arXiv subjects

Bin Yan

Publications and source records attributed to Bin Yan.

At least 73 records · Page 4Linked to original sources

arXiv2023

A Multimodal Visual Encoding Model Aided by Introducing Verbal Semantic Information

Biological research has revealed that the verbal semantic information in the brain cortex, as an additional source, participates in nonverbal semantic tasks, such as visual encoding. However, previous visual encoding models did not incorporate verbal semantic information, contradicting this biological finding. This paper proposes a multimodal visual information encoding network model based on stimulus images and associated textual information in response to this issue. Our visual information encoding network model takes stimulus images as input and leverages textual information generated by a text-image generation model as verbal semantic information. This approach injects new information into the visual encoding model. Subsequently, a Transformer network aligns image and text feature information, creating a multimodal feature space. A convolutional network then maps from this multimodal feature space to voxel space, constructing the multimodal visual information encoding network model. Experimental results demonstrate that the proposed multimodal visual information encoding network model outperforms previous models under the exact training cost. In voxel prediction of the left hemisphere of subject 1's brain, the performance improves by approximately 15.87%, while in the right hemisphere, the performance improves by about 4.6%. The multimodal visual encoding network model exhibits superior encoding performance. Additionally, ablation experiments indicate that our proposed model better simulates the brain's visual information processing.

cs.CV↗

arXiv2023

Probing the $Hgg$ coupling through the jet charge correlation in Higgs boson decay

The effective coupling of the Higgs boson to gluons determines the production rate of the Higgs boson at the LHC and plays a crucial role to measure the Higgs properties. In this paper, we propose a novel method to extract the Higgs gluon gauge coupling information at the future lepton colliders, by utilizing the jet charge asymmetry of the two leading jets from the Higgs boson decay. Owing to the jet charge correlation nature of the Higgs decay products, we demonstrate that this asymmetry value would be close to one for the signal $H\to gg$, while it is always larger than one for the dominated backgrounds. With the help of the jet charge information, it is possible to constrain the Higgs gluon effective coupling very well in the future Higgs factory and this conclusion is not sensitive to the jet charge parameter $κ$.

hep-ph↗

arXiv2023

Universal Instance Perception as Object Discovery and Retrieval

All instance perception tasks aim at finding certain objects specified by some queries such as category names, language expressions, and target annotations, but this complete field has been split into multiple independent subtasks. In this work, we present a universal instance perception model of the next generation, termed UNINEXT. UNINEXT reformulates diverse instance perception tasks into a unified object discovery and retrieval paradigm and can flexibly perceive different types of objects by simply changing the input prompts. This unified formulation brings the following benefits: (1) enormous data from different tasks and label vocabularies can be exploited for jointly training general instance-level representations, which is especially beneficial for tasks lacking in training data. (2) the unified model is parameter-efficient and can save redundant computation when handling multiple tasks simultaneously. UNINEXT shows superior performance on 20 challenging benchmarks from 10 instance-level tasks including classical image-level tasks (object detection and instance segmentation), vision-and-language tasks (referring expression comprehension and segmentation), and six video-level object tracking tasks. Code is available at https://github.com/MasterBin-IIAU/UNINEXT.

cs.CV↗

arXiv2023

Topologically protected Grover's oracle for the partition problem

The Number Partitioning Problem (NPP) is one of the NP-complete computational problems. Its definite exact solution generally requires a check of all $N$ solution candidates, which is exponentially large. Here we describe a path to the fast solution of this problem in $\sqrt{N}$ quasi-adiabatic quantum annealing steps. We argue that the errors due to the finite duration of the quantum annealing can be suppressed if the annealing time scales with $N$ only logarithmically. Moreover, our adiabatic oracle is topologically protected, in the sense that it is robust against small uncertainty and slow time-dependence of the physical parameters or the choice of the annealing protocol.

quant-ph↗

arXiv2023

Exploring Transformers for Open-world Instance Segmentation

Open-world instance segmentation is a rising task, which aims to segment all objects in the image by learning from a limited number of base-category objects. This task is challenging, as the number of unseen categories could be hundreds of times larger than that of seen categories. Recently, the DETR-like models have been extensively studied in the closed world while stay unexplored in the open world. In this paper, we utilize the Transformer for open-world instance segmentation and present SWORD. Firstly, we introduce to attach the stop-gradient operation before classification head and further add IoU heads for discovering novel objects. We demonstrate that a simple stop-gradient operation not only prevents the novel objects from being suppressed as background, but also allows the network to enjoy the merit of heuristic label assignment. Secondly, we propose a novel contrastive learning framework to enlarge the representations between objects and background. Specifically, we maintain a universal object queue to obtain the object center, and dynamically select positive and negative samples from the object queries for contrastive learning. While the previous works only focus on pursuing average recall and neglect average precision, we show the prominence of SWORD by giving consideration to both criteria. Our models achieve state-of-the-art performance in various open-world cross-category and cross-dataset generalizations. Particularly, in VOC to non-VOC setup, our method sets new state-of-the-art results of 40.0% on ARb100 and 34.9% on ARm100. For COCO to UVO generalization, SWORD significantly outperforms the previous best open-world model by 5.9% on APm and 8.1% on ARm100.

