arXiv subjects

Yujun Zhang

Publications and source records attributed to Yujun Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2025

MamNet: A Novel Hybrid Model for Time-Series Forecasting and Frequency Pattern Analysis in Network Traffic

The abnormal fluctuations in network traffic may indicate potential security threats or system failures. Therefore, efficient network traffic prediction and anomaly detection methods are crucial for network security and traffic management. This paper proposes a novel network traffic prediction and anomaly detection model, MamNet, which integrates time-domain modeling and frequency-domain feature extraction. The model first captures the long-term dependencies of network traffic through the Mamba module (time-domain modeling), and then identifies periodic fluctuations in the traffic using Fourier Transform (frequency-domain feature extraction). In the feature fusion layer, multi-scale information is integrated to enhance the model's ability to detect network traffic anomalies. Experiments conducted on the UNSW-NB15 and CAIDA datasets demonstrate that MamNet outperforms several recent mainstream models in terms of accuracy, recall, and F1-Score. Specifically, it achieves an improvement of approximately 2% to 4% in detection performance for complex traffic patterns and long-term trend detection. The results indicate that MamNet effectively captures anomalies in network traffic across different time scales and is suitable for anomaly detection tasks in network security and traffic management. Future work could further optimize the model structure by incorporating external network event information, thereby improving the model's adaptability and stability in complex network environments.

cs.LG↗

arXiv2025

NET-SA: An Efficient Secure Aggregation Architecture Based on In-Network Computing

Privacy-preserving machine learning (PPML) enables clients to collaboratively train deep learning models without sharing private datasets, but faces privacy leakage risks due to gradient leakage attacks. Prevailing methods leverage secure aggregation strategies to enhance PPML, where clients leverage masks and secret sharing to further protect gradient data while tolerating participant dropouts. These methods, however, require frequent inter-client communication to negotiate keys and perform secret sharing, leading to substantial communication overhead. To tackle this issue, we propose NET-SA, an efficient secure aggregation architecture for PPML based on in-network computing. NET-SA employs seed homomorphic pseudorandom generators for local gradient masking and utilizes programmable switches for seed aggregation. Accurate and secure gradient aggregation is then performed on the central server based on masked gradients and aggregated seeds. This design effectively reduces communication overhead due to eliminating the communication-intensive phases of seed agreement and secret sharing, with enhanced dropout tolerance due to overcoming the threshold limit of secret sharing. Extensive experiments on server clusters and Intel Tofino programmable switch demonstrate that NET-SA achieves up to 77x and 12x enhancements in runtime and 2x decrease in total client communication cost compared with state-of-the-art methods.

cs.CR↗

arXiv2024

Spectral-Based Graph Neural Networks for Complementary Item Recommendation

Modeling complementary relationships greatly helps recommender systems to accurately and promptly recommend the subsequent items when one item is purchased. Unlike traditional similar relationships, items with complementary relationships may be purchased successively (such as iPhone and Airpods Pro), and they not only share relevance but also exhibit dissimilarity. Since the two attributes are opposites, modeling complementary relationships is challenging. Previous attempts to exploit these relationships have either ignored or oversimplified the dissimilarity attribute, resulting in ineffective modeling and an inability to balance the two attributes. Since Graph Neural Networks (GNNs) can capture the relevance and dissimilarity between nodes in the spectral domain, we can leverage spectral-based GNNs to effectively understand and model complementary relationships. In this study, we present a novel approach called Spectral-based Complementary Graph Neural Networks (SComGNN) that utilizes the spectral properties of complementary item graphs. We make the first observation that complementary relationships consist of low-frequency and mid-frequency components, corresponding to the relevance and dissimilarity attributes, respectively. Based on this spectral observation, we design spectral graph convolutional networks with low-pass and mid-pass filters to capture the low-frequency and mid-frequency components. Additionally, we propose a two-stage attention mechanism to adaptively integrate and balance the two attributes. Experimental results on four e-commerce datasets demonstrate the effectiveness of our model, with SComGNN significantly outperforming existing baseline models.

