Searcharxiv⌕ Search

arXiv subjects

Jing Li

Publications and source records attributed to Jing Li.

At least 451 records · Page 25Linked to original sources
arXiv2023

Data-Augmented and Retrieval-Augmented Context Enrichment in Chinese Media Bias Detection

With the increasing pursuit of objective reports, automatically understanding media bias has drawn more attention in recent research. However, most of the previous work examines media bias from Western ideology, such as the left and right in the political spectrum, which is not applicable to Chinese outlets. Based on the previous lexical bias and informational bias structure, we refine it from the Chinese perspective and go one step further to craft data with 7 fine-grained labels. To be specific, we first construct a dataset with Chinese news reports about COVID-19 which is annotated by our newly designed system, and then conduct substantial experiments on it to detect media bias. However, the scale of the annotated data is not enough for the latest deep-learning technology, and the cost of human annotation in media bias, which needs a lot of professional knowledge, is too expensive. Thus, we explore some context enrichment methods to automatically improve these problems. In Data-Augmented Context Enrichment (DACE), we enlarge the training data; while in Retrieval-Augmented Context Enrichment (RACE), we improve information retrieval methods to select valuable information and integrate it into our models to better understand bias. Extensive experiments are conducted on both our dataset and an English dataset BASIL. Our results show that both methods outperform our baselines, while the RACE methods are more efficient and have more potential.

cs.CY↗
arXiv2023

VIBE: Topic-Driven Temporal Adaptation for Twitter Classification

Language features are evolving in real-world social media, resulting in the deteriorating performance of text classification in dynamics. To address this challenge, we study temporal adaptation, where models trained on past data are tested in the future. Most prior work focused on continued pretraining or knowledge updating, which may compromise their performance on noisy social media data. To tackle this issue, we reflect feature change via modeling latent topic evolution and propose a novel model, VIBE: Variational Information Bottleneck for Evolutions. Concretely, we first employ two Information Bottleneck (IB) regularizers to distinguish past and future topics. Then, the distinguished topics work as adaptive features via multi-task training with timestamp and class label prediction. In adaptive learning, VIBE utilizes retrieved unlabeled data from online streams created posterior to training data time. Substantial Twitter experiments on three classification tasks show that our model, with only 3% of data, significantly outperforms previous state-of-the-art continued-pretraining methods.

cs.CL↗
arXiv2023

Novel models for fatigue life prediction under wideband random loads based on machine learning

Machine learning as a data-driven solution has been widely applied in the field of fatigue lifetime prediction. In this paper, three models for wideband fatigue life prediction are built based on three machine learning models, i.e. support vector machine (SVM), Gaussian process regression (GPR) and artificial neural network (ANN). The generalization ability of the models is enhanced by employing numerous power spectra samples with different bandwidth parameters and a variety of material properties related to fatigue life. Sufficient Monte Carlo numerical simulations demonstrate that the newly developed machine learning models are superior to the traditional frequency-domain models in terms of life prediction accuracy and the ANN model has the best overall performance among the three developed machine learning models.

cond-mat.mtrl-sci↗
arXiv2023

TSP-Transformer: Task-Specific Prompts Boosted Transformer for Holistic Scene Understanding

Holistic scene understanding includes semantic segmentation, surface normal estimation, object boundary detection, depth estimation, etc. The key aspect of this problem is to learn representation effectively, as each subtask builds upon not only correlated but also distinct attributes. Inspired by visual-prompt tuning, we propose a Task-Specific Prompts Transformer, dubbed TSP-Transformer, for holistic scene understanding. It features a vanilla transformer in the early stage and tasks-specific prompts transformer encoder in the lateral stage, where tasks-specific prompts are augmented. By doing so, the transformer layer learns the generic information from the shared parts and is endowed with task-specific capacity. First, the tasks-specific prompts serve as induced priors for each task effectively. Moreover, the task-specific prompts can be seen as switches to favor task-specific representation learning for different tasks. Extensive experiments on NYUD-v2 and PASCAL-Context show that our method achieves state-of-the-art performance, validating the effectiveness of our method for holistic scene understanding. We also provide our code in the following link https://github.com/tb2-sy/TSP-Transformer.

