arXiv subjects

Jing Zhang

Publications and source records attributed to Jing Zhang.

At least 613 records · Page 34Linked to original sources

arXiv2023

MOEF: Modeling Occasion Evolution in Frequency Domain for Promotion-Aware Click-Through Rate Prediction

Promotions are becoming more important and prevalent in e-commerce to attract customers and boost sales, leading to frequent changes of occasions, which drives users to behave differently. In such situations, most existing Click-Through Rate (CTR) models can't generalize well to online serving due to distribution uncertainty of the upcoming occasion. In this paper, we propose a novel CTR model named MOEF for recommendations under frequent changes of occasions. Firstly, we design a time series that consists of occasion signals generated from the online business scenario. Since occasion signals are more discriminative in the frequency domain, we apply Fourier Transformation to sliding time windows upon the time series, obtaining a sequence of frequency spectrum which is then processed by Occasion Evolution Layer (OEL). In this way, a high-order occasion representation can be learned to handle the online distribution uncertainty. Moreover, we adopt multiple experts to learn feature representations from multiple aspects, which are guided by the occasion representation via an attention mechanism. Accordingly, a mixture of feature representations is obtained adaptively for different occasions to predict the final CTR. Experimental results on real-world datasets validate the superiority of MOEF and online A/B tests also show MOEF outperforms representative CTR models significantly.

cs.LG↗

arXiv2023

Dynamic Focus-aware Positional Queries for Semantic Segmentation

The DETR-like segmentors have underpinned the most recent breakthroughs in semantic segmentation, which end-to-end train a set of queries representing the class prototypes or target segments. Recently, masked attention is proposed to restrict each query to only attend to the foreground regions predicted by the preceding decoder block for easier optimization. Although promising, it relies on the learnable parameterized positional queries which tend to encode the dataset statistics, leading to inaccurate localization for distinct individual queries. In this paper, we propose a simple yet effective query design for semantic segmentation termed Dynamic Focus-aware Positional Queries (DFPQ), which dynamically generates positional queries conditioned on the cross-attention scores from the preceding decoder block and the positional encodings for the corresponding image features, simultaneously. Therefore, our DFPQ preserves rich localization information for the target segments and provides accurate and fine-grained positional priors. In addition, we propose to efficiently deal with high-resolution cross-attention by only aggregating the contextual tokens based on the low-resolution cross-attention scores to perform local relation aggregation. Extensive experiments on ADE20K and Cityscapes show that with the two modifications on Mask2former, our framework achieves SOTA performance and outperforms Mask2former by clear margins of 1.1%, 1.9%, and 1.1% single-scale mIoU with ResNet-50, Swin-T, and Swin-B backbones on the ADE20K validation set, respectively. Source code is available at https://github.com/ziplab/FASeg

cs.CV↗

arXiv2023

Model-free screening procedure for ultrahigh-dimensional survival data based on Hilbert-Schmidt independence criterion

How to select the active variables which have significant impact on the event of interest is a very important and meaningful problem in the statistical analysis of ultrahigh-dimensional data. Sure independent screening procedure has been demonstrated to be an effective method to reduce the dimensionality of data from a large scale to a relatively moderate scale. For censored survival data, the existing screening methods mainly adopt the Kaplan--Meier estimator to handle censoring, which may not perform well for scenarios which have heavy censoring rate. In this article, we propose a model-free screening procedure based on the Hilbert-Schmidt independence criterion (HSIC). The proposed method avoids the complication to specify an actual model from a large number of covariates. Compared with existing screening procedures, this new approach has several advantages. First, it does not involve the Kaplan--Meier estimator, thus its performance is much more robust for the cases with a heavy censoring rate. Second, the empirical estimate of HSIC is very simple as it just depends on the trace of a product of Gram matrices. In addition, the proposed procedure does not require any complicated numerical optimization, so the corresponding calculation is very simple and fast. Finally, the proposed procedure which employs the kernel method is substantially more resistant to outliers. Extensive simulation studies demonstrate that the proposed method has favorable exhibition over the existing methods. As an illustration, we apply the proposed method to analyze the diffuse large-B-cell lymphoma (DLBCL) data and the ovarian cancer data.