cs.IR↗

arXiv2024

Enhance Graph Alignment for Large Language Models

Graph-structured data is prevalent in the real world. Recently, due to the powerful emergent capabilities, Large Language Models (LLMs) have shown promising performance in modeling graphs. The key to effectively applying LLMs on graphs is converting graph data into a format LLMs can comprehend. Graph-to-token approaches are popular in enabling LLMs to process graph information. They transform graphs into sequences of tokens and align them with text tokens through instruction tuning, where self-supervised instruction tuning helps LLMs acquire general knowledge about graphs, and supervised fine-tuning specializes LLMs for the downstream tasks on graphs. Despite their initial success, we find that existing methods have a misalignment between self-supervised tasks and supervised downstream tasks, resulting in negative transfer from self-supervised fine-tuning to downstream tasks. To address these issues, we propose Graph Alignment Large Language Models (GALLM) to benefit from aligned task templates. In the self-supervised tuning stage, we introduce a novel text matching task using templates aligned with downstream tasks. In the task-specific tuning stage, we propose two category prompt methods that learn supervision information from additional explanation with further aligned templates. Experimental evaluations on four datasets demonstrate substantial improvements in supervised learning, multi-dataset generalizability, and particularly in zero-shot capability, highlighting the model's potential as a graph foundation model.

cs.CL↗

arXiv2024

Photo-induced phase transition on black samarium monosulfide

To investigate the role of the excitons for the origin of the pressure-induced phase transition (BGT) from the black-colored insulator (BI) to the golden-yellow-colored metal (GM) of samarium monosulfide (SmS), optical reflectivity, Sm $3d$ X-ray absorption spectroscopy (XAS), and X-ray diffraction (XRD) with the creation of excitons by photoexcitation (PE) are reported. In the pump-probe reflectivity measurement, following a huge reflectivity change of about 22 %, three different relaxation times with a vibration component were observed. The fast component with the relaxation time ($τ$) of less than 1 ps is due to the excitation and relaxation of electrons into the conduction band, and the slowest one with $τ> {\rm several} 100$ ps originates from the appearance of the photo-induced (PI) state. The components with $τ\sim 10$ ps and vibration originate from the appearance of the PI state and the interference between the reflection lights at the sample surface and the boundary between the BI and PI states, suggesting that the electronic structure of the PI phase is different from that of the BI state. XAS spectra indicate that the Sm mean valence is shifted from the Sm$^{2+}$ dominant to the intermediate between Sm$^{2+}$ and Sm$^{3+}$ by PE but did not change to that of the GM phase across BGT, consistent with the reflectivity data. The XRD result after PE shows that the PI state has much less lattice contraction than the GM phase. These results suggest that the BGT cannot be achieved solely by creating excitons after PE but requires other effects, such as a lattice contraction.

cond-mat.str-el↗

arXiv2024

Flexomagnetoelectric effect in Sr2IrO4 thin films

Symmetry engineering is explicitly effective to manipulate and even create phases and orderings in strongly correlated materials. Flexural stress is universally practical to break the space-inversion or time-reversal symmetry. Here, by introducing strain gradient in a centrosymmetric antiferromagnet Sr2IrO4, the space-inversion symmetry is broken accompanying a non-equivalent O p-Ir d orbital hybridization along z axis. Thus, emergent polar phase and out-of-plane magnetic moment have been simultaneously observed in these asymmetric Sr2IrO4 thin films, which both are absent in its ground state. Furthermore, upon the application of magnetic field, such polarization can be controlled by modifying the occupied d orbitals through spin-orbit interaction, giving rise to a flexomagnetoelectric effect. This work provides a general strategy to artificially design multiple symmetries and ferroic orderings in strongly correlated systems.

