arXiv subjects

Qing Wang

Publications and source records attributed to Qing Wang.

At least 127 records · Page 7Linked to original sources

arXiv2025

Improving Significant Wave Height Prediction Using Chronos Models

Accurate wave height prediction is critical for maritime safety and coastal resilience, yet conventional physics-based models and traditional machine learning methods face challenges in computational efficiency and nonlinear dynamics modeling. This study introduces Chronos, the first implementation of a large language model (LLM)-powered temporal architecture (Chronos) optimized for wave forecasting. Through advanced temporal pattern recognition applied to historical wave data from three strategically chosen marine zones in the Northwest Pacific basin, our framework achieves multimodal improvements: (1) 14.3% reduction in training time with 2.5x faster inference speed compared to PatchTST baselines, achieving 0.575 mean absolute scaled error (MASE) units; (2) superior short-term forecasting (1-24h) across comprehensive metrics; (3) sustained predictive leadership in extended-range forecasts (1-120h); and (4) demonstrated zero-shot capability maintaining median performance (rank 4/12) against specialized operational models. This LLM-enhanced temporal modeling paradigm establishes a new standard in wave prediction, offering both computationally efficient solutions and a transferable framework for complex geophysical systems modeling.

cs.LG↗

arXiv2025

MMC: Iterative Refinement of VLM Reasoning via MCTS-based Multimodal Critique

Visual language models (VLMs) have demonstrated strong performance across diverse multimodal reasoning tasks but still face challenges such as hallucinations, resulting in incorrect reasoning outcomes. Inspired by recent research on external feedback mechanisms in large language models (LLMs), we propose a multimodal actor-critic framework to enhance VLM reasoning capabilities. Specifically, the actor model generates step-by-step reasoning paths based on image and text inputs, while the critic model evaluates these reasoning paths and provides corrective feedback. The actor model iteratively refines its reasoning based on the feedback until the reasoning outcome is deemed satisfactory by the critic model. To reduce reliance on costly manual annotations, we introduce an automated method for constructing multimodal critique datasets. By leveraging Monte Carlo Tree Search (MCTS), we systematically guide the actor model to explore diverse reasoning paths. To obtain critique data for correcting erroneous reasoning steps, we prompt an annotator model to compare pairs of reasoning paths diverging from a shared ancestor node - one leading to a correct conclusion and the other to an incorrect one. This approach enables us to construct the MMC (MCTS-based Multimodal Critique) dataset, upon which we further develop a comprehensive training and inference pipeline. Extensive experiments conducted on several public benchmark datasets and mainstream VLMs demonstrate that our approach significantly improves the performance of VLM on complex multimodal reasoning tasks, underscoring its effectiveness and wide applicability.

cs.MM↗

arXiv2025

The Tenth NTIRE 2025 Efficient Super-Resolution Challenge Report

This paper presents a comprehensive review of the NTIRE 2025 Challenge on Single-Image Efficient Super-Resolution (ESR). The challenge aimed to advance the development of deep models that optimize key computational metrics, i.e., runtime, parameters, and FLOPs, while achieving a PSNR of at least 26.90 dB on the $\operatorname{DIV2K\_LSDIR\_valid}$ dataset and 26.99 dB on the $\operatorname{DIV2K\_LSDIR\_test}$ dataset. A robust participation saw \textbf{244} registered entrants, with \textbf{43} teams submitting valid entries. This report meticulously analyzes these methods and results, emphasizing groundbreaking advancements in state-of-the-art single-image ESR techniques. The analysis highlights innovative approaches and establishes benchmarks for future research in the field.

cs.CV↗

arXiv2025

Twisted heterobilayer photonic crystal based on stacking and selective etching of 2D materials

