arXiv subjects

Xu Zhang

Publications and source records attributed to Xu Zhang.

At least 199 records · Page 11Linked to original sources

arXiv2025

Submillimeter and Mid-Infrared Variability of Young Stellar Objects in the M17 HII Region

We conducted a comprehensive analysis of young stellar object (YSO) variability at submillimeter and mid-infrared (mid-IR) wavelengths for the M\,17 \ion{H}{2} region, using 3.5 years monitoring data from the JCMT Transient Survey at $450$ and $850\,μ$m and 9 years mid-IR monitoring data from the NEOWISE mission. Our study encompasses observations of 198 and 164 bright submillimeter peaks identified within the deep JCMT coadded maps at 450 and $850\,μ$m, and 66 YSOs seen by NEOWISE W2 that were previously identified in mid-IR observations. We find one robust linear submillimeter variable, an intermediate mass protostar, with a $4\%$ peak flux change in 3.5 years of JCMT monitoring that sets a lower limit of $16\%$ luminosity increase for the source. At mid-IR wavelengths, our analysis reveals secular and stochastic variability in 22 YSOs, with the highest fraction of secular variability occurring at the earliest evolutionary stage. This mid-IR fractional variability as a function of evolutionary stage result is similar to what has previously been found for YSO variability within the Gould Belt and the intermediate-mass star formation region M17\,SWex, though overall less variability is detected in M\,17 in submillimeter and mid-IR. We suspect that this lower detection of YSO variability is due to both the greater distance to M\,17 and the strong feedback from the \ion{H}{2} region. Our findings showcase the utility of multiwavelength observations to better capture the complex variability phenomena inherent to star formation processes and demonstrate the importance of years-long monitoring of a diverse selection of star-forming environments.

astro-ph.GA↗

arXiv2025

Reconfigurable Ultrafast Thermal Metamaterial Pixel Arrays by Dual-Gate Graphene Transistors

Thermal signatures represent ubiquitous infrared appearances of objects, carrying their unique spectral fingerprints. Despite extensive efforts to decipher and manipulate thermal-infrared signals, the ability to fully control them across spatial, temporal and spectral domains remains a significant challenge due to the slow speed, diffuse and broadband emitting nature of thermal emission in most materials. Here, we demonstrate a reconfigurable ultrafast thermal metamaterial pixel array that integrates active metasurfaces with dual-gate graphene transistors (Gr-FETs). The Gr-FETs with dual-gate control in each pixel achieve the heater-switch dual functionalities. As broadband transparent microheaters, Gr-FETs support the arbitrary design of integrated metasurfaces to achieve multi-color, narrowband infrared emission and operate at ultrafast modulation speeds. Concurrently as electrical switches, they enable a unified control scheme for pixel arrays of various sizes over large areas without compromising emission intensity. By decoupling the thermal generation and emission design processes, our approach provides an unprecedented degree of flexibility in programming thermal output across space, time, and wavelength. Our fabricated thermal pixel array experimentally demonstrated 26 alphabetical letters by applying progressive scanning, thus paving the way for practical realization of universal thermal signature controls for advanced thermal-infrared applications.

physics.optics↗

arXiv2025

Cryogenic in-memory computing using magnetic topological insulators

Machine learning algorithms have been proven effective for essential quantum computation tasks such as quantum error correction and quantum control. Efficient hardware implementation of these algorithms at cryogenic temperatures is essential. Here, we utilize magnetic topological insulators as memristors (termed magnetic topological memristors) and introduce a cryogenic in-memory computing scheme based on the coexistence of the chiral edge state and the topological surface state. The memristive switching and reading of the giant anomalous Hall effect exhibit high energy efficiency, high stability, and low stochasticity. We achieve high accuracy in a proof-of-concept classification task using four magnetic topological memristors. Furthermore, our algorithm-level and circuit-level simulations of large-scale neural networks demonstrate software-level accuracy and lower energy consumption for image recognition and quantum state preparation compared with existing magnetic memristor and CMOS technologies. Our results not only showcase a new application of chiral edge states but also may inspire further topological quantum physics-based novel computing schemes.