cond-mat.str-el↗

arXiv2023

4$f$ electron temperature driven ultrafast electron localization

Valence transitions in strongly correlated electron systems are caused by orbital hybridization and Coulomb interactions between localized and delocalized electrons. The transition can be triggered by changes in the electronic structure and is sensitive to temperature variations, applications of magnetic fields, and physical or chemical pressure. Launching the transition by photoelectric fields can directly excite the electronic states and thus provides an ideal platform to study the correlation among electrons on ultrafast timescales. The EuNi$_2$(Si$_{0.21}$Ge$_{0.79}$)$_2$ mixed-valence metal is an ideal material to investigate the valence transition of the Eu ions via the amplified orbital hybridization by the photoelectric field on sub-picosecond timescales. A direct view on the 4$f$ electron occupancy of the Eu ions is required to understand the microscopic origin of the transition. Here we probe the 4$f$ electron states of EuNi$_2$(Si$_{0.21}$Ge$_{0.79}$)$_2$ at the sub-ps timescale after photoexcitation by X-ray absorption spectroscopy across the Eu $M_5$-absorption edge. The observed spectral changes due to the excitation indicate a population change of total angular momentum multiplet states $J$ = 0, 1, 2, and 3 of Eu$^{3+}$, and the Eu$^{2+}$ $J$ = 7/2 multiplet state caused by an increase in 4$f$ electron temperature that results in a 4$f$ localization process. This electronic temperature increase combined with fluence-dependent screening accounts for the strongly non-linear effective valence change. The data allow us to extract a time-dependent determination of an effective temperature of the 4$f$ shell, which is also of great relevance in the understanding of metallic systems' properties, such as the ultrafast demagnetization of ferromagnetic rare-earth intermetallics and their all-optical magnetization switching.

cond-mat.str-el↗

arXiv2023

Overheated Topological Hall Effect

The topological Hall effect (THE) originates from the real-space Berry phase that an electron gains when its spin follows the spatially varying non-trivial magnetization textures, such as skyrmions. Such topologically protected magnetization textures can provide great potential for information storage and processing. Since directly imaging the skyrmions or detecting the magnetic diffraction of skyrmion lattice are significantly more challenging than conducting Hall measurements, THE has been widely used to attest the presence of skyrmions. However, the key feature of THE, namely the bump/dip in the Hall signal is not sufficient proof of THE. Here, we use empirical numerical modeling to demonstrate all possible THE-like signals that two anomalous Hall effect (AHE) signals with opposite signs can superpose. Besides the reproduction of many published results by the numerical model, we propose an exotic {\lq THE\rq} could, in principle, emerge with finely tuned two-channel AHE. The importance of the scrupulous re-examination of the THE observed in experiments cannot be exaggerated.

cond-mat.mtrl-sci↗

arXiv2023

Photocarrier Transport of Ferroelectric Photovoltaic Thin Films Detected by the Magnetic Dynamics of Adjacent Ferromagnetic Layers

We have observed photocarrier transport behaviors in BiFeO$_3$/La$_{1-x}$Sr$_x$MnO$_3$~(BFO/LSMO) heterostructures by using time-resolved synchrotron x-ray magnetic circular dichroism in reflectivity. The magnetization of LSMO layers was used as a probe of photo-induced carrier dynamics in the photovoltaic BFO layers. During the photo-induced demagnetization process, the decay time of LSMO~($x$=0.2) magnetization strongly depends on the ferroelectric polarization direction of the BFO layer. The variation of decay time should be attributed to the different sign of accumulated photocarriers at the BFO/LSMO interface induced by the photovoltaic effect of the BFO layer. The photocarriers can reach the BFO/LSMO interface and influence the magnetization distribution in the LSMO layers within the timescale of $\sim$100~ps. Our results provide a novel strategy to investigate carrier dynamics and mechanisms of optical control of magnetization in thin film heterostructures.

cond-mat.mtrl-sci↗

arXiv2023

Significant Unconventional Anomalous Hall Effect in Heavy Metal/Antiferromagnetic Insulator Heterostructures

The anomalous Hall effect (AHE) is a quantum coherent transport phenomenon that conventionally vanishes at elevated temperatures because of thermal dephasing. Therefore, it is puzzling that the AHE can survive in heavy metal (HM)/antiferromagnetic (AFM) insulator (AFMI) heterostructures at high temperatures yet disappears at low temperatures. In this paper, we report that an unconventional high-temperature AHE in HM/AFMI is observed only around the Néel temperature of AFM, with large anomalous Hall resistivity up to 40 n$Ω$ cm. This mechanism is attributed to the emergence of a noncollinear AFM spin texture with a non-zero net topological charge. Atomistic spin dynamics simulation shows that such a unique spin texture can be stabilized by the subtle interplay among the collinear AFM exchange coupling, interfacial Dyzaloshinski-Moriya interaction, thermal fluctuation, and bias magnetic field.

