arXiv subjects

Zhang Yue

Publications and source records attributed to Zhang Yue.

6 recordsLinked to original sources

arXiv2025

Supermassive Black Hole and Broad-line Region in NGC 5548: 2023 Reverberation Mapping Results

We present the results of the 2023 spectroscopic reverberation mapping (RM) campaign for active galactic nuclei (AGN) of NGC 5548, continuing our long-term monitoring program. Using the Lijiang 2.4-meter telescope, we obtained 74 spectra with a median cadence of 1.9 days. Through detailed spectral decomposition, we measured the light curves of the optical continuum at 5100~\AA\ and the broad He~{\sc ii}, He~{\sc i}, H$\gamma$, and H$\beta$ emission lines. The time lags of these lines relative to the continuum are measured as $1.3^{+1.6}_{-0.6}$, $2.3^{+1.5}_{-2.1}$, $10.0^{+2.0}_{-1.8}$, and $15.6^{+2.6}_{-2.9}$ days (rest-frame), respectively. Velocity-resolved lag profiles for H$\gamma$ and H$\beta$ were constructed. Combined with data from previous seasons (2015$-$2021), we find that the radial ionization stratification of the broad-line region (BLR) is stable; the average virial mass of the supermassive black hole in NGC~5548 is $(2.6\pm1.1)\times 10^{8}M_{\odot}$, consistent with the $M_{\rm BH}-\sigma_*$ relation; the broad He~{\sc ii} line exhibits the largest responsivity, followed by broad He~{\sc i} (or H$\gamma$) and H$\beta$ lines; the BLR kinematics show significant temporal evolution, transitioning from virialized motions to signatures of inflow and outflow. Furthermore, an analysis of 35 years of historical data confirms a 3.5-year time lag between variations in the optical luminosity and the BLR radius, potentially implicating the role of radiation pressure or dynamical structure changes in the inner accretion disk. Long-term campaign demonstrates that the BLR in NGC 5548 is a robust yet dynamically evolving entity, providing crucial insights into AGN structure and accretion physics.

astro-ph.GA↗

arXiv2025

Accretion-Regulated Type Transitions in Changing-Look AGNs: Evidence from Two-Epoch Spectral Analysis

The changing-look active galactic nucleus (CL-AGN), an extraordinary subpopulation of supermassive black holes, has attracted growing attention for understanding its nature. We present an analysis of the spectral properties of 203 low-redshift CL-AGNs ($z<0.35$) using two-epoch spectra from SDSS DR16 and DESI DR1 with time baseline ranging from $\sim$1000 to 8000 days, based on spectral fitting and decomposition. The sample consists of 11.3\% Type 1.0, 26.6\% Type 1.2, 43.1\% Type 1.5, and 19\% Type 1.8/2.0 AGNs. The total sample is divided into two datasets: Dataset A (110 objects) with minor spectral type variations, likely general AGN variability, and Dataset B (93 objects) showing significant type transitions and characteristic turn-on or turn-off behavior. Our results reveal clear optical continuum and emission-line variability, showing both bluer-when-brighter and redder-when-brighter trends. A strong correlation between the broad H$\beta$/[O~{\sc iii}] ratio and broad H$\alpha$ luminosity ($L_{\rm H\alpha}$), ${\rm log(H\beta/[O~III])}=(0.63\pm 0.07){\rm log}(L_{\rm H\alpha})-(26.49\pm2.96)\pm0.48$ for Dataset B, as well as the correlation between H$\beta$/[O~{\sc iii}] and Eddington ratio ($L_{\rm bol}/L_{\rm Edd}$), ${\rm log(H\beta/[O~III])}=(0.59\pm 0.08){\rm log}(L_{\rm bol}/L_{\rm Edd})+(1.02\pm0.15)\pm0.53$ for Dataset B, suggests that accretion rate variations drive changes in ionizing flux within the broad-line region, thereby triggering AGN type transitions. These findings underscore the critical role of supermassive black hole accretion processes in refining the AGN unification model. Future work should investigate potential connections between stellar evolution in outer accretion disk and the observed scatter in these correlations.