Nanophotonic devices with moiré superlattice is currently attracting broad interest due to the unique periodicity and high efficiency control of photons. Till now, experimental investigations mainly focus on the single layer device, i.e., two or more layers of photonic crystal patterns are merged and etched in a single layer of material. By comparison, twisted photonic crystal with multilayer materials raises challenges in the nanofabrication technology, because the growth of upper layer material usually requires a smooth bottom layer without nanostructures. Hereby, we fabricate twisted heterobilayer photonic crystal in the graphite/Si$_3$N$_4$ heterostructure. We use dry transfer method to stack the graphite on top of bottom Si$_3$N$_4$ with pre-etched photonic crystal patterns. Selective dry etching recipes are used to etch two photonic crystal layers individually, which improves the quality and accuracy in alignment. The cavity photonic mode at the visible wavelength $\sim 700$ nm arsing from the moiré site is clearly observed in experiment. These results reveal the experimental diagram of heterobilayer nanophotonic devices and open the way to design flexibility and control of photons in new degrees of freedom.

physics.optics↗

arXiv2025

NTR-Gaussian: Nighttime Dynamic Thermal Reconstruction with 4D Gaussian Splatting Based on Thermodynamics

Thermal infrared imaging offers the advantage of all-weather capability, enabling non-intrusive measurement of an object's surface temperature. Consequently, thermal infrared images are employed to reconstruct 3D models that accurately reflect the temperature distribution of a scene, aiding in applications such as building monitoring and energy management. However, existing approaches predominantly focus on static 3D reconstruction for a single time period, overlooking the impact of environmental factors on thermal radiation and failing to predict or analyze temperature variations over time. To address these challenges, we propose the NTR-Gaussian method, which treats temperature as a form of thermal radiation, incorporating elements like convective heat transfer and radiative heat dissipation. Our approach utilizes neural networks to predict thermodynamic parameters such as emissivity, convective heat transfer coefficient, and heat capacity. By integrating these predictions, we can accurately forecast thermal temperatures at various times throughout a nighttime scene. Furthermore, we introduce a dynamic dataset specifically for nighttime thermal imagery. Extensive experiments and evaluations demonstrate that NTR-Gaussian significantly outperforms comparison methods in thermal reconstruction, achieving a predicted temperature error within 1 degree Celsius.

cs.CV↗

arXiv2025

Depth-Adaptive Graph Neural Networks via Learnable Bakry-'Emery Curvature

Graph Neural Networks (GNNs) have demonstrated strong representation learning capabilities for graph-based tasks. Recent advances on GNNs leverage geometric properties, such as curvature, to enhance its representation capabilities by modeling complex connectivity patterns and information flow within graphs. However, most existing approaches focus solely on discrete graph topology, overlooking diffusion dynamics and task-specific dependencies essential for effective learning. To address this, we propose integrating Bakry-Émery curvature, which captures both structural and task-driven aspects of information propagation. We develop an efficient, learnable approximation strategy, making curvature computation scalable for large graphs. Furthermore, we introduce an adaptive depth mechanism that dynamically adjusts message-passing layers per vertex based on its curvature, ensuring efficient propagation. Our theoretical analysis establishes a link between curvature and feature distinctiveness, showing that high-curvature vertices require fewer layers, while low-curvature ones benefit from deeper propagation. Extensive experiments on benchmark datasets validate the effectiveness of our approach, showing consistent performance improvements across diverse graph learning tasks.

cs.LG↗

arXiv2025

Permutation-Invariant Graph Partitioning:How Graph Neural Networks Capture Structural Interactions?

Graph Neural Networks (GNNs) have paved the way for being a cornerstone in graph-related learning tasks. Yet, the ability of GNNs to capture structural interactions within graphs remains under-explored. In this work, we address this gap by drawing on the insight that permutation invariant graph partitioning enables a powerful way of exploring structural interactions. We establish theoretical connections between permutation invariant graph partitioning and graph isomorphism, and then propose Graph Partitioning Neural Networks (GPNNs), a novel architecture that efficiently enhances the expressive power of GNNs in learning structural interactions. We analyze how partitioning schemes and structural interactions contribute to GNN expressivity and their trade-offs with complexity. Empirically, we demonstrate that GPNNs outperform existing GNN models in capturing structural interactions across diverse graph benchmark tasks.

