objective is to maximize the sum secrecy rate of the network. Due to the uncertainty of the channel, it is difficult to calculate the exact sum secrecy rate of the system. Thus, the maximum of lower bound of sum secrecy rate is considered. The optimization of the lower bound of sum secrecy rate still makes the considered beamforming design problem difficult to handle. To solve this problem, a beamforming design scheme is proposed to transform the original problem into a convex approximation problem, by employing semidefinite relaxation and first-order approximation technique based on Taylor expansion. Besides, with the advantage of low complexity, a zero-forcing based beamforming method is presented in the case that base station is able to nullify the eavesdroppers&#39; rate. When the base station doesn&#39;t have the ability, user selection algorithm would be in use. Numerical results show that the former strategy achieves better performance than the latter one, which is mainly due to the ability of optimizing beamforming direction, and both outperform the signal-to-leakage-and-noise ratio based algorithm.

0 citations0 reviewsNo code linkedOpen access