cs.CV↗
arXiv2023

Proactively incremental-learning QAOA

Solving optimization problems with high performance is the target of existing works of Quantum Approximate Optimization Algorithm (QAOA). With this intention, we propose an advanced QAOA based on incremental learning, where the training trajectory is proactively segmented into incremental phases. Taking the MaxCut problem as our example, we randomly select a small subgraph from the whole graph and train the quantum circuit to get optimized parameters for the MaxCut of the subgraph in the first phase. Then in each subsequent incremental phase, a portion of the remaining nodes and edges are added to the current subgraph, and the circuit is retrained to get new optimized parameters. The above operation is repeated until the MaxCut problem on the whole graph is solved. The key point is that the optimized parameters of the previous phase will be reused in the initial parameters of the current phase. Numerous simulation experiments show our method has superior performance on Approximation Ratio (AR) and training time compared to prevalent works of QAOA. Specifically, the AR is higher than standard QAOA by 13.17% on weighted random graphs.

quant-ph↗
arXiv2023

Sanitized Clustering against Confounding Bias

Real-world datasets inevitably contain biases that arise from different sources or conditions during data collection. Consequently, such inconsistency itself acts as a confounding factor that disturbs the cluster analysis. Existing methods eliminate the biases by projecting data onto the orthogonal complement of the subspace expanded by the confounding factor before clustering. Therein, the interested clustering factor and the confounding factor are coarsely considered in the raw feature space, where the correlation between the data and the confounding factor is ideally assumed to be linear for convenient solutions. These approaches are thus limited in scope as the data in real applications is usually complex and non-linearly correlated with the confounding factor. This paper presents a new clustering framework named Sanitized Clustering Against confounding Bias (SCAB), which removes the confounding factor in the semantic latent space of complex data through a non-linear dependence measure. To be specific, we eliminate the bias information in the latent space by minimizing the mutual information between the confounding factor and the latent representation delivered by Variational Auto-Encoder (VAE). Meanwhile, a clustering module is introduced to cluster over the purified latent representations. Extensive experiments on complex datasets demonstrate that our SCAB achieves a significant gain in clustering performance by removing the confounding bias. The code is available at \url{https://github.com/EvaFlower/SCAB}.

cs.LG↗
arXiv2023

Graph Representation Learning for Infrared and Visible Image Fusion

Infrared and visible image fusion aims to extract complementary features to synthesize a single fused image. Many methods employ convolutional neural networks (CNNs) to extract local features due to its translation invariance and locality. However, CNNs fail to consider the image's non-local self-similarity (NLss), though it can expand the receptive field by pooling operations, it still inevitably leads to information loss. In addition, the transformer structure extracts long-range dependence by considering the correlativity among all image patches, leading to information redundancy of such transformer-based methods. However, graph representation is more flexible than grid (CNN) or sequence (transformer structure) representation to address irregular objects, and graph can also construct the relationships among the spatially repeatable details or texture with far-space distance. Therefore, to address the above issues, it is significant to convert images into the graph space and thus adopt graph convolutional networks (GCNs) to extract NLss. This is because the graph can provide a fine structure to aggregate features and propagate information across the nearest vertices without introducing redundant information. Concretely, we implement a cascaded NLss extraction pattern to extract NLss of intra- and inter-modal by exploring interactions of different image pixels in intra- and inter-image positional distance. We commence by preforming GCNs on each intra-modal to aggregate features and propagate information to extract independent intra-modal NLss. Then, GCNs are performed on the concatenate intra-modal NLss features of infrared and visible images, which can explore the cross-domain NLss of inter-modal to reconstruct the fused image. Ablation studies and extensive experiments illustrates the effectiveness and superiority of the proposed method on three datasets.

cs.CV↗
arXiv2023

Large-Time Behavior of the 2D Compressible Navier-Stokes System in Bounded Domains with Large Data and Vacuum