cond-mat.mes-hall↗

arXiv2025

Generalization Performance of Ensemble Clustering: From Theory to Algorithm

Ensemble clustering has demonstrated great success in practice; however, its theoretical foundations remain underexplored. This paper examines the generalization performance of ensemble clustering, focusing on generalization error, excess risk and consistency. We derive a convergence rate of generalization error bound and excess risk bound both of $\mathcal{O}(\sqrt{\frac{\log n}{m}}+\frac{1}{\sqrt{n}})$, with $n$ and $m$ being the numbers of samples and base clusterings. Based on this, we prove that when $m$ and $n$ approach infinity and $m$ is significantly larger than log $n$, i.e., $m,n\to \infty, m\gg \log n$, ensemble clustering is consistent. Furthermore, recognizing that $n$ and $m$ are finite in practice, the generalization error cannot be reduced to zero. Thus, by assigning varying weights to finite clusterings, we minimize the error between the empirical average clusterings and their expectation. From this, we theoretically demonstrate that to achieve better clustering performance, we should minimize the deviation (bias) of base clustering from its expectation and maximize the differences (diversity) among various base clusterings. Additionally, we derive that maximizing diversity is nearly equivalent to a robust (min-max) optimization model. Finally, we instantiate our theory to develop a new ensemble clustering algorithm. Compared with SOTA methods, our approach achieves average improvements of 6.1%, 7.3%, and 6.0% on 10 datasets w.r.t. NMI, ARI, and Purity. The code is available at https://github.com/xuz2019/GPEC.

cs.LG↗

arXiv2025

Optimizing Robustness and Accuracy in Mixture of Experts: A Dual-Model Approach

Mixture of Experts (MoE) have shown remarkable success in leveraging specialized expert networks for complex machine learning tasks. However, their susceptibility to adversarial attacks presents a critical challenge for deployment in robust applications. This paper addresses the critical question of how to incorporate robustness into MoEs while maintaining high natural accuracy. We begin by analyzing the vulnerability of MoE components, finding that expert networks are notably more susceptible to adversarial attacks than the router. Based on this insight, we propose a targeted robust training technique that integrates a novel loss function to enhance the adversarial robustness of MoE, requiring only the robustification of one additional expert without compromising training or inference efficiency. Building on this, we introduce a dual-model strategy that linearly combines a standard MoE model with our robustified MoE model using a smoothing parameter. This approach allows for flexible control over the robustness-accuracy trade-off. We further provide theoretical foundations by deriving certified robustness bounds for both the single MoE and the dual-model. To push the boundaries of robustness and accuracy, we propose a novel joint training strategy JTDMoE for the dual-model. This joint training enhances both robustness and accuracy beyond what is achievable with separate models. Experimental results on CIFAR-10 and TinyImageNet datasets using ResNet18 and Vision Transformer (ViT) architectures demonstrate the effectiveness of our proposed methods. The code is publicly available at https://github.com/TIML-Group/Robust-MoE-Dual-Model.

cs.LG↗

arXiv2025

A General Knowledge Injection Framework for ICD Coding

ICD Coding aims to assign a wide range of medical codes to a medical text document, which is a popular and challenging task in the healthcare domain. To alleviate the problems of long-tail distribution and the lack of annotations of code-specific evidence, many previous works have proposed incorporating code knowledge to improve coding performance. However, existing methods often focus on a single type of knowledge and design specialized modules that are complex and incompatible with each other, thereby limiting their scalability and effectiveness. To address this issue, we propose GKI-ICD, a novel, general knowledge injection framework that integrates three key types of knowledge, namely ICD Description, ICD Synonym, and ICD Hierarchy, without specialized design of additional modules. The comprehensive utilization of the above knowledge, which exhibits both differences and complementarity, can effectively enhance the ICD coding performance. Extensive experiments on existing popular ICD coding benchmarks demonstrate the effectiveness of GKI-ICD, which achieves the state-of-the-art performance on most evaluation metrics. Code is available at https://github.com/xuzhang0112/GKI-ICD.

cs.CL↗

arXiv2025

RSTeller: Scaling Up Visual Language Modeling in Remote Sensing with Rich Linguistic Semantics from Openly Available Data and Large Language Models