cs.LG↗

arXiv2025

PhenoProfiler: Advancing Phenotypic Learning for Image-based Drug Discovery

In the field of image-based drug discovery, capturing the phenotypic response of cells to various drug treatments and perturbations is a crucial step. However, existing methods require computationally extensive and complex multi-step procedures, which can introduce inefficiencies, limit generalizability, and increase potential errors. To address these challenges, we present PhenoProfiler, an innovative model designed to efficiently and effectively extract morphological representations, enabling the elucidation of phenotypic changes induced by treatments. PhenoProfiler is designed as an end-to-end tool that processes whole-slide multi-channel images directly into low-dimensional quantitative representations, eliminating the extensive computational steps required by existing methods. It also includes a multi-objective learning module to enhance robustness, accuracy, and generalization in morphological representation learning. PhenoProfiler is rigorously evaluated on large-scale publicly available datasets, including over 230,000 whole-slide multi-channel images in end-to-end scenarios and more than 8.42 million single-cell images in non-end-to-end settings. Across these benchmarks, PhenoProfiler consistently outperforms state-of-the-art methods by up to 20%, demonstrating substantial improvements in both accuracy and robustness. Furthermore, PhenoProfiler uses a tailored phenotype correction strategy to emphasize relative phenotypic changes under treatments, facilitating the detection of biologically meaningful signals. UMAP visualizations of treatment profiles demonstrate PhenoProfiler ability to effectively cluster treatments with similar biological annotations, thereby enhancing interpretability. These findings establish PhenoProfiler as a scalable, generalizable, and robust tool for phenotypic learning.

cs.LG↗

arXiv2025

Mimicking the Familiar: Dynamic Command Generation for Information Theft Attacks in LLM Tool-Learning System

Information theft attacks pose a significant risk to Large Language Model (LLM) tool-learning systems. Adversaries can inject malicious commands through compromised tools, manipulating LLMs to send sensitive information to these tools, which leads to potential privacy breaches. However, existing attack approaches are black-box oriented and rely on static commands that cannot adapt flexibly to the changes in user queries and the invocation chain of tools. It makes malicious commands more likely to be detected by LLM and leads to attack failure. In this paper, we propose AutoCMD, a dynamic attack comment generation approach for information theft attacks in LLM tool-learning systems. Inspired by the concept of mimicking the familiar, AutoCMD is capable of inferring the information utilized by upstream tools in the toolchain through learning on open-source systems and reinforcement with target system examples, thereby generating more targeted commands for information theft. The evaluation results show that AutoCMD outperforms the baselines with +13.2% $ASR_{Theft}$, and can be generalized to new tool-learning systems to expose their information leakage risks. We also design four defense methods to effectively protect tool-learning systems from the attack.

cs.AI↗

arXiv2025

From Allies to Adversaries: Manipulating LLM Tool-Calling through Adversarial Injection

Tool-calling has changed Large Language Model (LLM) applications by integrating external tools, significantly enhancing their functionality across diverse tasks. However, this integration also introduces new security vulnerabilities, particularly in the tool scheduling mechanisms of LLM, which have not been extensively studied. To fill this gap, we present ToolCommander, a novel framework designed to exploit vulnerabilities in LLM tool-calling systems through adversarial tool injection. Our framework employs a well-designed two-stage attack strategy. Firstly, it injects malicious tools to collect user queries, then dynamically updates the injected tools based on the stolen information to enhance subsequent attacks. These stages enable ToolCommander to execute privacy theft, launch denial-of-service attacks, and even manipulate business competition by triggering unscheduled tool-calling. Notably, the ASR reaches 91.67% for privacy theft and hits 100% for denial-of-service and unscheduled tool calling in certain cases. Our work demonstrates that these vulnerabilities can lead to severe consequences beyond simple misuse of tool-calling systems, underscoring the urgent need for robust defensive strategies to secure LLM Tool-calling systems.

cs.CR↗

arXiv2025

NPU-NTU System for Voice Privacy 2024 Challenge

Speaker anonymization is an effective privacy protection solution that conceals the speaker's identity while preserving the linguistic content and paralinguistic information of the original speech. To establish a fair benchmark and facilitate comparison of speaker anonymization systems, the VoicePrivacy Challenge (VPC) was held in 2020 and 2022, with a new edition planned for 2024. In this paper, we describe our proposed speaker anonymization system for VPC 2024. Our system employs a disentangled neural codec architecture and a serial disentanglement strategy to gradually disentangle the global speaker identity and time-variant linguistic content and paralinguistic information. We introduce multiple distillation methods to disentangle linguistic content, speaker identity, and emotion. These methods include semantic distillation, supervised speaker distillation, and frame-level emotion distillation. Based on these distillations, we anonymize the original speaker identity using a weighted sum of a set of candidate speaker identities and a randomly generated speaker identity. Our system achieves the best trade-off of privacy protection and emotion preservation in VPC 2024.