cs.CV↗

arXiv2023

Dynamic ensemble selection based on Deep Neural Network Uncertainty Estimation for Adversarial Robustness

The deep neural network has attained significant efficiency in image recognition. However, it has vulnerable recognition robustness under extensive data uncertainty in practical applications. The uncertainty is attributed to the inevitable ambient noise and, more importantly, the possible adversarial attack. Dynamic methods can effectively improve the defense initiative in the arms race of attack and defense of adversarial examples. Different from the previous dynamic method depend on input or decision, this work explore the dynamic attributes in model level through dynamic ensemble selection technology to further protect the model from white-box attacks and improve the robustness. Specifically, in training phase the Dirichlet distribution is apply as prior of sub-models' predictive distribution, and the diversity constraint in parameter space is introduced under the lightweight sub-models to construct alternative ensembel model spaces. In test phase, the certain sub-models are dynamically selected based on their rank of uncertainty value for the final prediction to ensure the majority accurate principle in ensemble robustness and accuracy. Compared with the previous dynamic method and staic adversarial traning model, the presented approach can achieve significant robustness results without damaging accuracy by combining dynamics and diversity property.

cs.LG↗

arXiv2023

Discriminating between Higgs Production Mechanisms via Jet Charge at the LHC

Discriminating the Higgs production mechanisms plays a crucial role in directly measuring the couplings of Higgs to gauge bosons for probing the nature of the electroweak symmetry breaking. We propose a novel method to distinguish the Higgs production mechanisms at the LHC by utilizing the jet charge asymmetry of the two leading forward jets in Higgs plus two jets production. This novel observable provides a way to disentangle the $W$-fusion from the $Z$-fusion and gluon fusion processes for the first time, due to the electric charge correlation of the two leading jets in the events. We show that the Higgs couplings to gauge bosons can be well constrained and its conclusion does not depend on the other possible new physics effects which modify the Higgs total or partial width. We also discuss the complementary roles between the proposed jet charge asymmetry measurement and the Higgs signal strength measurements at the HL-LHC in determining the Higgs couplings.

hep-ph↗

arXiv2023

Probing the $Zb\bar{b}$ coupling at the $Z$-pole of future lepton colliders

The determination of the $Zb\bar{b}$ coupling in experiments has been a long-standing challenge, as the limited precision of off $Z$-pole measurements at the LEP has resulted in two degenerate solutions remained to be resolved. In this paper, we propose a novel method to probe the $Zb\bar{b}$ coupling by measuring the forward-backward asymmetry of the bottom quark, $A^{b}_\text{FB}$, in the $b\bar{b}$ system of the $e^+ e^- \to b \bar{b} γ$ and/or $e^+ e^- \to b \bar{b} g$ processes at the $Z$-pole of future lepton colliders. The additional hard photon or light jet radiation can mimic the energy scanning of the $e^+e^-\to b\bar{b}$ process, and the $A_{\rm FB}^b$ distribution from the $γ$-$Z$ interference process is linearly sensitive to the $Zb\bar{b}$ coupling. By combining the expected measurements of $R^0_b$ and $A^{0,b}_\text{FB}$ at the $Z$-pole at the CEPC, the $A^{b}_\text{FB}$ distributions can break the degeneracy observed at the LEP, leading to a unique determination of the $Zb\bar{b}$ coupling through $Z$-pole running alone.

hep-ph↗

arXiv2023

Muti-scale Graph Neural Network with Signed-attention for Social Bot Detection: A Frequency Perspective