cond-mat.mes-hall↗

arXiv2022

Nuclear phase retrieval spectroscopy using resonant x-ray scattering

Light-matter interaction is exploited in spectroscopic techniques to access information about molecular, atomic or nuclear constituents of the sample of interest. While scattered light carries both amplitude and phase information of the electromagnetic field, most of the time the latter is lost in intensity measurements. However, often the phase information is paramount to reconstruct the desired information of the target, as it is well known from coherent x-ray imaging. Here we introduce a new phase retrieval algorithm which allows us to reconstruct the field phase information from two-dimensional time- and energy-resolved spectra. We apply this method to the particular case of x-ray scattering off Mössbauer nuclei at a synchrotron radiation source. Knowledge of the phase allows also for an excellent reconstruction of the energy spectra from experimental data, which could not be achieved with this resolution otherwise. Our approach provides an efficient novel data analysis tool which will benefit x-ray quantum optics and Mössbauer spectroscopy with synchrotron radiation alike.

physics.ins-det↗

arXiv2021

VIL-100: A New Dataset and A Baseline Model for Video Instance Lane Detection

Lane detection plays a key role in autonomous driving. While car cameras always take streaming videos on the way, current lane detection works mainly focus on individual images (frames) by ignoring dynamics along the video. In this work, we collect a new video instance lane detection (VIL-100) dataset, which contains 100 videos with in total 10,000 frames, acquired from different real traffic scenarios. All the frames in each video are manually annotated to a high-quality instance-level lane annotation, and a set of frame-level and video-level metrics are included for quantitative performance evaluation. Moreover, we propose a new baseline model, named multi-level memory aggregation network (MMA-Net), for video instance lane detection. In our approach, the representation of current frame is enhanced by attentively aggregating both local and global memory features from other frames. Experiments on the new collected dataset show that the proposed MMA-Net outperforms state-of-the-art lane detection methods and video object segmentation methods. We release our dataset and code at https://github.com/yujun0-0/MMA-Net.

cs.CV↗

arXiv2021

Photo-induced Antiferromagnetic-ferromagnetic Transition and Electronic Structure Modulation in GdBaCo$_2$O$_{5.5}$ Thin Film

We investigate both the ferromagnetic (FM) and antiferromagnetic (AFM) ultrafast dynamics of a strongly correlated oxide system, GdBaCo$_2$O$_{5.5}$ thin film, by time-resolved x-ray magnetic circular dichroism in reflectivity (XMCDR) and resonant magnetic x-ray diffraction (RMXD). A photo-induced AFM-FM transition characterized by an increase of the transient XMCDR beyond the unpumped value and a decay of RMXD was observed. The photon-energy dependence of the transient XMCDR and reflectivity can likely be interpreted as a concomitant photo-induced spinstate transition.

cond-mat.str-el↗

arXiv2020

Berry phase manipulation in ultrathin SrRuO$_3$ films

A notion of the Berry phase is a powerful means to unravel the non-trivial role of topology in various novel phenomena observed in chiral magnetic materials and structures. A celebrated example is the intrinsic anomalous Hall effect (AHE) driven by the non-vanishing Berry phase in the momentum space. As the AHE is highly dependent on details of the band structure near the Fermi edge, the Berry phase and AHE can be altered in thin films whose chemical potential is tunable by dimensionality and disorder. Here, we demonstrate that in ultrathin SrRuO$_3$ films the Berry phase can be effectively manipulated by the effects of disorder on the intrinsic Berry phase contribution to the AHE, which is corroborated by our numerically exact calculations. In addition, our findings provide ample experimental evidence for the superficial nature of the topological Hall effect attribution to the protected spin texture and instead lend strong support to the multi-channel AHE scenario in ultrathin SrRuO$_3$.

cond-mat.str-el↗

arXiv2020

Adaptive In-network Collaborative Caching for Enhanced Ensemble Deep Learning at Edge