The large time behavior of the unique strong solution to the barotropic compressible Navier-Stokes system is studied with large external forces and initial data, where the shear viscosity is a positive constant and the bulk one is proportional to a power of the density. Some uniform estimates on the Lp-norm of the density are established, and then deduce that the density converges to its steady state in Lp-spaces, which transforms the large external force into a small one in some sense. Moreover, to deal with the obstacles brought by boundary, the conformal mapping and the pull back Green function are applied to give a point-wise representation of the effective viscous flux, and then make use of slip boundary conditions to cancel out the singularity.

math.AP↗
arXiv2023

The Wagging Plasma Tail of Comet C/2020 S3 (Erasmus)

Long-period comet C/2020 S3 (Erasmus) reached perihelion at 0.398 au on UT 2020 December 12.67, making it a bright, near-Sun object. Images taken between mid-November and December 2020 using the HI-1 camera and COR2 coronagraph onboard STEREO-A, as well as the LASCO/C3 coronagraph onboard SoHO, show significant variations in the plasma tail position angles. To analyze these variations, a simple technique was developed to calculate the aberration angles. These angles are defined as the angle between the sun-comet line and the tail axis, measured in the orbital plane. The aberration angles were found to range from $1.2^\circ$ to $46.8^\circ$, with an average (median) value of approximately $20.3^\circ$ ($16.3^\circ$). By considering the aberration angles, the solar wind radial velocities during the observations were inferred to range from 73.9 km/s to 573.5 km/s, with mean (median) values of approximately 205.5 km/s (182.3 km/s). Throughout the observations, two periods were identified where the tails showed forward tilting, which cannot be explained by aberration alone. In one case, this anomalous position angle was sustained for at least 11 days and is possibly due to co-rotating interaction regions. In the other case, the tail exhibited dramatic excursions from 180$^\circ$ to 150$^\circ$ back to 210$^\circ$ over a limited period of around 34 hours. This behavior is tentatively explained as a consequence of the interaction with a halo Coronal Mass Ejection that was launched from NOAA 12786 and arrived at comet C/2020 S3 during the time when the tail displayed its wagging behavior.

astro-ph.EP↗
arXiv2023

Mock Observations: Formation and Evolution of diffuse light in Galaxy Groups and Clusters in the IllustrisTNG Simulations

In this paper, by analyzing mock images from the IllustrisTNG100-1 simulation, we examine the properties of the diffuse light and compare them to those of central and satellite galaxies. Our findings suggest that the majority of the diffuse light originates from satellites. This claim is supported by the similarity between the age and metallicity distributions of the diffuse light and those of the satellites. Notably, the color distribution of the diffuse light gradually evolves to resemble that of the centrals at lower redshifts, suggesting a coevolution or passive process. The radial profiles of the diffuse light reveal distinct trends, with the inner regions displaying a relatively flat distribution and the outer regions showing a descending pattern. This finding suggests that the formation of the diffuse light is influenced by both major mergers and stellar tidal stripping. Moreover, strong correlations are found between the stellar mass of the diffuse light and the overall stellar mass of the satellites, as well as between the stellar mass of the diffuse light and the number of satellites within groups or clusters. These relationships can be described by power-law and logarithmic functions. Overall, the diffuse light components predominantly originate from satellites with intermediate ages and metallicities. These satellites typically fall within the stellar mass range of $\rm 8<\log_{10}M_{star}/M_{\odot}< 10$ and the color range of $\rm -1<[g-r]^{0.1}< 0$. As the redshift decreases, the growth of the diffuse light is primarily influenced by the redder satellites, while the most massive and reddest satellites have minimal roles in its growth.

astro-ph.GA↗
arXiv2023

Cloud-Magnetic Resonance Imaging System: In the Era of 6G and Artificial Intelligence

