About

Weiqiang Wang is the Gordon Whyburn Professor of Mathematics at the University of Virginia. His research focuses on Representation Theory, Quantum Groups, and related areas within mathematics. He is associated with the Department of Mathematics at the University of Virginia, located in Kerchof Hall. His contact information includes a phone number (434-924-7905) and an email address (ww9c@virginia.edu).

Research topics

• Computer science
• Artificial intelligence
• Mathematics
• Computer vision
• Pure mathematics

Selected publications

• Gas-solid two-phase flow erosion of needle throttle valve in shale gas field based on CFD-DEM model

  Petroleum Science · 2026-02-01 · 1 citations

  articleOpen access

  Needle throttle valve (NTV) is a key equipment to ensure the safe production of shale gas fields, but it is often seriously eroded by the solid particles in the produced gas. At this work, the CFD-DEM coupling calculation method is adopted to investigate the internal flow field characteristics and the erosion rate of each component of the NTV under gas-solid two-phase flow. The accuracy of the numerical model is validated by the comparison results of simulation and experiment. The results show that with the increase of particle diameter, the maximum erosion rate of each part of the valve generally increases. When valve opening degree (VOD) is equal to 0.5, it has the best effect on suppressing erosion rates at high flow velocity. There is a “vulnerable zone” of the spool cone that is not affected by the changes in particle diameter and VOD, which occupies 2/3 of the height of the spool cone. This work not only predicts the vulnerable zone of each component of the valve, but also reveals the erosion mechanism of the NTV, and can provide a reference for the design and maintenance of the valve.

  PublisherDOI

• Saliency-based dual-layer contrast method for infrared small target detection

  2025-09-19

  article

  Traditional infrared small target detection can easily cause false alarm detection in complex environment, therefore, to address this issue, we present a saliency-based dual-layer contrast method for infrared small target detection. Firstly, the possible region of the target can be obtained by calculating the spectral residuals of the original infrared images. Secondly, the double-layer local contrast detection method can be used for identification in this area. Finally, the small target can be detected by adaptive threshold segmentation method. Experimental results indicate that compared with traditional detection algorithms, the proposed algorith can enhance targets, suppress background and noise in different scenes, higher detection efficiency, and meet the real-time requirements.

  PublisherDOI

• GenPTW: Latent Image Watermarking for Provenance Tracing and Tamper Localization

  ArXiv.org · 2025-04-28

  preprintOpen access

  The proliferation of generative image models has revolutionized AIGC creation while amplifying concerns over content provenance and manipulation forensics. Existing methods are typically either unable to localize tampering or restricted to specific generative settings, limiting their practical utility. We propose \textbf{GenPTW}, a \textbf{Gen}eral watermarking framework that unifies \textbf{P}rovenance tracing and \textbf{T}amper localization in latent space. It supports both in-generation and post-generation embedding without altering the generative process, and is plug-and-play compatible with latent diffusion models (LDMs) and visual autoregressive (VAR) models. To achieve precise provenance tracing and tamper localization, we embed the watermark using two complementary mechanisms: cross-attention fusion aligned with latent semantics and spatial fusion providing explicit spatial guidance for edit sensitivity. A tamper-aware extractor jointly conducts provenance tracing and tamper localization by leveraging watermark features together with high-frequency features. Experiments show that GenPTW maintains high visual fidelity and strong robustness against diverse AIGC-editing.