stat.ME↗

arXiv2023

Vision Transformer with Quadrangle Attention

Window-based attention has become a popular choice in vision transformers due to its superior performance, lower computational complexity, and less memory footprint. However, the design of hand-crafted windows, which is data-agnostic, constrains the flexibility of transformers to adapt to objects of varying sizes, shapes, and orientations. To address this issue, we propose a novel quadrangle attention (QA) method that extends the window-based attention to a general quadrangle formulation. Our method employs an end-to-end learnable quadrangle regression module that predicts a transformation matrix to transform default windows into target quadrangles for token sampling and attention calculation, enabling the network to model various targets with different shapes and orientations and capture rich context information. We integrate QA into plain and hierarchical vision transformers to create a new architecture named QFormer, which offers minor code modifications and negligible extra computational cost. Extensive experiments on public benchmarks demonstrate that QFormer outperforms existing representative vision transformers on various vision tasks, including classification, object detection, semantic segmentation, and pose estimation. The code will be made publicly available at \href{https://github.com/ViTAE-Transformer/QFormer}{QFormer}.

cs.CV↗

arXiv2023

LPFF: A Portrait Dataset for Face Generators Across Large Poses

The creation of 2D realistic facial images and 3D face shapes using generative networks has been a hot topic in recent years. Existing face generators exhibit exceptional performance on faces in small to medium poses (with respect to frontal faces) but struggle to produce realistic results for large poses. The distorted rendering results on large poses in 3D-aware generators further show that the generated 3D face shapes are far from the distribution of 3D faces in reality. We find that the above issues are caused by the training dataset's pose imbalance. In this paper, we present LPFF, a large-pose Flickr face dataset comprised of 19,590 high-quality real large-pose portrait images. We utilize our dataset to train a 2D face generator that can process large-pose face images, as well as a 3D-aware generator that can generate realistic human face geometry. To better validate our pose-conditional 3D-aware generators, we develop a new FID measure to evaluate the 3D-level performance. Through this novel FID measure and other experiments, we show that LPFF can help 2D face generators extend their latent space and better manipulate the large-pose data, and help 3D-aware face generators achieve better view consistency and more realistic 3D reconstruction results.

cs.CV↗

arXiv2023

Referring Image Matting

Different from conventional image matting, which either requires user-defined scribbles/trimap to extract a specific foreground object or directly extracts all the foreground objects in the image indiscriminately, we introduce a new task named Referring Image Matting (RIM) in this paper, which aims to extract the meticulous alpha matte of the specific object that best matches the given natural language description, thus enabling a more natural and simpler instruction for image matting. First, we establish a large-scale challenging dataset RefMatte by designing a comprehensive image composition and expression generation engine to automatically produce high-quality images along with diverse text attributes based on public datasets. RefMatte consists of 230 object categories, 47,500 images, 118,749 expression-region entities, and 474,996 expressions. Additionally, we construct a real-world test set with 100 high-resolution natural images and manually annotate complex phrases to evaluate the out-of-domain generalization abilities of RIM methods. Furthermore, we present a novel baseline method CLIPMat for RIM, including a context-embedded prompt, a text-driven semantic pop-up, and a multi-level details extractor. Extensive experiments on RefMatte in both keyword and expression settings validate the superiority of CLIPMat over representative methods. We hope this work could provide novel insights into image matting and encourage more follow-up studies. The dataset, code and models are available at https://github.com/JizhiziLi/RIM.

cs.CV↗

arXiv2023

A Survey on Class Imbalance in Federated Learning

Federated learning, which allows multiple client devices in a network to jointly train a machine learning model without direct exposure of clients' data, is an emerging distributed learning technique due to its nature of privacy preservation. However, it has been found that models trained with federated learning usually have worse performance than their counterparts trained in the standard centralized learning mode, especially when the training data is imbalanced. In the context of federated learning, data imbalance may occur either locally one one client device, or globally across many devices. The complexity of different types of data imbalance has posed challenges to the development of federated learning technique, especially considering the need of relieving data imbalance issue and preserving data privacy at the same time. Therefore, in the literature, many attempts have been made to handle class imbalance in federated learning. In this paper, we present a detailed review of recent advancements along this line. We first introduce various types of class imbalance in federated learning, after which we review existing methods for estimating the extent of class imbalance without the need of knowing the actual data to preserve data privacy. After that, we discuss existing methods for handling class imbalance in FL, where the advantages and disadvantages of the these approaches are discussed. We also summarize common evaluation metrics for class imbalanced tasks, and point out potential future directions.