Abundant, well-annotated multimodal data in remote sensing are pivotal for aligning complex visual remote sensing (RS) scenes with human language, enabling the development of specialized vision language models across diverse RS interpretation tasks. However, annotating RS images with rich linguistic semantics at scale demands expertise in RS and substantial human labor, making it costly and often impractical. In this study, we propose a workflow that leverages large language models (LLMs) to generate multimodal datasets with semantically rich captions at scale from plain OpenStreetMap (OSM) data for images sourced from the Google Earth Engine (GEE) platform. This approach facilitates the generation of paired remote sensing data and can be readily scaled up using openly available data. Within this framework, we present RSTeller, a multimodal dataset comprising over 1.3 million RS images, each accompanied by two descriptive captions. Extensive experiments demonstrate that RSTeller enhances the performance of multiple existing vision language models for RS scene understanding through continual pre-training. Our methodology significantly reduces the manual effort and expertise needed for annotating remote sensing imagery while democratizing access to high-quality annotated data. This advancement fosters progress in visual language modeling and encourages broader participation in remote sensing research and applications. The RSTeller dataset is available at https://github.com/SlytherinGe/RSTeller.

cs.CV↗

arXiv2025

Ultra Lowrate Image Compression with Semantic Residual Coding and Compression-aware Diffusion

Existing multimodal large model-based image compression frameworks often rely on a fragmented integration of semantic retrieval, latent compression, and generative models, resulting in suboptimal performance in both reconstruction fidelity and coding efficiency. To address these challenges, we propose a residual-guided ultra lowrate image compression named ResULIC, which incorporates residual signals into both semantic retrieval and the diffusion-based generation process. Specifically, we introduce Semantic Residual Coding (SRC) to capture the semantic disparity between the original image and its compressed latent representation. A perceptual fidelity optimizer is further applied for superior reconstruction quality. Additionally, we present the Compression-aware Diffusion Model (CDM), which establishes an optimal alignment between bitrates and diffusion time steps, improving compression-reconstruction synergy. Extensive experiments demonstrate the effectiveness of ResULIC, achieving superior objective and subjective performance compared to state-of-the-art diffusion-based methods with - 80.7%, -66.3% BD-rate saving in terms of LPIPS and FID. Project page is available at https: //njuvision.github.io/ResULIC/.

cs.CV↗

arXiv2025

Angle-Tuned Gross-Neveu Quantum Criticality in Twisted Bilayer Graphene: A Quantum Monte Carlo Study

The fascinating quantum many-body states in twisted bilayber graphene (TBG) at magic angle, due to the interplay of Coulomb interactions and the quantum metrics of flat bands, have been well understood both experimentally and theoretically. However, the phase diagram and excitations as functions of twist angle and permittivity are still largely unknown. Here, via a newly developed momentum-space continuous-field quantum Monte Carlo method fully taking into account long-ranged Coulomb interactions and flat bands' quantum metrics with system sizes that were not accessible before, we show that charge-neutral TBG realizes an angle-tuned quantum phase transition from a gapped Kramers intervalley coherence (KIVC) state to a Dirac semimetal with critical angles around 1.2$°$. In single-particle spectra we demonstrate the evolution of a minimum gap at $Γ$ at the magic angle and towards touching points at Brillouin zone corners as the angle increases. The free energy and KIVC order parameter show that the transition belongs to fermionic Gross-Neveu criticality, and is robust upon varying the permittivity or the interlayer hopping.

cond-mat.str-el↗

arXiv2025

Perceive-IR: Learning to Perceive Degradation Better for All-in-One Image Restoration

Existing All-in-One image restoration methods often fail to perceive degradation types and severity levels simultaneously, overlooking the importance of fine-grained quality perception. Moreover, these methods often utilize highly customized backbones, which hinder their adaptability and integration into more advanced restoration networks. To address these limitations, we propose Perceive-IR, a novel backbone-agnostic All-in-One image restoration framework designed for fine-grained quality control across various degradation types and severity levels. Its modular structure allows core components to function independently of specific backbones, enabling seamless integration into advanced restoration models without significant modifications. Specifically, Perceive-IR operates in two key stages: 1) multi-level quality-driven prompt learning stage, where a fine-grained quality perceiver is meticulously trained to discern three tier quality levels by optimizing the alignment between prompts and images within the CLIP perception space. This stage ensures a nuanced understanding of image quality, laying the groundwork for subsequent restoration; 2) restoration stage, where the quality perceiver is seamlessly integrated with a difficulty-adaptive perceptual loss, forming a quality-aware learning strategy. This strategy not only dynamically differentiates sample learning difficulty but also achieves fine-grained quality control by driving the restored image toward the ground truth while pulling it away from both low- and medium-quality samples.