Magnetic Resonance Imaging (MRI) plays an important role in medical diagnosis, generating petabytes of image data annually in large hospitals. This voluminous data stream requires a significant amount of network bandwidth and extensive storage infrastructure. Additionally, local data processing demands substantial manpower and hardware investments. Data isolation across different healthcare institutions hinders cross-institutional collaboration in clinics and research. In this work, we anticipate an innovative MRI system and its four generations that integrate emerging distributed cloud computing, 6G bandwidth, edge computing, federated learning, and blockchain technology. This system is called Cloud-MRI, aiming at solving the problems of MRI data storage security, transmission speed, AI algorithm maintenance, hardware upgrading, and collaborative work. The workflow commences with the transformation of k-space raw data into the standardized Imaging Society for Magnetic Resonance in Medicine Raw Data (ISMRMRD) format. Then, the data are uploaded to the cloud or edge nodes for fast image reconstruction, neural network training, and automatic analysis. Then, the outcomes are seamlessly transmitted to clinics or research institutes for diagnosis and other services. The Cloud-MRI system will save the raw imaging data, reduce the risk of data loss, facilitate inter-institutional medical collaboration, and finally improve diagnostic accuracy and work efficiency.

eess.IV↗
arXiv2023

InfoDiffusion: Information Entropy Aware Diffusion Process for Non-Autoregressive Text Generation

Diffusion models have garnered considerable interest in the field of text generation. Several studies have explored text diffusion models with different structures and applied them to various tasks, including named entity recognition and summarization. However, there exists a notable disparity between the "easy-first" text generation process of current diffusion models and the "keyword-first" natural text generation process of humans, which has received limited attention. To bridge this gap, we propose InfoDiffusion, a non-autoregressive text diffusion model. Our approach introduces a "keyinfo-first" generation strategy and incorporates a noise schedule based on the amount of text information. In addition, InfoDiffusion combines self-conditioning with a newly proposed partially noising model structure. Experimental results show that InfoDiffusion outperforms the baseline model in terms of generation quality and diversity, as well as exhibiting higher sampling efficiency.

cs.CL↗
arXiv2023

RoomDesigner: Encoding Anchor-latents for Style-consistent and Shape-compatible Indoor Scene Generation

Indoor scene generation aims at creating shape-compatible, style-consistent furniture arrangements within a spatially reasonable layout. However, most existing approaches primarily focus on generating plausible furniture layouts without incorporating specific details related to individual furniture pieces. To address this limitation, we propose a two-stage model integrating shape priors into the indoor scene generation by encoding furniture as anchor latent representations. In the first stage, we employ discrete vector quantization to encode furniture pieces as anchor-latents. Based on the anchor-latents representation, the shape and location information of the furniture was characterized by a concatenation of location, size, orientation, class, and our anchor latent. In the second stage, we leverage a transformer model to predict indoor scenes autoregressively. Thanks to incorporating the proposed anchor-latents representations, our generative model produces shape-compatible and style-consistent furniture arrangements and synthesis furniture in diverse shapes. Furthermore, our method facilitates various human interaction applications, such as style-consistent scene completion, object mismatch correction, and controllable object-level editing. Experimental results on the 3D-Front dataset demonstrate that our approach can generate more consistent and compatible indoor scenes compared to existing methods, even without shape retrieval. Additionally, extensive ablation studies confirm the effectiveness of our design choices in the indoor scene generation model.

cs.CV↗
arXiv2023

Quantum control of classical motion: piston dynamics in a Rabi-coupled Bose-Einstein condensate

We explore the dynamics of a hybrid classical-quantum system consisting of a classical piston and a self-interacting pseudospin 1/2 Bose-Einstein condensate with a time-dependent Rabi coupling. We investigate the mechanical work produced by the piston moving as a result of the quantum pressure of the condensate. The time-dependent Rabi field redistributes the condensate density between the spin components and, as a result, causes a time dependent pressure acting on the piston. Correspondingly, the motion of the piston produces quantum evolution of the condensate mass- and spin density profiles. We show how by optimised design of the time-dependent direction of the Rabi field, one can control position and velocity of the piston.

quant-ph↗
arXiv2023

Magnetic Resonance Spectroscopy Quantification Aided by Deep Estimations of Imperfection Factors and Macromolecular Signal