The presence of a large number of bots on social media has adverse effects. The graph neural network (GNN) can effectively leverage the social relationships between users and achieve excellent results in detecting bots. Recently, more and more GNN-based methods have been proposed for bot detection. However, the existing GNN-based bot detection methods only focus on low-frequency information and seldom consider high-frequency information, which limits the representation ability of the model. To address this issue, this paper proposes a Multi-scale with Signed-attention Graph Filter for social bot detection called MSGS. MSGS could effectively utilize both high and low-frequency information in the social graph. Specifically, MSGS utilizes a multi-scale structure to produce representation vectors at different scales. These representations are then combined using a signed-attention mechanism. Finally, multi-scale representations via MLP after polymerization to produce the final result. We analyze the frequency response and demonstrate that MSGS is a more flexible and expressive adaptive graph filter. MSGS can effectively utilize high-frequency information to alleviate the over-smoothing problem of deep GNNs. Experimental results on real-world datasets demonstrate that our method achieves better performance compared with several state-of-the-art social bot detection methods.

cs.CV↗

arXiv2023

RF-GNN: Random Forest Boosted Graph Neural Network for Social Bot Detection

The presence of a large number of bots on social media leads to adverse effects. Although Random forest algorithm is widely used in bot detection and can significantly enhance the performance of weak classifiers, it cannot utilize the interaction between accounts. This paper proposes a Random Forest boosted Graph Neural Network for social bot detection, called RF-GNN, which employs graph neural networks (GNNs) as the base classifiers to construct a random forest, effectively combining the advantages of ensemble learning and GNNs to improve the accuracy and robustness of the model. Specifically, different subgraphs are constructed as different training sets through node sampling, feature selection, and edge dropout. Then, GNN base classifiers are trained using various subgraphs, and the remaining features are used for training Fully Connected Netural Network (FCN). The outputs of GNN and FCN are aligned in each branch. Finally, the outputs of all branches are aggregated to produce the final result. Moreover, RF-GNN is compatible with various widely-used GNNs for node classification. Extensive experimental results demonstrate that the proposed method obtains better performance than other state-of-the-art methods.

cs.LG↗

arXiv2023

MGTAB: A Multi-Relational Graph-Based Twitter Account Detection Benchmark

The development of social media user stance detection and bot detection methods rely heavily on large-scale and high-quality benchmarks. However, in addition to low annotation quality, existing benchmarks generally have incomplete user relationships, suppressing graph-based account detection research. To address these issues, we propose a Multi-Relational Graph-Based Twitter Account Detection Benchmark (MGTAB), the first standardized graph-based benchmark for account detection. To our knowledge, MGTAB was built based on the largest original data in the field, with over 1.55 million users and 130 million tweets. MGTAB contains 10,199 expert-annotated users and 7 types of relationships, ensuring high-quality annotation and diversified relations. In MGTAB, we extracted the 20 user property features with the greatest information gain and user tweet features as the user features. In addition, we performed a thorough evaluation of MGTAB and other public datasets. Our experiments found that graph-based approaches are generally more effective than feature-based approaches and perform better when introducing multiple relations. By analyzing experiment results, we identify effective approaches for account detection and provide potential future research directions in this field. Our benchmark and standardized evaluation procedures are freely available at: https://github.com/GraphDetec/MGTAB.

cs.CV↗

arXiv2023

Searching for the axion-like particle at the EIC

The axion-like particle (ALP) is a well motivated new particle candidate of beyond the Standard Model. In this work, we propose to probe the ALP through the photon fusion scattering at the upcoming Electron-Ion Collider (EIC) with electron and proton energy $E_e=20~{\rm GeV}$ and $E_p=250~{\rm GeV}$. It shows that we could constrain the effective coupling strength between ALP and photons to be $0.2~{\rm TeV}^{-1}$ at $2σ$ confidence level with the integrated luminosity of $300~{\rm fb}^{-1}$ for the mass range $m_a\in [5,40]~{\rm GeV}$. Such bound could be much improved if we consider the nucleus beam at the EIC. We also demonstrate that the limits from the EIC could be stronger than the off $Z$-pole measurement at the LEP and the Light-by-Light scattering with pp collision at the LHC.

hep-ph↗

arXiv2023

An adiabatic oracle for Grover's algorithm

Grover's search algorithm was originally proposed for circuit-based quantum computers. A crucial part of it is to query an oracle -- a black-box unitary operation. Generation of this oracle is formally beyond the original algorithm design. Here, we propose a realization of Grover's oracle for a large class of searching problems using a quantum annealing step. The time of this step grows only logarithmically with the size of the database. This suggests an efficient path to application of Grover's algorithm to practically important problems, such as finding the ground state of a Hamiltonian with a spectral gap over its ground state.