cs.LG↗

arXiv2023

DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting

End-to-end text spotting aims to integrate scene text detection and recognition into a unified framework. Dealing with the relationship between the two sub-tasks plays a pivotal role in designing effective spotters. Although Transformer-based methods eliminate the heuristic post-processing, they still suffer from the synergy issue between the sub-tasks and low training efficiency. In this paper, we present DeepSolo, a simple DETR-like baseline that lets a single Decoder with Explicit Points Solo for text detection and recognition simultaneously. Technically, for each text instance, we represent the character sequence as ordered points and model them with learnable explicit point queries. After passing a single decoder, the point queries have encoded requisite text semantics and locations, thus can be further decoded to the center line, boundary, script, and confidence of text via very simple prediction heads in parallel. Besides, we also introduce a text-matching criterion to deliver more accurate supervisory signals, thus enabling more efficient training. Quantitative experiments on public benchmarks demonstrate that DeepSolo outperforms previous state-of-the-art methods and achieves better training efficiency. In addition, DeepSolo is also compatible with line annotations, which require much less annotation cost than polygons. The code is available at https://github.com/ViTAE-Transformer/DeepSolo.

cs.CV↗

arXiv2023

Dual Path Modeling for Semantic Matching by Perceiving Subtle Conflicts

Transformer-based pre-trained models have achieved great improvements in semantic matching. However, existing models still suffer from insufficient ability to capture subtle differences. The modification, addition and deletion of words in sentence pairs may make it difficult for the model to predict their relationship. To alleviate this problem, we propose a novel Dual Path Modeling Framework to enhance the model's ability to perceive subtle differences in sentence pairs by separately modeling affinity and difference semantics. Based on dual-path modeling framework we design the Dual Path Modeling Network (DPM-Net) to recognize semantic relations. And we conduct extensive experiments on 10 well-studied semantic matching and robustness test datasets, and the experimental results show that our proposed method achieves consistent improvements over baselines.

cs.CL↗

arXiv2023

Atomic Bose-Einstein condensate in a twisted-bilayer optical lattice

Observation of strong correlations and superconductivity in twisted-bilayer-graphene have stimulated tremendous interest in fundamental and applied physics. In this system, the superposition of two twisted honeycomb lattices, generating a Moir$\acute{\mathrm{e}}$ pattern, is the key to the observed flat electronic bands, slow electron velocity and large density of states. Despite these observations, a full understanding of the emerging superconductivity from the coupled insulating layers and the appearance of a small magic angle remain a hot topic of research. Here, we demonstrate a quantum simulation platform to study superfluids in twisted bilayer lattices based on Bose-Einstein condensates loaded into spin-dependent optical lattices. The lattices are made of two sets of laser beams that independently address atoms in different spin states, which form the synthetic dimension of the two layers. The twisted angle of the two lattices is controlled by the relative angle of the laser beams. We show that atoms in each spin state only feel one set of the lattice and the interlayer coupling can be controlled by microwave coupling between the spin states. Our system allows for flexible control of both the inter- and intralayer couplings. Furthermore we directly observe the spatial Moir$\acute{\mathrm{e}}$ pattern and the momentum diffraction, which confirm the presence of atomic superfluid in the bilayer lattices. Our system constitutes a powerful platform to investigate the physics underlying the superconductivity in twisted-bilayer-graphene and to explore other novel quantum phenomena difficult to realize in materials.

cond-mat.quant-gas↗

arXiv2023

Generation of large-scale continuous-variable cluster states multiplexed both in time and frequency domains