eess.AS↗

arXiv2025

Stable Tree Labelling for Accelerating Distance Queries on Dynamic Road Networks

Finding the shortest-path distance between two arbitrary vertices is an important problem in road networks. Due to real-time traffic conditions, road networks undergo dynamic changes all the time. Current state-of-the-art methods incrementally maintain a distance labelling based on a hierarchy among vertices to support efficient distance computation. However, their labelling sizes are often large and cannot be efficiently maintained. To combat these issues, we present a simple yet efficient labelling method, namely \emph{Stable Tree Labelling} (STL), for answering distance queries on dynamic road networks. We observe that the properties of an underlying hierarchy play an important role in improving and balancing query and update performance. Thus, we introduce the notion of \emph{stable tree hierarchy} which lays the ground for developing efficient maintenance algorithms on dynamic road networks. Based on stable tree hierarchy, STL can be efficiently constructed as a 2-hop labelling. A crucial ingredient of STL is to only store distances within subgraphs in labels, rather than distances in the entire graph, which restricts the labels affected by dynamic changes. We further develop two efficient maintenance algorithms upon STL: \emph{Label Search algorithm} and \emph{Pareto Search algorithm}. Label Search algorithm identifies affected ancestors in a stable tree hierarchy and performs efficient searches to update labels from those ancestors. Pareto Search algorithm explores the interaction between search spaces of different ancestors, and combines searches from multiple ancestors into only two searches for each update, eliminating duplicate graph traversals. The experiments show that our algorithms significantly outperform state-of-the-art dynamic methods in maintaining the labelling and query processing, while requiring an order of magnitude less space.

cs.DS↗

arXiv2025

Divergence-Augmented Policy Optimization

In deep reinforcement learning, policy optimization methods need to deal with issues such as function approximation and the reuse of off-policy data. Standard policy gradient methods do not handle off-policy data well, leading to premature convergence and instability. This paper introduces a method to stabilize policy optimization when off-policy data are reused. The idea is to include a Bregman divergence between the behavior policy that generates the data and the current policy to ensure small and safe policy updates with off-policy data. The Bregman divergence is calculated between the state distributions of two policies, instead of only on the action probabilities, leading to a divergence augmentation formulation. Empirical experiments on Atari games show that in the data-scarce scenario where the reuse of off-policy data becomes necessary, our method can achieve better performance than other state-of-the-art deep reinforcement learning algorithms.

cs.LG↗

arXiv2025

HECLIP: Histology-Enhanced Contrastive Learning for Imputation of Transcriptomics Profiles

Histopathology, particularly hematoxylin and eosin (H\&E) staining, plays a critical role in diagnosing and characterizing pathological conditions by highlighting tissue morphology. However, H\&E-stained images inherently lack molecular information, requiring costly and resource-intensive methods like spatial transcriptomics to map gene expression with spatial resolution. To address these challenges, we introduce HECLIP (Histology-Enhanced Contrastive Learning for Imputation of Profiles), an innovative deep learning framework that bridges the gap between histological imaging and molecular profiling. HECLIP is specifically designed to infer gene expression profiles directly from H\&E-stained images, eliminating the need for expensive spatial transcriptomics assays. HECLIP leverages an advanced image-centric contrastive loss function to optimize image representation learning, ensuring that critical morphological patterns in histology images are effectively captured and translated into accurate gene expression profiles. This design enhances the predictive power of the image modality while minimizing reliance on gene expression data. Through extensive benchmarking on publicly available datasets, HECLIP demonstrates superior performance compared to existing approaches, delivering robust and biologically meaningful predictions. Detailed ablation studies further underscore its effectiveness in extracting molecular insights from histology images. Additionally, HECLIP's scalable and cost-efficient approach positions it as a transformative tool for both research and clinical applications, driving advancements in precision medicine. The source code for HECLIP is openly available at https://github.com/QSong-github/HECLIP.