cs.CV↗

arXiv2025

Attosecond-Resolved Photoionization Dynamics and Interference-Enhanced Photoelectron Circular Dichroism in Chiral Molecules

Chiral molecules exhibit enantiosensitive light-matter interactions, with photoelectron circular dichroism (PECD) serving as a sensitive probe of molecular chirality through the asymmetry in the photoelectron wavepacket amplitude. Here, we demonstrate a photoelectron interferometric approach to access the phase of the photoelectron wavepacket and uncover attosecond dynamics in chiral molecule photoionization. Using circularly polarized attosecond XUV pulse trains synchronized with IR fields, we reveal distinct time delays between forward- and backward-ejected photoelectrons in a randomly oriented ensemble of chiral molecules. Moreover, we predict a pronounced enhancement of PECD due to the interference of the two photoionization pathways. The forward-backward time delay difference and the PECD are more prominent when the IR field counter-rotates with the XUV field. These results imply the counter-rotating IR field is more efficient in generating odd-parity photoelectron wavepackets in continuum-continuum transitions, highlighting the critical role of long-range chiral potential. Our work demonstrates a way of coherent control over the chiral photoelectron wavepackets, providing a route to enhance chiral signals and manipulate ultrafast chiral dynamics on attosecond time scales.

physics.atom-ph↗

arXiv2025

An Enriched Immersed Finite Element Method for 3D Interface Problems

We introduce an enriched immersed finite element method for addressing interface problems characterized by general non-homogeneous jump conditions. Unlike many existing unfitted mesh methods, our approach incorporates a homogenization concept. The IFE trial function set is composed of two components: the standard homogeneous IFE space and additional enrichment IFE functions. These enrichment functions are directly determined by the jump data, without adding extra degrees of freedom to the system. Meanwhile, the homogeneous IFE space is isomorphic to the standard finite element space on the same mesh. This isomorphism remains stable regardless of interface location relative to the mesh, ensuring optimal $\mathcal{O}(h^2)$ conditioning that is independent of the interface location and facilitates an immediate development of a multigrid fast solver; namely the iteration numbers are independent of not only the mesh size but also the relative interface location. Theoretical analysis and extensive numerical experiments are carried out in the efforts to demonstrate these features.

math.NA↗

arXiv2025

Unveiling the Underwater World: CLIP Perception Model-Guided Underwater Image Enhancement

High-quality underwater images are essential for both machine vision tasks and viewers with their aesthetic appeal.However, the quality of underwater images is severely affected by light absorption and scattering. Deep learning-based methods for Underwater Image Enhancement (UIE) have achieved good performance. However, these methods often overlook considering human perception and lack sufficient constraints within the solution space. Consequently, the enhanced images often suffer from diminished perceptual quality or poor content restoration.To address these issues, we propose a UIE method with a Contrastive Language-Image Pre-Training (CLIP) perception loss module and curriculum contrastive regularization. Above all, to develop a perception model for underwater images that more aligns with human visual perception, the visual semantic feature extraction capability of the CLIP model is leveraged to learn an appropriate prompt pair to map and evaluate the quality of underwater images. This CLIP perception model is then incorporated as a perception loss module into the enhancement network to improve the perceptual quality of enhanced images. Furthermore, the CLIP perception model is integrated with the curriculum contrastive regularization to enhance the constraints imposed on the enhanced images within the CLIP perceptual space, mitigating the risk of both under-enhancement and over-enhancement. Specifically, the CLIP perception model is employed to assess and categorize the learning difficulty level of negatives in the regularization process, ensuring comprehensive and nuanced utilization of distorted images and negatives with varied quality levels. Extensive experiments demonstrate that our method outperforms state-of-the-art methods in terms of visual quality and generalization ability.