Objective: Magnetic Resonance Spectroscopy (MRS) is an important technique for biomedical detection. However, it is challenging to accurately quantify metabolites with proton MRS due to serious overlaps of metabolite signals, imperfections because of non-ideal acquisition conditions, and interference with strong background signals mainly from macromolecules. The most popular method, LCModel, adopts complicated non-linear least square to quantify metabolites and addresses these problems by designing empirical priors such as basis-sets, imperfection factors. However, when the signal-to-noise ratio of MRS signal is low, the solution may have large deviation. Methods: Linear Least Squares (LLS) is integrated with deep learning to reduce the complexity of solving this overall quantification. First, a neural network is designed to explicitly predict the imperfection factors and the overall signal from macromolecules. Then, metabolite quantification is solved analytically with the introduced LLS. In our Quantification Network (QNet), LLS takes part in the backpropagation of network training, which allows the feedback of the quantification error into metabolite spectrum estimation. This scheme greatly improves the generalization to metabolite concentrations unseen for training compared to the end-to-end deep learning method. Results: Experiments show that compared with LCModel, the proposed QNet, has smaller quantification errors for simulated data, and presents more stable quantification for 20 healthy in vivo data at a wide range of signal-to-noise ratio. QNet also outperforms other end-to-end deep learning methods. Conclusion: This study provides an intelligent, reliable and robust MRS quantification. Significance: QNet is the first LLS quantification aided by deep learning.

physics.med-ph↗
arXiv2023

Detecting multipartite entanglement via complete orthogonal basis

We study genuine tripartite entanglement and multipartite entanglement in arbitrary $n$-partite quantum systems based on complete orthogonal basis (COB). While the usual Bloch representation of a density matrix uses three types of generators, the density matrix with COB operators has one uniformed type of generators which may simplify related computations. We take the advantage of this simplicity to derive useful and operational criteria to detect genuine tripartite entanglement and multipartite entanglement. We first convert the general states to simpler forms by using the relationship between general symmetric informationally complete measurements and COB. Then we derive an operational criteria to detect genuine tripartite entanglement. We study multipartite entanglement in arbitrary dimensional multipartite systems. By providing detailed examples, we demonstrate that our criteria can detect more genuine entangled and multipartite entangled states than the previously existing criteria.

quant-ph↗
arXiv2023

Label Supervised LLaMA Finetuning

The recent success of Large Language Models (LLMs) has gained significant attention in both academia and industry. Substantial efforts have been made to enhance the zero- and few-shot generalization capabilities of open-source LLMs through finetuning. Currently, the prevailing approach is instruction-tuning, which trains LLMs to complete real-world tasks by generating responses guided by natural language instructions. It is worth noticing that such an approach may underperform in sequence and token classification tasks. Unlike text generation tasks, classification tasks have a limited label space, where precise label prediction is more appreciated than generating diverse and human-like responses. Prior research has unveiled that instruction-tuned LLMs cannot outperform BERT, prompting us to explore the potential of leveraging latent representations from LLMs for supervised label prediction. In this paper, we introduce a label-supervised adaptation for LLMs, which aims to finetuning the model with discriminant labels. We evaluate this approach with Label Supervised LLaMA (LS-LLaMA), based on LLaMA-2-7B, a relatively small-scale LLM, and can be finetuned on a single GeForce RTX4090 GPU. We extract latent representations from the final LLaMA layer and project them into the label space to compute the cross-entropy loss. The model is finetuned by Low-Rank Adaptation (LoRA) to minimize this loss. Remarkably, without intricate prompt engineering or external knowledge, LS-LLaMA substantially outperforms LLMs ten times its size in scale and demonstrates consistent improvements compared to robust baselines like BERT-Large and RoBERTa-Large in text classification. Moreover, by removing the causal mask from decoders, LS-unLLaMA achieves the state-of-the-art performance in named entity recognition (NER). Our work will shed light on a novel approach to adapting LLMs for various downstream tasks.

cs.CL↗