Large-scale continuous variable (CV) cluster state is necessary in quantum information processing based on measurement-based quantum computing (MBQC). Specially, generating large-scale CV cluster state multiplexed in time domain is easier to implement and has strong scalability in experiment. Here one-dimensional (1D) large-scale dual-rail CV cluster states multiplexed both in time and frequency domains are parallelly generated, which can be further extended to three-dimensional (3D) CV cluster state by combining two time-delay NOPA systems with beamsplitters. It is shown that the number of parallel arrays depends on the corresponding frequency comb lines and the partite number of each array can be very large (million), and scale of the 3D cluster state can be ultra-large. This scheme provides some special-structured CV cluster states, which will be valuable for quantum computing of hybrid domains.

quant-ph↗

arXiv2023

Transmission-Guided Bayesian Generative Model for Smoke Segmentation

Smoke segmentation is essential to precisely localize wildfire so that it can be extinguished in an early phase. Although deep neural networks have achieved promising results on image segmentation tasks, they are prone to be overconfident for smoke segmentation due to its non-rigid shape and transparent appearance. This is caused by both knowledge level uncertainty due to limited training data for accurate smoke segmentation and labeling level uncertainty representing the difficulty in labeling ground-truth. To effectively model the two types of uncertainty, we introduce a Bayesian generative model to simultaneously estimate the posterior distribution of model parameters and its predictions. Further, smoke images suffer from low contrast and ambiguity, inspired by physics-based image dehazing methods, we design a transmission-guided local coherence loss to guide the network to learn pair-wise relationships based on pixel distance and the transmission feature. To promote the development of this field, we also contribute a high-quality smoke segmentation dataset, SMOKE5K, consisting of 1,400 real and 4,000 synthetic images with pixel-wise annotation. Experimental results on benchmark testing datasets illustrate that our model achieves both accurate predictions and reliable uncertainty maps representing model ignorance about its prediction. Our code and dataset are publicly available at: https://github.com/redlessme/Transmission-BVM.

cs.CV↗

arXiv2023

GLM-Dialog: Noise-tolerant Pre-training for Knowledge-grounded Dialogue Generation

We present GLM-Dialog, a large-scale language model (LLM) with 10B parameters capable of knowledge-grounded conversation in Chinese using a search engine to access the Internet knowledge. GLM-Dialog offers a series of applicable techniques for exploiting various external knowledge including both helpful and noisy knowledge, enabling the creation of robust knowledge-grounded dialogue LLMs with limited proper datasets. To evaluate the GLM-Dialog more fairly, we also propose a novel evaluation method to allow humans to converse with multiple deployed bots simultaneously and compare their performance implicitly instead of explicitly rating using multidimensional metrics.Comprehensive evaluations from automatic to human perspective demonstrate the advantages of GLM-Dialog comparing with existing open source Chinese dialogue models. We release both the model checkpoint and source code, and also deploy it as a WeChat application to interact with users. We offer our evaluation platform online in an effort to prompt the development of open source models and reliable dialogue evaluation systems. The additional easy-to-use toolkit that consists of short text entity linking, query generation, and helpful knowledge classification is also released to enable diverse applications. All the source code is available on Github.

cs.CL↗

arXiv2023

Localizing Scan Targets from Human Pose for Autonomous Lung Ultrasound Imaging

Ultrasound is progressing toward becoming an affordable and versatile solution to medical imaging. With the advent of COVID-19 global pandemic, there is a need to fully automate ultrasound imaging as it requires trained operators in close proximity to patients for a long period of time, therefore increasing risk of infection. In this work, we investigate the important yet seldom-studied problem of scan target localization, under the setting of lung ultrasound imaging. We propose a purely vision-based, data driven method that incorporates learning-based computer vision techniques. We combine a human pose estimation model with a specially designed regression model to predict the lung ultrasound scan targets, and deploy multiview stereo vision to enhance the consistency of 3D target localization. While related works mostly focus on phantom experiments, we collect data from 30 human subjects for testing. Our method attains an accuracy level of 16.00(9.79) mm for probe positioning and 4.44(3.75) degree for probe orientation, with a success rate above 80% under an error threshold of 25mm for all scan targets. Moreover, our approach can serve as a general solution to other types of ultrasound modalities. The code for implementation has been released.