cs.CV↗

arXiv2025

TVPR: Text-to-Video Person Retrieval and a New Benchmark

Most existing methods for text-based person retrieval focus on text-to-image person retrieval. Nevertheless, due to the lack of dynamic information provided by isolated frames, the performance is hampered when the person is obscured or variable motion details are missed in isolated frames. To overcome this, we propose a novel Text-to-Video Person Retrieval (TVPR) task. Since there is no dataset or benchmark that describes person videos with natural language, we construct a large-scale cross-modal person video dataset containing detailed natural language annotations, termed as Text-to-Video Person Re-identification (TVPReid) dataset. In this paper, we introduce a Multielement Feature Guided Fragments Learning (MFGF) strategy, which leverages the cross-modal text-video representations to provide strong text-visual and text-motion matching information to tackle uncertain occlusion conflicting and variable motion details. Specifically, we establish two potential cross-modal spaces for text and video feature collaborative learning to progressively reduce the semantic difference between text and video. To evaluate the effectiveness of the proposed MFGF, extensive experiments have been conducted on TVPReid dataset. To the best of our knowledge, MFGF is the first successful attempt to use video for text-based person retrieval task and has achieved state-of-the-art performance on TVPReid dataset. The TVPReid dataset will be publicly available to benefit future research.

cs.CV↗

arXiv2025

C-FAITH: A Chinese Fine-Grained Benchmark for Automated Hallucination Evaluation

Despite the rapid advancement of large language models, they remain highly susceptible to generating hallucinations, which significantly hinders their widespread application. Hallucination research requires dynamic and fine-grained evaluation. However, most existing hallucination benchmarks (especially in Chinese language) rely on human annotations, making automatical and cost-effective hallucination evaluation challenging. To address this, we introduce HaluAgent, an agentic framework that automatically constructs fine-grained QA dataset based on some knowledge documents. Our experiments demonstrate that the manually designed rules and prompt optimization can improve the quality of generated data. Using HaluAgent, we construct C-FAITH, a Chinese QA hallucination benchmark created from 1,399 knowledge documents obtained from web scraping, totaling 60,702 entries. We comprehensively evaluate 16 mainstream LLMs with our proposed C-FAITH, providing detailed experimental results and analysis.

cs.CL↗

arXiv2025

Pay More Attention to the Robustness of Prompt for Instruction Data Mining

Instruction tuning has emerged as a paramount method for tailoring the behaviors of LLMs. Recent work has unveiled the potential for LLMs to achieve high performance through fine-tuning with a limited quantity of high-quality instruction data. Building upon this approach, we further explore the impact of prompt's robustness on the selection of high-quality instruction data. This paper proposes a pioneering framework of high-quality online instruction data mining for instruction tuning, focusing on the impact of prompt's robustness on the data mining process. Our notable innovation, is to generate the adversarial instruction data by conducting the attack for the prompt of online instruction data. Then, we introduce an Adversarial Instruction-Following Difficulty metric to measure how much help the adversarial instruction data can provide to the generation of the corresponding response. Apart from it, we propose a novel Adversarial Instruction Output Embedding Consistency approach to select high-quality online instruction data. We conduct extensive experiments on two benchmark datasets to assess the performance. The experimental results serve to underscore the effectiveness of our proposed two methods. Moreover, the results underscore the critical practical significance of considering prompt's robustness.

cs.AI↗

arXiv2025

Enabling Versatile Controls for Video Diffusion Models

Despite substantial progress in text-to-video generation, achieving precise and flexible control over fine-grained spatiotemporal attributes remains a significant unresolved challenge in video generation research. To address these limitations, we introduce VCtrl (also termed PP-VCtrl), a novel framework designed to enable fine-grained control over pre-trained video diffusion models in a unified manner. VCtrl integrates diverse user-specified control signals-such as Canny edges, segmentation masks, and human keypoints-into pretrained video diffusion models via a generalizable conditional module capable of uniformly encoding multiple types of auxiliary signals without modifying the underlying generator. Additionally, we design a unified control signal encoding pipeline and a sparse residual connection mechanism to efficiently incorporate control representations. Comprehensive experiments and human evaluations demonstrate that VCtrl effectively enhances controllability and generation quality. The source code and pre-trained models are publicly available and implemented using the PaddlePaddle framework at http://github.com/PaddlePaddle/PaddleMIX/tree/develop/ppdiffusers/examples/ppvctrl.

cs.CV↗

arXiv2025

Constraints from anti-unitary symmetries on phase diagrams of sign problem-free models

A powerful way to guarantee the absence of a sign problem in determinantal quantum Monte Carlo simulations is imposing a particular type of anti-unitary symmetries. It is shown that these same symmetries give rise to constraints on correlation functions, which can be used to identify local operators whose correlations upper bound those of a large class of physically relevant operators, including for example superconducting order parameters. Our results also help understand why it is difficult to realize generic finite-momentum orders in sign problem-free models.

cond-mat.str-el↗