To enhance the quality and speed of data processing and protect the privacy and security of the data, edge computing has been extensively applied to support data-intensive intelligent processing services at edge. Among these data-intensive services, ensemble learning-based services can in natural leverage the distributed computation and storage resources at edge devices to achieve efficient data collection, processing, analysis. Collaborative caching has been applied in edge computing to support services close to the data source, in order to take the limited resources at edge devices to support high-performance ensemble learning solutions. To achieve this goal, we propose an adaptive in-network collaborative caching scheme for ensemble learning at edge. First, an efficient data representation structure is proposed to record cached data among different nodes. In addition, we design a collaboration scheme to facilitate edge nodes to cache valuable data for local ensemble learning, by scheduling local caching according to a summarization of data representations from different edge nodes. Our extensive simulations demonstrate the high performance of the proposed collaborative caching scheme, which significantly reduces the learning latency and the transmission overhead.

cs.NI↗

arXiv2020

From Anomalous to Normal: Temperature Dependence of the Band Gap in Two-Dimensional Black Phosphorus

The temperature dependence of the band gap is crucial to a semiconductor. Bulk black phosphorus (BP) is known to exhibit an anomalous behavior. Through optical spectroscopy, here we show that the temperature effect on BP band gap gradually evolves with decreasing layer number, eventually turns into a normal one in the monolayer limit, rendering a crossover from the anomalous to the normal. Meanwhile, the temperature-induced shift in optical resonance also differs with different transition indices for the same thickness sample. A comprehensive analysis reveals that the temperature-tunable interlayer coupling is responsible for the observed diverse scenario. Our study provides a key to the apprehension of the anomalous temperature behavior in certain layered semiconductors.

cond-mat.mtrl-sci↗

arXiv2020

Spin-orbital liquid in Ba$_3$CuSb$_2$O$_9$ stabilized by oxygen holes

Both the Jahn-Teller distortion of Cu$^{2+}$O$_6$ octahedra and magnetic ordering are absent in hexagonal Ba$_3$CuSb$_2$O$_9$ suggesting a Cu 3$d$ spin-orbital liquid state. Here, by means of resonant x-ray scattering and absorption experiment, we show that oxygen 2$p$ holes play crucial role in stabilizing this spin-orbital liquid state. These oxygen holes appear due to the "reaction" Sb$^{5+}$$\rightarrow$Sb$^{3+}$ $+$ two oxygen holes, with these holes being able to attach to Cu ions. The hexagonal phase with oxygen 2$p$ holes exhibits also a novel charge-orbital dynamics which is absent in the orthorhombic phase of Ba$_3$CuSb$_2$O$_9$ with Jahn-Teller distortion and Cu 3$d$ orbital order. The present work opens up a new avenue towards spin-charge-orbital entangled liquid state in transition-metal oxides with small or negative charge transfer energy.

cond-mat.str-el↗

arXiv2020

Interfacial-hybridization-modified Ir Ferromagnetism and Electronic Structure in LaMnO$_3$/SrIrO$_3$ Superlattices

Artificially fabricated 3$d$/5$d$ superlattices (SLs) involve both strong electron correlation and spin-orbit coupling in one material by means of interfacial 3$d$-5$d$ coupling, whose mechanism remains mostly unexplored. In this work we investigated the mechanism of interfacial coupling in LaMnO$_3$/SrIrO$_3$ SLs by several spectroscopic approaches. Hard x-ray absorption, magnetic circular dichroism and photoemission spectra evidence the systematic change of the Ir ferromagnetism and the electronic structure with the change of the SL repetition period. First-principles calculations further reveal the mechanism of the SL-period dependence of the interfacial electronic structure and the local properties of the Ir moments, confirming that the formation of Ir-Mn molecular orbital is responsible for the interfacial coupling effects. The SL-period dependence of the ratio between spin and orbital components of the Ir magnetic moments can be attributed to the realignment of electron spin during the formation of the interfacial molecular orbital. Our results clarify the nature of interfacial coupling in this prototypical 3$d$/5$d$ SL system and the conclusion will shed light on the study of other strongly correlated and spin-orbit coupled oxide hetero-interfaces.

cond-mat.str-el↗