eess.IV↗

arXiv2023

Audio-Visual Segmentation

We propose to explore a new problem called audio-visual segmentation (AVS), in which the goal is to output a pixel-level map of the object(s) that produce sound at the time of the image frame. To facilitate this research, we construct the first audio-visual segmentation benchmark (AVSBench), providing pixel-wise annotations for the sounding objects in audible videos. Two settings are studied with this benchmark: 1) semi-supervised audio-visual segmentation with a single sound source and 2) fully-supervised audio-visual segmentation with multiple sound sources. To deal with the AVS problem, we propose a novel method that uses a temporal pixel-wise audio-visual interaction module to inject audio semantics as guidance for the visual segmentation process. We also design a regularization loss to encourage the audio-visual mapping during training. Quantitative and qualitative experiments on the AVSBench compare our approach to several existing methods from related tasks, demonstrating that the proposed method is promising for building a bridge between the audio and pixel-wise visual semantics. Code is available at https://github.com/OpenNLPLab/AVSBench.

cs.CV↗

arXiv2023

Feature Decomposition for Reducing Negative Transfer: A Novel Multi-task Learning Method for Recommender System

In recent years, thanks to the rapid development of deep learning (DL), DL-based multi-task learning (MTL) has made significant progress, and it has been successfully applied to recommendation systems (RS). However, in a recommender system, the correlations among the involved tasks are complex. Therefore, the existing MTL models designed for RS suffer from negative transfer to different degrees, which will injure optimization in MTL. We find that the root cause of negative transfer is feature redundancy that features learned for different tasks interfere with each other. To alleviate the issue of negative transfer, we propose a novel multi-task learning method termed Feature Decomposition Network (FDN). The key idea of the proposed FDN is reducing the phenomenon of feature redundancy by explicitly decomposing features into task-specific features and task-shared features with carefully designed constraints. We demonstrate the effectiveness of the proposed method on two datasets, a synthetic dataset and a public datasets (i.e., Ali-CCP). Experimental results show that our proposed FDN can outperform the state-of-the-art (SOTA) methods by a noticeable margin.

cs.IR↗

arXiv2023

Spectral conditions for forbidden subgraphs in bipartite graphs

A graph $G$ is $H$-free, if it contains no $H$ as a subgraph. A graph is said to be \emph{$H$-minor free}, if it does not contain $H$ as a minor. In recent years, Nikiforov asked that what is the maximum spectral radius of an $H$-free graph of order $n$? In this paper, we consider about some Brualdi-Solheid-Turán type problems on bipartite graphs. In 2015, Zhai, Lin and Gong proved that if $G$ is a bipartite graph with order $n \geq 2k+2$ and $ρ(G)\geq ρ(K_{k,n-k})$, then $G$ contains a $C_{2k+2}$ unless $G \cong K_{k,n-k}$ [Linear Algebra Appl. 471 (2015)]. Firstly, we give a new and more simple proof for the above theorem. Secondly, we prove that if $G$ is a bipartite graph with order $n \geq 2k+2$ and $ρ(G)\geq ρ(K_{k,n-k})$, then $G$ contains all $T_{2k+3}$ unless $G \cong K_{k,n-k}$. Finally, we prove that among all outerplanar bipartite graphs on $n>344569$ vertices, $K_{1,n-1}$ attains the maximum spectral radius.

math.CO↗

arXiv2023

Convergence of a quantum lattice Boltzmann scheme to the nonlinear Dirac equation for Gross-Neveu model in $1+1$ dimensions

This paper studies the quantum lattice Boltzmann scheme for the nonlinear Dirac equations for Gross-Neveu model in $1+1$ dimensions. The initial data for the scheme are assumed to be convergent in $L^2$. Then for any $T\ge 0$ the corresponding solutions for the quantum lattice Boltzmann scheme are shown to be convergent in $C([0,T];L^2(R^1))$ to the strong solution to the nonlinear Dirac equations as the mesh sizes converge to zero. In the proof, at first a Glimm type functional is introduced to establish the stability estimates for the difference between two solutions for the corresponding quantum lattice Boltzmann scheme, which leads to the compactness of the set of the solutions for the quantum lattice Boltzmann scheme. Finally, the limit of any convergent subsequence of the solutions for the quantum lattice Boltzmann scheme is shown to coincide with the strong solution to a Cauchy problem for the nonlinear Dirac equations.

math.AP↗