  PublisherOA PDFDOI

• Veritas: Generalizable Deepfake Detection via Pattern-Aware Reasoning

  ArXiv.org · 2025-08-28

  preprintOpen access

  Deepfake detection remains a formidable challenge due to the complex and evolving nature of fake content in real-world scenarios. However, existing academic benchmarks suffer from severe discrepancies from industrial practice, typically featuring homogeneous training sources and low-quality testing images, which hinder the practical deployments of current detectors. To mitigate this gap, we introduce HydraFake, a dataset that simulates real-world challenges with hierarchical generalization testing. Specifically, HydraFake involves diversified deepfake techniques and in-the-wild forgeries, along with rigorous training and evaluation protocol, covering unseen model architectures, emerging forgery techniques and novel data domains. Building on this resource, we propose Veritas, a multi-modal large language model (MLLM) based deepfake detector. Different from vanilla chain-of-thought (CoT), we introduce pattern-aware reasoning that involves critical reasoning patterns such as "planning" and "self-reflection" to emulate human forensic process. We further propose a two-stage training pipeline to seamlessly internalize such deepfake reasoning capacities into current MLLMs. Experiments on HydraFake dataset reveal that although previous detectors show great generalization on cross-model scenarios, they fall short on unseen forgeries and data domains. Our Veritas achieves significant gains across different OOD scenarios, and is capable of delivering transparent and faithful detection outputs.

  PublisherOA PDFDOI

• Scalable Autoregressive Monocular Depth Estimation

  2025-06-10 · 1 citations

  article

  This paper proposes a new autoregressive model as an effective and scalable monocular depth estimator. Our idea is simple: We tackle the monocular depth estimation (MDE) task with an autoregressive prediction paradigm, based on two core designs. First, our depth autoregressive model (DAR) treats the depth map of different resolutions as a set of tokens, and conducts the low-to-high resolution autoregressive objective with a patch-wise causal mask. Second, our DAR recursively discretizes the entire depth range into more compact intervals, and attains the coarse-to-fine granularity autoregressive objective in an ordinal-regression manner. By coupling these two autoregressive objectives, our DAR establishes new state-of-the-art (SOTA) on KITTI and NYU Depth v2 by clear margins. Further, our scalable approach allows us to scale the model up to 2.0B and achieve the best RMSE of 1.799 on the KITTI dataset (5% improvement) compared to 1.896 by the current SOTA (Depth Anything). DAR further showcases zero-shot generalization ability on unseen datasets. These results suggest that DAR yields superior performance with an autoregressive prediction paradigm, providing a promising approach to equip modern autoregressive large models (e.g., GPT-4o) with depth estimation capabilities. Project page: https://depth-ar.github.io/.

  PublisherDOI

• VQualA 2025 Challenge on Image Super-Resolution Generated Content Quality Assessment: Methods and Results

  ArXiv.org · 2025-09-08

  preprintOpen access

  This paper presents the ISRGC-Q Challenge, built upon the Image Super-Resolution Generated Content Quality Assessment (ISRGen-QA) dataset, and organized as part of the Visual Quality Assessment (VQualA) Competition at the ICCV 2025 Workshops. Unlike existing Super-Resolution Image Quality Assessment (SR-IQA) datasets, ISRGen-QA places a greater emphasis on SR images generated by the latest generative approaches, including Generative Adversarial Networks (GANs) and diffusion models. The primary goal of this challenge is to analyze the unique artifacts introduced by modern super-resolution techniques and to evaluate their perceptual quality effectively. A total of 108 participants registered for the challenge, with 4 teams submitting valid solutions and fact sheets for the final testing phase. These submissions demonstrated state-of-the-art (SOTA) performance on the ISRGen-QA dataset. The project is publicly available at: https://github.com/Lighting-YXLI/ISRGen-QA.

  PublisherOA PDFDOI

• Effect of NaF Additive on Energy Consumption and Coating Properties in Micro-arc Oxidation of Aluminum Alloy