quant-ph↗

arXiv2022

High-Performance Transformer Tracking

Correlation has a critical role in the tracking field, especially in recent popular Siamese-based trackers. The correlation operation is a simple fusion method that considers the similarity between the template and the search region. However, the correlation operation is a local linear matching process, losing semantic information and easily falling into a local optimum, which may be the bottleneck in designing high-accuracy tracking algorithms. In this work, to determine whether a better feature fusion method exists than correlation, a novel attention-based feature fusion network, inspired by the transformer, is presented. This network effectively combines the template and search region features using attention. Specifically, the proposed method includes an ego-context augment module based on self-attention and a cross-feature augment module based on cross-attention. First, we present a transformer tracking (named TransT) method based on the Siamese-like feature extraction backbone, the designed attention-based fusion mechanism, and the classification and regression head. Based on the TransT baseline, we further design a segmentation branch to generate an accurate mask. Finally, we propose a stronger version of TransT by extending TransT with a multi-template scheme and an IoU prediction head, named TransT-M. Experiments show that our TransT and TransT-M methods achieve promising results on seven popular datasets. Code and models are available at https://github.com/chenxin-dlut/TransT-M.

cs.CV↗

arXiv2022

Decoherence factor as a convolution: an interplay between a Gaussian and an exponential coherence loss

This paper identifies and investigates nature of the transition between Gaussian and exponential forms of decoherence. We show that the decoherence factor (that controls the time dependence of the suppression of the off-diagonal terms when the density matrix is expressed in the pointer basis representation) can be described by the convolution of Gaussian and exponential functions, their contributions modulated by the strength of the system-environment interaction. In the strong and weak coupling limits, decoherence reduces to the familiar Gaussian and exponential forms, respectively. The mechanism is demonstrated with two paradigmatic examples of decoherence -- a spin-bath model and the quantum Brownian motion.

quant-ph↗

arXiv2022

Probing the $Hγγ$ coupling via Higgs exclusive decay into quarkonia plus a photon at the HL-LHC

It is well know that the precise measurement of $H\toγγ$ can generate two degenerate parameter spaces of $Hγγ$ anomalous coupling. We propose to utilize the exclusive Higgs rare decays $H\toΥ(ns)+γ$ to break above mentioned degeneracy and further to constrain the $Hγγ$ anomalous coupling at the HL-LHC. We demonstrate that the branching ratios of $H\toΥ(ns)+γ$ can be significantly enhanced in the non-SM-like parameter space from $H\toγγ$ measurement, due to the destructive interference between the direct and indirect production of $Υ(ns)$ in the SM. By applying the NRQCD factorization formalism, we calculate the partial decay widths of $H\toΥ(ns)+γ$ at the NLO accuracy of $α_s$. We show that it is hopeful to break such degenerate parameter space at the HL-LHC if we can further highly suppress the background and enhance the signal efficiency compared to the ATLAS preliminary simulation.

hep-ph↗

arXiv2022

Randomized channel-state duality

Channel-state duality is a central result in quantum information science. It refers to the correspondence between a dynamical process (quantum channel) and a static quantum state in an enlarged Hilbert space. Since the corresponding dual state is generally mixed, it is described by a Hermitian matrix. In this article, we present a randomized channel-state duality. In other words, a quantum channel is represented by a collection of $N$ pure quantum states that are produced from a random source. The accuracy of this randomized duality relation is given by $1/N$, with regard to an appropriate distance measure. For large systems, $N$ is much smaller than the dimension of the exact dual matrix of the quantum channel. This provides a highly accurate low-rank approximation of any quantum channel, and, as a consequence of the duality relation, an efficient data compression scheme for mixed quantum states. We demonstrate these two immediate applications of the randomized channel-state duality with a chaotic $1$-dimensional spin system.

quant-ph↗

arXiv2022

Dynamic Stochastic Ensemble with Adversarial Robust Lottery Ticket Subnetworks

Adversarial attacks are considered the intrinsic vulnerability of CNNs. Defense strategies designed for attacks have been stuck in the adversarial attack-defense arms race, reflecting the imbalance between attack and defense. Dynamic Defense Framework (DDF) recently changed the passive safety status quo based on the stochastic ensemble model. The diversity of subnetworks, an essential concern in the DDF, can be effectively evaluated by the adversarial transferability between different networks. Inspired by the poor adversarial transferability between subnetworks of scratch tickets with various remaining ratios, we propose a method to realize the dynamic stochastic ensemble defense strategy. We discover the adversarial transferable diversity between robust lottery ticket subnetworks drawn from different basic structures and sparsity. The experimental results suggest that our method achieves better robust and clean recognition accuracy by adversarial transferable diversity, which would decrease the reliability of attacks.

cs.CV↗