cs.CE↗

arXiv2025

An Experimental Study on Joint Modeling for Sound Event Localization and Detection with Source Distance Estimation

In traditional sound event localization and detection (SELD) tasks, the focus is typically on sound event detection (SED) and direction-of-arrival (DOA) estimation, but they fall short of providing full spatial information about the sound source. The 3D SELD task addresses this limitation by integrating source distance estimation (SDE), allowing for complete spatial localization. We propose three approaches to tackle this challenge: a novel method with independent training and joint prediction, which firstly treats DOA and distance estimation as separate tasks and then combines them to solve 3D SELD; a dual-branch representation with source Cartesian coordinate used for simultaneous DOA and distance estimation; and a three-branch structure that jointly models SED, DOA, and SDE within a unified framework. Our proposed method ranked first in the DCASE 2024 Challenge Task 3, demonstrating the effectiveness of joint modeling for addressing the 3D SELD task. The relevant code for this paper will be open-sourced in the future.

cs.SD↗

arXiv2025

Classical Limit of Yukawa theory from quantum state perspective

We derive the quantum states corresponding to classical scalar fields in the representation expanded by the eigenstates of quantum field operators. This allows us to directly observe the spatial entanglement structure of quantum states and explore the differences and relationships between quantum superposition and classical superposition. We find that if two classical fields are identical in a certain spatial region, then their corresponding quantum states have the same reduced density matrix in that region. This indicates that knowing the classical field in a local region is sufficient to derive the reduced density matrix for that region. According to the correspondence between classical quantities and quantum states, we derive the equation of motion of the classical theory from the evolution of quantum states in Yukawa theory. This leads to the relativistic classical Yukawa theory, and we further obtain the relativistic corrections to the Yukawa potential.

hep-th↗

arXiv2025

DiffAttack: Diffusion-based Timbre-reserved Adversarial Attack in Speaker Identification

Being a form of biometric identification, the security of the speaker identification (SID) system is of utmost importance. To better understand the robustness of SID systems, we aim to perform more realistic attacks in SID, which are challenging for both humans and machines to detect. In this study, we propose DiffAttack, a novel timbre-reserved adversarial attack approach that exploits the capability of a diffusion-based voice conversion (DiffVC) model to generate adversarial fake audio with distinct target speaker attribution. By introducing adversarial constraints into the generative process of the diffusion-based voice conversion model, we craft fake samples that effectively mislead target models while preserving speaker-wise characteristics. Specifically, inspired by the use of randomly sampled Gaussian noise in conventional adversarial attacks and diffusion processes, we incorporate adversarial constraints into the reverse diffusion process. These constraints subtly guide the reverse diffusion process toward aligning with the target speaker distribution. Our experiments on the LibriTTS dataset indicate that DiffAttack significantly improves the attack success rate compared to vanilla DiffVC and other methods. Moreover, objective and subjective evaluations demonstrate that introducing adversarial constraints does not compromise the speech quality generated by the DiffVC model.

cs.SD↗

arXiv2024

Understanding Individual Agent Importance in Multi-Agent System via Counterfactual Reasoning

Explaining multi-agent systems (MAS) is urgent as these systems become increasingly prevalent in various applications. Previous work has proveided explanations for the actions or states of agents, yet falls short in understanding the black-boxed agent's importance within a MAS and the overall team strategy. To bridge this gap, we propose EMAI, a novel agent-level explanation approach that evaluates the individual agent's importance. Inspired by counterfactual reasoning, a larger change in reward caused by the randomized action of agent indicates its higher importance. We model it as a MARL problem to capture interactions across agents. Utilizing counterfactual reasoning, EMAI learns the masking agents to identify important agents. Specifically, we define the optimization function to minimize the reward difference before and after action randomization and introduce sparsity constraints to encourage the exploration of more action randomization of agents during training. The experimental results in seven multi-agent tasks demonstratee that EMAI achieves higher fidelity in explanations than baselines and provides more effective guidance in practical applications concerning understanding policies, launching attacks, and patching policies.

cs.AI↗