  Progress in Chinese Materials Sciences · 2025-06-28

  articleOpen accessSenior author

  In the process of achieving the goal of automotive lightweighting, aluminum alloy has become an indispensable key material with its excellent characteristics, but their insufficient surface properties often require enhancement through technology modification. Micro-arc oxidation (MAO) , as an effective surface treatment technique, can in situ grow high-performance ceramic coatings on aluminum alloy surfaces. However, the high energy consumption limits its industrial application. In this paper, by regulating the NaF concentration in the electrolyte, the effects on the MAO discharge characteristics, coating structure and performance, and MAO energy consumption were systematically investigated. The results showed that the addition of NaF significantly reduced the breakdown voltage, and the unit energy consumption decreased with the increase of NaF concentration, in which the unit energy consumption of the 10g/L NaF system (MF10) was reduced by about 52. 2% compared with that of the unadded system (MF0) . The coating morphology showed that the moderate amount of NaF could reduce the coating pores and enhance the densification, while the excessive amount of NaF led to large pores and surface ablation. XRD analysis showed that NaF promoted the generation of α-Al2O3, mullite, and AlF3 phases. Coating performance tests showed that the corrosion current density of the MF10 samples was as low as 2. 50×10-8 A·cm-2 and the minimum abrasion depth was obtained in friction tests. This study provides a theoretical basis and technical reference for the development of a low-energy and high-performance MAO process for aluminum alloys.

  PublisherOA PDFDOI

• Dynamic traffic encountering scenario generation for testing of ship collision avoidance algorithm

  Ocean Engineering · 2025-12-05 · 1 citations

  article1st author
  PublisherDOI

• COCO-Inpaint: A Benchmark for Detecting and Localizing Inpainting-Based Image Manipulations

  ArXiv.org · 2025-04-25

  preprintOpen access

  Recent advances in image manipulation have enabled highly photorealistic content generation, but also lowered the barrier to arbitrary editing, raising concerns about multimedia authenticity and security. Existing Image Manipulation Detection and Localization (IMDL) methods mainly target splicing or copy-move forgeries, while benchmarks for inpainting-based manipulations remain limited. To bridge this gap, we present COCO-Inpaint, a comprehensive benchmark specifically designed for inpainting detection and localization, with three key contributions: 1) High-quality inpainting samples generated by six state-of-the-art inpainting models, 2) Diverse generation scenarios enabled by four mask generation strategies with optional text guidance, and 3) Large-scale coverage of 238,302 inpainted images with rich semantic diversity. Our benchmark is constructed to highlight intrinsic inconsistencies between inpainted and authentic regions, rather than superficial semantic artifacts such as object shapes. We further establish a rigorous evaluation protocol with three standard metrics to benchmark existing IMDL methods and reveal current trends and challenges.

  PublisherOA PDFDOI

• Response Error Prediction and Feedback Control Method for Electro-Hydraulic Actuators Based on LSTM

  Electronics · 2025-05-13 · 1 citations

  articleOpen access

  The application of hydraulic systems in aerospace and engineering machines is becoming widespread. With the use of electro-hydraulic actuators, designing efficient and intelligent controllers can help the rapid expansion of electromechanical equipment in various scenarios. In response to the difficulty of slow response in the EHA control process, the paper designs an error prediction algorithm to predict the system response curve and replace the real-time error of PID input, achieving advanced correction of the controller. The experiment shows that the proposed method has a lower response time and smoother control curve while ensuring accuracy. It might have potential value in improving hydraulic system efficiency, reducing switching shock, and increasing system service life.

  PublisherOA PDFDOI

Recent grants

• Duality between representations of Lie superalgebras and Lie algebras via Kazhdan-Lusztig theory

  NSF · $100k · 2005–2008

• Canonical Bases, Categorification, and Modular Representations

  NSF · $318k · 2017–2020

• Quantum Symmetric Pairs, Categorification, and Geometry

  NSF · $345k · 2020–2024

• Affine algebras, Lie superalgebras, Hecke algebras, and representations

  NSF · $171k · 2008–2011

Frequent coauthors

• Zhenbo Qin

  Beijing Wuzi University

  22 shared

• Wei-Ping Li

  Zhengzhou Railway Vocational & Technical College

  19 shared

• Wen Gao

  15 shared

• Shao Huang

  14 shared

• Ke Lü

  University of Chinese Academy of Sciences

  13 shared

• Shun‐Jen Cheng

  12 shared

• Yuanqiang Cai

  12 shared

• Ke Lv

  West China Hospital of Sichuan University

  9 shared