astro-ph.GA↗

arXiv2022

Pairwise Comparison Network for Remote Sensing Scene Classification

Remote sensing scene classification aims to assign a specific semantic label to a remote sensing image. Recently, convolutional neural networks have greatly improved the performance of remote sensing scene classification. However, some confused images may be easily recognized as the incorrect category, which generally degrade the performance. The differences between image pairs can be used to distinguish image categories. This paper proposed a pairwise comparison network, which contains two main steps: pairwise selection and pairwise representation. The proposed network first selects similar image pairs, and then represents the image pairs with pairwise representations. The self-representation is introduced to highlight the informative parts of each image itself, while the mutual-representation is proposed to capture the subtle differences between image pairs. Comprehensive experimental results on two challenging datasets (AID, NWPU-RESISC45) demonstrate the effectiveness of the proposed network. The codes are provided in https://github.com/spectralpublic/PCNet.git.

cs.CV↗

arXiv2020

Porous Lattice-based Transformer Encoder for Chinese NER

Incorporating lattices into character-level Chinese named entity recognition is an effective method to exploit explicit word information. Recent works extend recurrent and convolutional neural networks to model lattice inputs. However, due to the DAG structure or the variable-sized potential word set for lattice inputs, these models prevent the convenient use of batched computation, resulting in serious inefficient. In this paper, we propose a porous lattice-based transformer encoder for Chinese named entity recognition, which is capable to better exploit the GPU parallelism and batch the computation owing to the mask mechanism in transformer. We first investigate the lattice-aware self-attention coupled with relative position representations to explore effective word information in the lattice structure. Besides, to strengthen the local dependencies among neighboring tokens, we propose a novel porous structure during self-attentional computation processing, in which every two non-neighboring tokens are connected through a shared pivot node. Experimental results on four datasets show that our model performs up to 9.47 times faster than state-of-the-art models, while is roughly on a par with its performance. The source code of this paper can be obtained from https://github.com/xxx/xxx.

cs.CL↗

arXiv2019

Cross-Lingual Dependency Parsing Using Code-Mixed TreeBank

Treebank translation is a promising method for cross-lingual transfer of syntactic dependency knowledge. The basic idea is to map dependency arcs from a source treebank to its target translation according to word alignments. This method, however, can suffer from imperfect alignment between source and target words. To address this problem, we investigate syntactic transfer by code mixing, translating only confident words in a source treebank. Cross-lingual word embeddings are leveraged for transferring syntactic knowledge to the target from the resulting code-mixed treebank. Experiments on University Dependency Treebanks show that code-mixed treebanks are more effective than translated treebanks, giving highly competitive performances among cross-lingual parsing methods.

cs.CL↗

arXiv2019

Secure Software-Defined Networking Based on Blockchain

Software-Defined Networking (SDN) separates the network control plane and data plane, which provides a network-wide view with centralized control (in the control plane) and programmable network configuration for data plane injected by SDN applications (in the application plane). With these features, a number of drawbacks of the traditional network architectures such as static configuration, non-scalability and low efficiency can be effectively avoided. However, SDN also brings with it some new security challenges, such as single-point failure of the control plane, malicious flows from applications, exposed network-wide resources and a vulnerable channel between the control plane and the data plane. In this paper, we design a monolithic security mechanism for SDN based on Blockchain. Our mechanism decentralizes the control plane to overcome single-point failure while maintaining a network-wide view. The mechanism also guarantees the authenticity, traceability, and accountability of application flows, and hence secures the programmable configuration. Moreover, the mechanism provides a fine-grained access control of network-wide resources and a secure controller-switch channel to further protect resources and communication in SDN.

cs.CR↗